pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
25376905 | PMC4245980 | pmc | 6,856 | {
"abstract": "MqsR-controlled\ncolanic acid and biofilm regulator (McbR, also\nknown as YncC) is the protein product of a highly induced gene in\nearly Escherichia coli biofilm development\nand has been regarded as an attractive target for blocking biofilm\nformation. This protein acts as a repressor for genes involved in\nexopolysaccharide production and an activator for genes involved in\nstress response. To better understand the role of McbR in governing\nthe switch from exponential growth to the biofilm state, we determined\nthe crystal structure of McbR to 2.1 Å. The structure reveals\nMcbR to be a member of the FadR C-terminal domain (FCD) family of\nthe GntR superfamily of transcriptional regulators (this family was\nnamed after the first identified member, GntR, a transcriptional repressor\nof the gluconate operon of Bacillus subtilis ). Previous to this study, only six of the predicted 2800 members\nof this family had been structurally characterized. Here, we identify\nthe residues that constitute the McbR effector and DNA binding sites.\nIn addition, comparison of McbR with other members of the FCD domain\nfamily shows that this family of proteins adopts highly distinct oligomerization\ninterfaces, which has implications for DNA binding and regulation.",
"discussion": "Results and Discussion McbR Is a Member of the VanR Subfamily of\nGntR Transcriptional\nRegulators Two constructs of McbR were screened for their\nability to form diffraction-quality crystals: McbR 1–221 and McbR 10–221 . The latter is missing the first\n9 amino acids, which were predicted to be disordered (PSIPRED, 26 IUPRED 27 , 28 ). Only McbR 10–221 formed crystals suitable for structure determination and is referred\nto hereafter as McbR. The crystal structure of McbR was determined\nby single-wavelength anomalous dispersion (SAD) using SeMet-labeled\nprotein, and the atomic model was refined to 2.1 Å resolution\n(Table 1 and Figures 1 B and 2 A). Two molecules of McbR are present\nin the asymmetric unit and are related by a nearly perfect 2-fold\naxis (179.8°; superposition using the C-terminal FCD domain;\nFigure 2 A). This is consistent with the observation\nthat McbR is predominantly a dimer in solution (Figure 2 B). McbR, like other members of the GntR family, consists\nof an N-terminal wHTH domain (residues 10–76; residues 38–46\nin the second subunit were not modeled due to a lack of clear electron\ndensity) and a C-terminal all α-helical effector binding domain\n(residues 77–219; Figure 1 B). The wHTH\ndomain is composed of three α-helices (α1−α3)\nand three β-strands (β1−β3), which form a\nsmall β-sheet that is a defining characteristic of the wHTH\nfold. The C-terminal domain is composed of six α-helices (α4−α9).\nThe secondary structure elements and topology of the C-terminal domain\nplaces McbR in the FadR C-terminal domain (FCD) family of GntR transcriptional\nregulators (Figure 1 A). Figure 2 Dimerization interface\nof McbR. (A) McbR dimer, with one monomer\ncolored in shades of blue and the second colored in shades of green.\nThe N-terminal wHTH domains are colored in light blue and light green,\nand the C-terminal FCD domains are colored in teal and dark green.\nThe residues between β1/α3 of monomer B are disordered\nand represented as a dotted line. The pseudo 2-fold axis is indicated\nby an arrow. (B) Size-exclusion chromatogram of McbR with elution\nvolumes of MW standards indicated (Bio-Rad; calculated molecular weight\nof the McbR monomer is ∼24.5 kDa). (C) Hydrophobic interactions\nthat stabilize the FCD domain dimerization interface; the pseudo 2-fold\naxis is indicated by a black circle. (D) Polar/salt bridge interactions\n(shown as black dashed lines) that stabilize the FCD domain dimerization\ninterface. (E) Interactions at the wHTH domain interface (polar/salt\nbridge interactions shown as black dashed lines); Leu56 is labeled\nin italics to highlight it. Whereas L56 A (light blue) is\nburied in the interface, Leu56 B (bright green) is not and\nis instead at the interface periphery. McbR Dimerization Interface Is Extensive and Includes Both the\nN- and C-Terminal Domains The FCD family has ∼2800\nmembers from more than 400 distinct species from archaea to eukaryota. 10 A structural homology search using Dali identified\nonly 6 other structures that have a high degree of similarity to McbR\n( Z -score > 7 using only the FCD domain; Table 4 and Figure 3 ). These represent\nthe only other members of the FCD family with known structures. The\ndimerization interface mediated by the C-terminal FCD domain is topologically\nconserved within the FCD family and is composed of the first helix\nin the FCD domain (α4 in McbR) and the N-terminal half of the\nkinked fourth helix (α7 in McbR). In McbR, the FCD dimerization\ninterface buries 1655 Å 2 of solvent-accessible surface\narea, which is 70% of the buried surface area (BSA) for the entire\nMcbR dimer. The hydrophobic core of the FCD dimerization interface\nis formed by residues Ile85, Ile88, Leu92, Met148, Ile150, Leu151,\nMet154, and Leu158 from both monomers, each of which is completely\noccluded from solvent (Figure 2 C). It is further\nstabilized by polar and salt bridge interactions, especially a bidentate\nhydrogen bond between Gln157 A and Gln157 B and\na bidentate salt bridge between Arg161 A and Glu153 B (A or B subscript indicates that the residue is from subunit\nA or B, respectively; Figure 2 D). Figure 3 Quaternary\nstructures of FCD family. FCD family members whose structures\nhave been determined are shown, with one monomer depicted in teal\nand one in gold. Metals bound to the FCD domains are depicted as light\nblue spheres. Ligands/molecules bound in the FCD ligand binding pockets\nare shown as magenta spheres or sticks. The dimerization helices (α4\nand α7 in McbR) are colored green and orange. The corresponding\nquaternary structures are depicted as cartoons, with the N-terminal\ndomains shown as triangles and the C-terminal domains as spheres.\n(A) Head-to-head dimerization in which both the wHTH domains and the\nFCD domains contribute to the dimerization interface. (B) Dimerization\nin which the wHTH domains do not interact with either one another\nor the FCD domains. (C) Head-to-head dimerization in which only the\nFCD domains contribute to the dimerization. (D) Domain swapping dimerization\nin which the wHTH domain of one monomer reaches across the FCD domain\ninterface to interact with the FCD domain of the second monomer. (E)\nSame as panel D except that the wHTH domains are inverted with respect\nto one another. Table 4 FCD Family\nMembers and Their Structural\nSimilarity to McbR a FCD only full-length name PDB R Z rmsd (Å) R Z rmsd (Å) Id (%) subfamily metal binding FCD ligands b McbR 4P9F VanR no UNK PS5454 3C7J 1 18.2 1.9 3 18.3 1.9 23 VanR yes Ni CGL2915 2DI3 2 15.5 2.6 5 15.5 4.0 20 FadR yes Zn RO03477 2HS5 3 15.4 2.6 2 19.7 3.1 19 VanR no Act Reut_B4629 3IHU 4 15.4 2.5 1 20.3 2.6 16 VanR no TM0439 3FMS 5 14.8 2.4 4 16.6 3.1 22 VanR yes Act, Ni FadR 1H9G 6 12.8 2.6 6 12.6 6.2 14 FadR no CoA-Myr a R , Z-score rank. Z , DALI Z-score.\nrmsd, root-mean-square deviation reported\nby DALI. Id, % sequence identity determined using FFAS. b Ligands/metals bound at the ligand\nbinding pocket; UNK, unknown; Act, acetate ion. In McbR, the wHTH domains also interact,\nextending the dimerization\ninterface beyond that typically observed in the FCD subfamily of GntR\nregulators. The wHTH interface buries 740 Å 2 of BSA,\nfor a total of 2395 Å 2 buried between the two McbR\nmonomers. Although the FCD domains are related by a near perfect 2-fold\naxis centered on Gln157 A/B , the wHTH are not. Instead,\nthey are related by a rotation of ∼172°. Thus, whereas\nLeu56 A is buried in the wHTH interface, the corresponding\nresidue (Leu56 B ) is located at the interface periphery\n(Figure 2 E). The wHTH dimerization is composed\nlargely of polar interactions (i.e., a hydrogen bond between Ser60 A/B and Arg57 A/B ) and a few hydrophobic interactions\n(Leu14 B and Leu56 A ); however, unlike the residues\nat the FCD interface, none of the wHTH interface residues become extensively\nburied upon complex formation (Figure 2 E).\nFinally, Asn62 A (wHTH domain) hydrogen bonds with Glu153 B (FCD domain); this is the only noncovalent interaction connecting\nthe two different domains from the distinct subunits in the dimer. Comparison of McbR with the other members of the FCD subfamily\nreveals that while the FCD dimerization interface is conserved within\nthe family, the relative orientation of the wHTH and FCD domains is\nnot. This gives rise to distinct differences in the orientations of\nthe wHTH domains and, in some cases, distinct quaternary structures\n(Figure 3 ). This is why the FCD family member\nidentified to be most similar to McbR using the DALI structural homology\nsearch database changes depending on whether the search is performed\nwith the McbR FCD domain alone (PS5454, PDB ID: 3C7J) or full-length\nMcbR (Reut_B4629, PDB ID: 3IHU; Table 4 ). As expected, the\nFCD proteins identified as most different from full-length McbR are\nFadR and CGL2915. These are also both members of the FadR subfamily,\nbut they have an additional helix between the wHTH and the FCD ligand\nbinding domains (Figure 4 ). The presence of\nthis helix leads to domain swapped quaternary structures, in which\nthe wHTH domain of subunit A crosses the dimerization interface to\nmake contacts with the FCD domain of subunit B. This domain swapping\nis not observed in the VanR subclass of FCD regulators, and instead,\nin these proteins, the wHTH and FCD domains of the same subunit are\nmore intimately associated. Figure 4 Ligand binding cavity of McbR and comparison\nto structural homologues.\n(A) Multiple sequence alignment showing high conservation at the N-terminal\ndomain (α1−β3) and the C-terminal domain (α4−α9)\nin McbR in comparison to the FCD family. Identical amino acids are\nhighlighted in black, and similar amino acids are highlighted in gray.\nMcbR e represents E. coli McbR, and McbR s represents S. typhimurium McbR. Alpha helices are depicted as cylinders above the sequence\nalignment, and the beta-strands, as arrows. Asterisks mark the residues\nin FadR that make base-specific contacts with DNA. Residues that define\nthe McbR ligand binding site are highlighted in yellow. Residues important\nfor metal binding in the FCD family are highlighted in orange. (B)\nStructural superposition of the three conserved histidines in metal\nbinding FCD family members and the corresponding residues in McbR\n(teal). Pseudomonas syringae PS5454\n(PDB ID: 3C7J) is shown in green, Thermotoga maritima TM0439 (PDB ID: 3FMS) is shown in orange, and Corynebacterium glutamicum CGL2915 (PDB ID: 2DI3) is shown in maroon. Each respective metal is shown as a sphere\nin the same color. (C) (Left) Cartoon depiction of McbR with the ligand\nbinding cavities represented as purple surfaces. (Right) Enlarged\nimage of the binding cavity highlighting the three conserved residues\nin McbR that appear to be important for ligand binding (colored as\nin the left panel). (D) Electron density for the ligand binding cavity\nin chain A of McbR. Positive density is shown as green chicken wire.\nResidues coordinating the unidentified entity (see text) are shown\nas teal sticks. Sigma level for the 2 F o – F c map is 1.0. Sigma level\nfor the F o – F c map is 3.0. The Structure of McbR Is Predominantly in a Ligand-Bound Conformation The C-terminal FCD domains are composed of either 6 (VanR subclass)\nor 7 (FadR subclass) α-helices that form an antiparallel helical\nbundle. McbR, which has 6 helices, is a member of the VanR subclass\n(Figure 1 A). The FCD domains have a large cavity\nin the center of this helical bundle, which is the location of the\nligand binding site. The structure of this cavity is identical between\nboth FCD domains in McbR, as the FCD domains superimpose with a root-mean-squared\ndeviation (rsmd) of only 0.24 Å (Figure 6 B). Although the helical topology is conserved among FCD domains,\nthe sequence conservation among FCD family members, especially the\nresidues that line the ligand binding cavities, is very low, likely\nreflecting their distinct ligand specificities (Figure 4 A). Recently, it was shown that the majority of FCD\nfamily members use three conserved histidines to bind a metal ion\nin the ligand binding cavity, suggesting that these regulators bind\nligands that interact directly with the bound metal (Figure 4 A,B). 10 In McbR, these\nhistidines are not conserved and are instead replaced by Arg139, Tyr185,\nand Ile207 (Figure 4 B). Thus, McbR is one of\nthe few FCD family members that does not bind a metal. Because of\nthis, the ligand pocket in McbR is large, with a volume of ∼200\nÅ 3 , nearly double that of the metal-binding FCD domains\n(Figure 4 C). Although the endogenous ligand\nfor McbR is still unknown, clear unambiguous density for a bound entity\nwas observed in the FCD ligand binding cavities of both monomers of\nMcbR (Figure 4 D). None of the protein and crystallization\nbuffer components, or derivatives thereof, fit the density. This is\nlikely because the density is rather undefined, potentially because\nit is not fully occupied, a phenomenon commonly observed without externally\nsupplied ligands and/or cofactors. Alternatively, the density could\ncorrespond to the biologically relevant ligand, as McbR is an E. coli protein and was expressed in E. coli . However, potential ligands, such as glucuronic\nacid, a component of colanic acid whose metabolism has been shown\nto be regulated by McbR, did not fit the density. 4 Finally, automated ligand fitting routines, such as the\nLigandFit program implemented in Phenix, also failed to identify a\nligand that satisfactorily fit the density. 29 , 30 Because the density did not enable the identity of the ligand to\nbe confidently determined, it has not been modeled. However,\nthe presence of the density did reveal the identity of\nthe residues that likely define the McbR ligand binding site. Namely,\nthe bound entity is strongly coordinated by two arginine residues,\nArg89 and Arg139, which themselves are organized via a shared salt\nbridge with Glu93 (Figures 4 C,D). Two neighboring\nasparagine residues, Asn135 and Asn211, also contribute to binding.\nTo investigate whether these residues are important for McbR function,\nwe generated two variants of McbR by mutating the residues that define\nthe entity binding site. Because mutating residues in the interior\nof a protein can also lead to protein unfolding, we generated two\ndistinct mutants: a double mutant in which Glu93 and Arg139 were substituted\nwith Ser and Phe, respectively, the structurally homologous residues\nin FadR (the residue structurally homologous to McbR Arg89 is also\nan Arg in FadR) and a triple mutant in which all three residues were\nmutated to alanines (Arg89Ala, Glu93Ala, and Arg139Ala). CD polarimetry\ndemonstrated that both McbR variants are folded, and EMSAs showed\nthey are functional ( Figure S1 ). The mutants\nwere somewhat less thermostable (Δ T m of −8.6 and −18.6 °C compared to WT for the double\nand triple mutant, respectively), but this was expected, as the mutations\nare in the interior of the protein; indeed, this is exactly why two\nmutants, one in which the residues were mutated to those present in\nFadR (the double mutant) and one in which the residues were simply\nmutated alanine (the triple mutant), were tested. McbR deletion from E. coli results in EPS overproduction and elicits\na mucoidy phenotype. 4 This mucoidy phenotype\nis substantially reduced upon producing McbR ectopically (Figure 5 ). However, cells producing McbR with triple mutations\n(Arg89Ala, Glu93Ala, and Arg139Ala) are mucoid (Figure 5 ). This observation demonstrates the importance of Arg89,\nGlu93, and Arg139 in binding the unknown ligand and, more importantly,\nthe physiological relevance of the unknown ligand in affecting EPS\nproduction. Both Arg89 and Arg139 are required for ligand binding,\nas cells producing McbR with only two mutations (Glu93Ser and Arg139Phe)\nalso remain less mucoid than cells with empty plasmid or cells producing\nMcbR with three mutations (Arg89Ala, Glu93Ala, and Arg139Ala) (Figure 5 ). Figure 5 Mucoidy level of E. coli MG1655\nΔ mcbR Δ Km producing\ndifferent McbR variants. Each strain was grown on LB agar supplemented\nwith 50 μg/mL kanamycin and 1 mM IPTG at 37 °C for 12 h.\nWT/empty, E. coli MG1655/pBS(Kan);\nΔ mcbR /empty, E. coli MG1655 Δ mcbR Δ Km /pBS(Kan);\nΔ mcbR /mcbR, E. coli MG1655 Δ mcbR Δ Km /pBS(Kan)- mcbR ; Δ mcbR /E93S-R139F, E. coli MG1655 Δ mcbR Δ Km /pBS(Kan)- mcbR -E93S-R139F; Δ mcbR /R89A-E93A-R139A, E. coli MG1655 Δ mcbR Δ Km /pBS(Kan)- mcbR -R89A-E93A-R139A. The Conformation of McbR Crystallized Is Likely Incompatible\nwith DNA Binding The wHTH domain is defined by helix α2,\na connecting turn, and helix α3 (HTH) and a small loop in the\nantiparallel β-sheet (the wing). The wHTH domain is slightly\nmore conserved than the FCD domain (5% identity, 13% similar) when\ncomparing the 7 structurally characterized FCD family members, with\nMcbR residues Leu20, Leu24, Leu29, Gly32, Leu35, Leu40, Leu44, Met46,\nVal51, Arg52, Glu53, Leu55, Leu58, and Leu64 being highly similar\n(Figure 4 A). The conserved hydrophobic residues\nfunction to stabilize the wHTH domain fold, whereas the two charged\nresidues are located at the wHTH dimerization interface (Figure 2 E). In McbR, the C-terminal portion of wHTH\nhelix α1 contributes to the top of the FCD binding cavity, with\nIle26 (helix α1) ∼12 Å away from the FCD domain\nligand coordinating arginines (Arg89 and Arg139; Figure 6 A). Thus, this wHTH–FCD\ninterface provides a conduit by which effector binding in the FCD\ndomain can be structurally communicated to wHTH DNA binding domain. 8 The conformation and orientation of the McbR\nwHTH domains appear to be incompatible with DNA binding. First, residues\n37–48, which comprise helix α2, are disordered in subunit\nB (Figure 2 A). Residues from helix α2\noften contribute to DNA recognition, as has been observed for the\nFCD transcription factor FadR. 9 Second,\nthe two domains in McbR differ not only in their relative orientations\nto the FCD domain but also in conformation, with an rmsd of 1.2 Å\n(Figure 6 B). This is due to a change in the\norientation of the wing between strands β2 and β3. Figure 6 N-Terminal\ndomain of McbR. (A) The pocket of α1 into the\nFCD domain. Helices α6 and α7 are colored teal (cartoon),\nand helix α1 is shown in light blue (sticks). (B) Superposition\nof McbR chain A (light blue/deep teal) and McbR chain B (light green/dark\ngreen). Whereas the C-terminal domains superimpose well (deep teal/dark\ngreen), the N-terminal domains (light blue/light green) do not. Implications for McbR Function Currently, E. coli FadR is the only\nmember of the GntR family\nwhose DNA-bound structure has been determined, 8 , 9 revealing\nthat FadR binds the short palindromic consensus sequence 5′-TGGNNNNNCCA-3′.\nPreviously, the E. coli McbR protein\nwas shown to bind upstream of the E. coli yciGFE promoter (P yciG ECO ). 6 Subsequent DNaseI footprinting identified two\ndistinct DNA sequences within P yciG ECO protected by McbR binding. To confirm that McbR binds this operator,\nwe performed EMSA experiments using WT McbR and P yciG ECO DNA. As shown in Figure 7 ,\nMcbR binds and shifts P yciG ECO DNA. Figure 7 McbR:P yciG ECO EMSA experiments. (A)\nSuperposition of the N-terminal domain of McbR (teal) and E. coli FadR (beige, PDB ID: 1HW2) bound to DNA. Residues\nmaking base-specific contacts in FadR and the structurally overlapping\nresidues in McbR are shown as sticks and labeled. (B) EMSA experiments\nusing biotin-labeled P yciG ECO and WT McbR\n(the migration of the DNA alone is shown in the left lane). (C) EMSA\nexperiments using biotin-labeled P yciG ECO and either WT McbR or the McbR variants as indicated; the migration\nof the DNA alone is shown in the left lane. All binding reactions\nin panels B and C contain the nonspecific poly(dI–dC) probe. In the FadR–DNA complex,\nArg35 (helix α2), Arg45 (helix\nα3), Thr46 (helix α3), and His65 (β2−β3\nwing) mediate base-specific contacts with the bound DNA. The corresponding\nresidues in McbR are Lys38, Ile48, Thr49, respectively, with no residue\ncorresponding to His65 (Figure 7 A). This suggests\nthat McbR likely interacts with DNA via helix α2 (Lys38) and\nhelix α3 (Thr49). Superposition of the FadR–DNA complex\nand McbR shows Gln70 as the only residue with a polar side chain in\nclose proximity with the DNA in the β2−β3 wing.\nAdditional basic residues in close proximity to the DNA include Arg34\nand Arg52 (Figure 7 A). We tested the role of\nthese residues in DNA binding using EMSA experiments performed with\nthe P yciG ECO promoter DNA and McbR mutants\n(we used CD to show that the variants are folded; Figure S1 ; the T m ’s of\nthe variants are within 3.7 °C of that of WT, which has a T m of 63.2 °C). The EMSA experiments show\nthat residues Arg34, Lys38, Thr49, and Arg52 are important for DNA\nbinding, as mutating these residues to alanine result in a loss of\nDNA binding compared to that of WT McbR (Figure 7 B). Furthermore, Arg34, Lys38, and Thr49 have the most debilitating\neffects, suggesting that β1, α2, and α3 play key\nroles in DNA binding. So, how is DNA binding regulated? As stated\nearlier, the GntR transcription\nfactors are typically regulated by ligands that are metabolic substrates/products/cofactors\nof the genes that they regulate. In many cases, these genes are often\nlocated next to or near the GntR gene itself. 7 McbR was previously shown to bind the promoter of yciGFE and ybiM . 4 , 6 While the molecular\nfunctions of the protein products of these genes are currently unknown, ybiM has been shown to effect colanic acid production in\na McbR-dependent manner, suggesting that colanic acid, or one of its\nconstituents, may be the biologically relevant ligand for McbR. 4 Currently, our results suggest that this is not\nthe case, as none of the components of colanic acid satisfactorily\nfit the ligand density in the McbR cavity. An examination of the genes\nnear mcbR in the E. coli chromosome shows that they are involved in a variety of biological\nprocesses (Table 5 ); our data again shows that\nMcbR is unlikely to be regulated by these metabolites (methionine,\ncurcumin/dihydrocurcumin, iron, asparagine, and glutathione), as they\nalso did not satisfactorily fit the density. However, sequence similarities\nbetween the E. coli and Salmonella McbR do suggest that they likely bind\nsimilar, if not identical, ligands. Namely, although the FCD domains\nof McbR from both organisms are less conserved than their corresponding\nwHTH domains (FCD domain sequence conservation: 46% identity, 74%\nsimilarity), the ligand binding residues are nearly perfectly conserved,\nincluding Arg89, Glu93, and Arg139 (Figure 4 A); the only differences in the ligand binding pocket are distal\nfrom the Arg-Glu-Arg pocket: Ile214 and Leu215 (Thr214 and Thr215\nin Salmonella ). Because these residues\nchange from hydrophobic ( E. coli ) to\npolar ( Salmonella ), the distal portion\nof the ligand may be slightly different between the organisms. Once\nthe biologically relevant ligand(s) of McbR have been confidently\nidentified, this ligand, or a derivative thereof, may be able to function\nas a novel therapeutic to target biofilms. Table 5 DNA Sequences\nSurrounding the mcbR ( yncC ) Gene\nin Escherichia\ncoli (MG1655) gene other names gene description yncA mnaT , b1448 Methionine\nN-acyltransferase; l -amino acid N-acyltransferase yncB curA , b1449 Curcumin/dihydrocurcumin\nreductase, NADPH-dependent yncD b1451 Predicted iron outer membrane\ntransporter yncE b1452 ATP-binding protein, periplasmic,\nfunction unknown yncF ansP , b1453 l -asparagine transporter yncG b1454 Glutathione S-transferase\nhomologue yncH b1455 Conserved protein, function\nunknown"
} | 5,992 |
20544086 | null | s2 | 6,857 | {
"abstract": "In this article, we developed a \"plant on a chip\" microfluidic platform that can control the local chemical environment around live roots of Arabidopsis thaliana with high spatial resolution using multi-laminar flow. We characterized the flow profile around the Arabidopsis root, and verified that the shear forces within the device ( approximately 10 dyne cm(-2)) did not impede growth of the roots. Our platform was able to deliver stimuli to the root at a spatial resolution of 10-800 microm. Further, the platform was validated by exposing desired regions of the root with a synthetic auxin derivative, 2,4-dichlorophenoxyacetic acid (2,4-D), and its inhibitor N-1-naphthylphthalamic acid (NPA). The response to the stimuli was observed using a DR5::GFP Arabidopsis line, where GFP expression is coupled to the auxin response regulator DR5. GFP expression in the root matched the position of the flow-focused stream containing 2,4-D. When the regions around the 2,4-D stimulus were exposed to the auxin transport inhibitor NPA, the active and passive transport mechanisms of auxin could be differentiated, as NPA blocks active cell-to-cell transport of auxin. Finally, we demonstrated that local 2,4-D stimulation in a approximately 10 microm root segment enhanced morphological changes such as epidermal hair growth. These experiments were proof-of-concept and agreed with the results expected based on known root biology, demonstrating that this \"root on a chip\" platform can be used to test how root development is affected by any chemical component of interest, including nitrogen, phosphate, salts, and other plant hormones."
} | 408 |
38930499 | PMC11205429 | pmc | 6,858 | {
"abstract": "Soil desertification is an important challenge in global soil management, and effectively and stably restoring soil function is an urgent problem. Using synthetic microbial communities (SynComs) is a burgeoning microbial strategy aimed at enhancing soil nutrients through functional synergies among diverse microorganisms; nevertheless, their effectiveness in restoring desertified soils remains unknown. In this study, we conducted a two-year field experiment using a SynCom constructed by in situ probiotic bacteria and set up control, chemical fertilizer, and combined SynCom–chemical fertilizer (combined fertilizer) treatments to investigate the linkage between microbial communities and soil multifunctionality in the soil surface layer (0–10 cm). Both the bacterial and fungal communities differed the most under the combined fertilizer treatment compared to the control. The bacterial communities differed more under treatments of the SynCom than the chemical fertilizer, while the fungal communities differed more under the chemical fertilizer treatment than the SynCom treatment. Regarding soil function, the SynCom strengthened the correlation between enzyme activities and both bacterial communities and functional properties. pH and available potassium were the main influencing factors under the chemical fertilizer and combined fertilizer treatments. The beta-diversity of the bacterial communities was significantly correlated with soil multifunctionality. Random forest analyses showed that the SynCom significantly enhanced the bacterial communities, driving soil multifunctionality, and that some potential microbial taxa drove multiple nutrient cycles simultaneously. In summary, the SynCom effectively increased the abundance of most carbon, nitrogen, and phosphorus functional genes as well as soil enzyme activities. The bacterial community composition contributed significantly to soil multifunctionality. Hence, the development of novel microbial agents holds significant potential for improving soil functionality and managing desertification.",
"conclusion": "5. Conclusions In conclusion, this study demonstrated that in desertified areas, the SynCom treatment was more effective for restoring soil multifunctionality compared to chemical fertilizers, with bacterial communities playing a major role. The SynCom induced significant changes in the bacterial communities, while the chemical fertilizers primarily affected fungal communities. While the chemical fertilizers increased the amount of soil nutrients, their association with microbial communities was weak. The SynCom significantly increased functional gene abundance, with biotic factors being the main drivers of microbial community and functional traits. The bacterial community under the SynCom was the primary driver of soil multifunctionality, with most potential microbial taxa being functionally diverse. The combined fertilization treatment shared microecological traits with the SynCom treatment and could be used for chemical fertilizer reduction in the future. Overall, our results highlight the positive impact of SynComs on microbial communities and ecological functions in desertified areas, providing insights into the use of microbial approaches to remediate soil environments.",
"introduction": "1. Introduction Land desertification, characterized by a decrease in soil cover and a gradual loss of ecological protection, is particularly prevalent in dry, semi-arid, and subhumid locations. External disturbances can easily disrupt fragile ecosystems, leading to soil nutrient imbalances and reduced fertility, thereby restricting plant growth. Fractional vegetation cover is crucial for controlling desertification. Measures such as setting up sand barriers, converting farmland to forests, and implementing grazing bans have achieved significant results in Northwest China [ 1 ]. While these natural restoration measures can slow desertification and restore vegetation growth, they pose challenges by being time-consuming and reducing food production [ 2 , 3 ]. Meanwhile, recent studies have confirmed that artificial vegetation restoration is more conducive to ecologically sustainable development [ 4 ]. According to local climate conditions and geographical environments, proper land management can optimize soil properties and increase vegetation productivity. Furthermore, artificial vegetation restoration has the advantages of a short cycle and strong targeting. Therefore, it is imperative to identify favorable measures to promote vegetation recovery for the ecological restoration of desertification. Chemical fertilizer is a frequently employed method in modern agriculture to provide nutrients essential for plant growth by increasing mineral nutrients. For example, in nitrogen-deficient desertified grasslands, the targeted application of nitrogen fertilizer can restore the growth of dominant grass species [ 5 ]. Yet, the rational utilization of chemical fertilizers is a challenge. Prolonged use or improper nutrient ratios can easily pollute the environment and accelerate soil degradation [ 6 ]. The development of microbial agents offers novel solutions for agricultural practices. Soil microorganisms can promote nutrient cycling and improve land productivity. For example, plant-growth-promoting rhizobacteria (PGPR) enhance nutrient acquisition, regulate physiological metabolism, and improve plant adaptability. Recently, synthetic microbial communities (SynComs) have emerged as a promising strategy, utilizing indigenous beneficial strains to enhance plant growth through functional integration [ 7 , 8 ]. Unlike commercial microbial inoculants, the microbial members of SynComs are better able to colonize the initial environment and interact with plants to achieve a “home-field advantage” [ 9 ]. An increasing body of research evidence indicates SynComs’ usefulness in enhancing plant productivity and resilience, especially in low-fertility areas [ 10 , 11 ]. SynComs can lead to a reduction in chemical fertilizer inputs and promote environmental sustainability. However, little is known about the effectiveness of SynComs in desertification management, and few research studies have looked specifically at their effects on soil biological functions. The functional characteristics and ecological niches of different microbial types sustain a wide range of ecological functions. According to studies, soil moisture has a substantial impact on the composition and function of soil microbial communities. Especially in desert soils, moisture is a key factor limiting microbial growth and metabolic activities. Increasing soil moisture not only increases microbial biomass, but also promotes organic matter decomposition, thereby improving soil fertility [ 12 , 13 ]. Different fertilization practices affect soil multifunctionality by altering microbial communities. Several studies have shown that the application of mineral fertilizers enhances microbial biomass and diversity and is more likely to affect fungal community structure [ 14 ]. However, irrational fertilization can inhibit microbial activity and reduce microbial communities’ ecological functions. SynComs restructure healthy microbial communities by introducing core functional strains [ 15 ]. A rising number of studies have proven that the application of SynComs enhances inter-root microbial synergism and improves plant productivity [ 16 ]. Positive correlations between microbial diversity and ecological function have been demonstrated in several ecosystems [ 17 ], with species richness favoring ecological function recovery. Microbial β-diversity, an important feature of community composition, has received little attention in studies involving ecological multifunctionality [ 18 ]. Therefore, whether the changes in microbial communities generated by SynComs are conducive to improving soil multifunctionality is key to their application. This knowledge is necessary for microbial ecological remediation and the restoration of ecosystem functionality in the context of global land degradation. As reported herein, we conducted a field experiment in a desertified area of Northwest China to investigate the association between microbial populations and soil multifunctionality at the soil surface (0–10 cm) using chemical fertilizer, SynCom, and SynCom–chemical fertilizer treatments. We hypothesized that the SynCom would alter the microbial communities to enhance soil multifunctionality, and that the bacterial communities would be more affected than the fungal communities. The objectives of this study were (1) to assess the ecological effects of SynComs in desertification management, and (2) to elucidate the link between microbial communities and ecological functions under different treatments. Our results demonstrated the critical role of microbial communities in desertification restoration and the positive effects of SynComs on ecological multifunctionality. This provides new ideas for solving environmental problems and contributes to ecological sustainability.",
"discussion": "4. Discussion 4.1. Fertilization Measures Have Altered Microbial Diversity and Composition Soil microorganisms are important for maintaining soil ecological balance and promoting plant growth. The level of microbial alpha-diversity increased under the different fertilizer applications in desertified areas ( Figure 2 a,e). The difference was that the chemical fertilizer (CF) only enhanced the fungal alpha-diversity, whereas the SynCom (SC) and combined fertilization (SCF) treatments significantly enhanced the bacterial and fungal alpha-diversity. External additives selectively impact microbial groups [ 29 ]. The SynCom affected the soil fertility by directly increasing the population of beneficial microorganisms, so the alpha-diversity changed as expected. Whereas chemical fertilizers enhance soil nutrients through nutrient supplementation, bacterial and fungal growth preferences and adaptations influence changes in microbial diversity. Inadequate nutrient accumulation may be responsible for the lack of significant changes in microbial alpha-diversity. The NMDS plot demonstrates that the fertilizing measures had a greater impact on the microbial population than the soil depth ( Figure 2 b,f). Meanwhile, the bacterial community was more susceptible to changing conditions than the fungal community. This reflects the fact that bacteria respond faster to changes in external factors, while fungi respond more slowly. This phenomenon has also been demonstrated in many ecosystem studies [ 30 ], which may be due to the faster growth rate of bacteria. A dissimilarity analysis showed that the SC treatment had a greater impact on the bacterial communities, while the CF treatment affected the fungal communities more significantly ( Figure 2 c,g). Interestingly, most of the bacterial genera enriched by the SC treatment are eutrophic bacteria. For example, Microvirga, Legionella, Polycyclovorans, and OM27_clade belong to the phylum Proteobacteria, and Ohtaekwangia belongs to the phylum Bacteroidetes ( Figure 3 c). Eutrophic bacteria can utilize abundant carbon sources for rapid growth and can be used in vegetation restoration [ 31 ]. However, this was not the case with the CF treatment, so the results suggest that the SC treatment altered the bacterial community in favor of the growth of eutrophic bacteria. Fungal communities are sensitive to CF [ 32 , 33 ], and our results are in agreement with this. The CF treatment increased the relative abundance of Mortierellomycota as a dominant phylum. Mortierellomycota often survive in nutrient-rich soils as saprophytes and have been strongly correlated with soil pH, nitrogen, and phosphorus nutrients in previous studies [ 34 ]. Combined fertilizer applications can avoid the problems of nutrient imbalance and poor structure associated with single applications [ 35 ]. In this study, the SCF treatment had the highest bacterial and fungal community differences compared to the CK treatment. This result suggests that the SCF treatment integrates the effects of two single fertilizer applications, proving highly beneficial in terms of fertilizer reduction and cost reduction. 4.2. SynCom Enhanced the Link between Microorganisms and Soil Biological Properties The composition of species can reflect the abundance changes in microbial communities. However, it should be noted that microorganisms within the same taxonomic group may not necessarily possess identical ecological functions. For example, different strains of the genus Pseudomonas exhibit significant differences in material degradation and plant growth promotion [ 36 ]. Therefore, we determined some C, N, and P cycle-related functional genes to understand the functional characteristics of microbial taxa ( Figure 4 and Table S4 ). The SC treatment significantly enhanced nitrogen fixation, nitrogen mineralization, nitrification, and denitrification processes in the N cycle. This suggests that soil microbial action promotes the N cycle. Microorganisms have also been demonstrated to play a vital role in controlling the nitrogen cycle in previous grassland restoration studies, in which a general increase in N inputs boosted vegetation restoration [ 37 ]. The AOA and AOB genes characterized the ammonia oxidation process of nitrification, and different types of ammonia-oxidizing microbial communities differed in terms of their sensitivity to the environment. The CF treatment only significantly increased AOB gene abundance, suggesting that it promoted nitrified nitrogen acquisition by enhancing ammonia-oxidizing bacteria. This result is consistent with that in the study of Allegrini et al. [ 38 ], which showed AOB genes are the main responders to the action of inorganic fertilizers. Here, we have found that the SC treatment has a universal effect on enhancing soil functional characteristics, which may be attributed to the reshaping of the microbial community, enabling multifunctional collaboration. C fixation and C degradation by microorganisms are important in maintaining soil C balance. In the present study, the SC treatment significantly enhanced cellulose degradation ( Fungcbhif gene) and reduced starch degradation ( GH31 gene). This suggests that the functional characteristics of microorganisms are influenced by nutrient limitations. Most of the C sources in barren environments are recalcitrant, which requires oligotrophic microorganisms for degradation to release nutrients [ 39 , 40 ]. Therefore, in the initial stage of vegetation restoration, it is beneficial to enhance nutrient accumulation and plant growth by enhancing the utilization of complex organic matter. In summary, the SC treatment significantly enhanced various functional characteristics, promoting nutrient restoration and vegetation growth in desertified areas. Additionally, we also found that the SC treatment maintained the functional stability of the soil surface, whereas many functional characteristics of the SCF treatment were affected by the soil depth. For example, the nifH , qnorB , and phoD genes were enhanced only at 0–2 cm by the SCF treatment and weakened by depth changes. These results suggest that the combined fertilization treatment may have weakened the impact of synthetic flora due to nutrient conflicts. Mantel visualizations were used to explore the links between microbial taxa and functional gene composition and soil factors ( Figure 5 ). In this study, AK and pH were the main influences on microbial community and function under the CF treatment. They had the same effect under the SCF treatment. This may be due to soil acidification caused by the addition of chemical fertilizer. Some studies have shown that pH is correlated with soil aggregate stability and affects carbon fluxes and available nutrients in desert grasslands [ 41 , 42 ]. The heat maps show that pH was also significantly correlated with seven soil properties, such as SOM and TN ( Figure 5 d), emphasizing pH as a key factor influencing microbial communities. In contrast, AN was an important influencing factor under the SC treatment. The enhancement of functional genes for several nitrogen cycling processes in this study suggests that the microbial regulation of N turnover is closely linked. The above results suggest that the SC and CF treatments used different nutrient strategies to influence the microbial communities and their functional traits. All enzyme activities significantly impacted microbial community composition and functional traits under the SC treatment. However, this was not found under the other treatments. This suggests that biotic factors are the main mode of action of microbial stimulation. Additionally, enzyme activity regulates nutrient cycling because most functional genes are involved in encoding the corresponding extracellular enzymes [ 43 ]. Numerous studies have demonstrated that beneficial microbial inoculation promotes enzyme activity in soil [ 44 ], and our study agrees with this view. 4.3. SynCom Increased the Contribution of Bacterial Communities to Soil Multifunctionality Our VPA analyses ( Figure S5 ) demonstrated how much the functional genes could be explained by the microbial community. In terms of C, N, and P cycling, the residuals of the CF treatment were similar to those of the CK treatment, while the SC treatment had the lowest residuals. The findings indicated that the SC treatment significantly enhanced the effect of the microbial community on multiple functional traits. This is consistent with our hypothesis that microbial communities play a greater role in soil functional improvement under SC treatments. Most studies have also confirmed that microbial taxa are significantly correlated with some ecological functions [ 45 , 46 ]. In this study, we suggest that the SC treatment may promote the effectiveness and synergy of microbial functions through the high adaptability of native microorganisms. Furthermore, we discovered that the bacterial community explained more than the fungal community, implying that the SC treatment may have enhanced the role of the bacterial community in restoring ecological services. In order to avoid the homogeneity of soil functions, we combined a variety of known soil indicators (enzyme activity, physicochemical properties, and functional gene abundance) to generate a multifunctionality index to explore the relationship between microbial communities and ecological multifunctionality in detail. Different agricultural management practices can enhance soil multifunctionality. Our results were in agreement with this, in that all of the fertilization treatments increased the multifunctionality index ( Figure 6 a). The SC treatment significantly enhanced soil multifunctionality at different depths in the top soil layer (0–10 cm). However, the CF treatment significantly increased multifunctionality only at 5–10 cm. This indicates that the SC treatment can effectively enhance soil multifunctionality in desertified areas, contributing to ecological restoration. There is growing evidence that microorganisms play a key role in regulating ecosystem functions [ 47 ]. The positive relationship between biodiversity and soil multifunctionality has been widely recognized, with specific microbial taxa driving multiple soil functions [ 48 , 49 ]. However, the effect of microbial community changes on ecological multifunctionality under different fertilization conditions remains unclear. Liu et al. [ 8 ]’s research has demonstrated that litter crusts in desert areas enhance soil nutrients by affecting bacterial communities but not fungal communities. The present study also proved that bacterial communities are more closely related to multifunctional indicators in desertified areas. Bacterial communities were linearly correlated with multifunctionality, whereas the correlation was weaker for fungi ( Figure 6 b). The SC treatment enhanced the linear relationship between the bacterial communities and functionality. A random forest model also confirmed that the SC treatment increased the importance of the bacterial communities. Therefore, the results are consistent with our second hypothesis that the bacterial community under the SynCom treatment plays a major role in soil multifunctional restoration. Bacteria exhibit a wide range of traits and functions, and they grow rapidly [ 17 ]. Therefore, SynComs are proven to be an effective measure for ecological restoration by regulating bacterial communities. 4.4. Potentially Functional Microbial Taxa Play an Important Role in Ecological Multifunctionality Microorganisms are key drivers of biogeochemical cycles. The random forest model demonstrated potential microbial predictors under the different treatments. Interestingly, key microbial genera under the SC and SCF treatments drove multiple elemental cycles, whereas key microbial genera under the CF treatments drove single elemental cycles. Herpetosiphon spp. are a class of filamentous slithering predatory bacteria capable of producing a wide range of secondary metabolites and hydrolytic enzymes [ 50 ]. The predator is able to selectively lyse its prey to release nutrients that promote the growth of other microorganisms. As a result, they play a significant role in reshaping microbial communities. Previous studies have reported that fertilization with added nutrients had less of an effect on microbial communities than predators [ 51 ]. This also explained why the SynCom did not introduce excessive nutrients, but was able to significantly change the structure of bacterial and fungal communities. Fusicolla spp. are important inter-root fungi and have an important role in biodegradation and plant promotion [ 52 ]. Zhu et al. [ 53 ] found a positive correlation between Fusicolla and soil nitrogen and phosphorus, which is consistent with our findings. These potentially key microorganisms can participate in multiple nutrient cycles, occupy a wide range of ecological niches, and enhance soil nutrient multifunctionality. These results suggest that SynCom highlights the importance of potentially functional taxa in ecological multifunctionality. Future studies should focus more on the physiological activities and metabolic characteristics of these multifunctional microbial taxa. This knowledge is crucial for us to apply microbial strategies for environmental remediation."
} | 5,611 |
37640908 | null | s2 | 6,859 | {
"abstract": "The combination of lithographic methods with two-dimensional DNA origami self-assembly has led, among others, to the development of photonic crystal cavity arrays and the exploration of sensing nanoarrays where molecular devices are patterned on the sub-micrometre scale. Here we extend this concept to the third dimension by mounting three-dimensional DNA origami onto nanopatterned substrates, followed by silicification to provide hybrid DNA-silica structures exhibiting mechanical and chemical stability and achieving feature sizes in the sub-10-nm regime. Our versatile and scalable method relying on self-assembly at ambient temperatures offers the potential to three-dimensionally position any inorganic and organic components compatible with DNA origami nanoarchitecture, demonstrated here with gold nanoparticles. This way of nanotexturing could provide a route for the low-cost production of complex and three-dimensionally patterned surfaces and integrated devices designed on the molecular level and reaching macroscopic dimensions."
} | 261 |
37640908 | null | s2 | 6,860 | {
"abstract": "The combination of lithographic methods with two-dimensional DNA origami self-assembly has led, among others, to the development of photonic crystal cavity arrays and the exploration of sensing nanoarrays where molecular devices are patterned on the sub-micrometre scale. Here we extend this concept to the third dimension by mounting three-dimensional DNA origami onto nanopatterned substrates, followed by silicification to provide hybrid DNA-silica structures exhibiting mechanical and chemical stability and achieving feature sizes in the sub-10-nm regime. Our versatile and scalable method relying on self-assembly at ambient temperatures offers the potential to three-dimensionally position any inorganic and organic components compatible with DNA origami nanoarchitecture, demonstrated here with gold nanoparticles. This way of nanotexturing could provide a route for the low-cost production of complex and three-dimensionally patterned surfaces and integrated devices designed on the molecular level and reaching macroscopic dimensions."
} | 261 |
22866957 | null | s2 | 6,861 | {
"abstract": "Bacteria have developed cell-to-cell communication mechanisms, termed quorum sensing (QS), that regulate bacterial gene expression in a cell population-dependent manner. Autoinducer-2 (AI-2), a class of QS signaling molecules derived from (4S)-4,5-dihydroxy-2,3-pentanedione (DPD), has been identified in both Gram-negative and Gram-positive bacteria. Despite considerable interest in the AI-2 QS system, the biomolecular communication used by distinct bacterial species still remains shrouded. Herein, we report the synthesis and evaluation of a new class of DPD analogues, C4-alkoxy-5-hydroxy-2,3-pentanediones, termed C4-alkoxy-HPDs. Remarkably, two of the analogues were more potent QS agonists than the natural ligand, DPD, in Vibrio harveyi. The findings presented extend insights into ligand-receptor recognition/signaling in the AI-2 mediated QS system."
} | 215 |
28948235 | PMC5602812 | pmc | 6,862 | {
"abstract": "Energy harvesting is a promising technology that powers the electronic devices via scavenging the ambient energy. Piezoelectric energy harvesters have attracted considerable interest for their high conversion efficiency and easy fabrication in minimized sensors and transducers. To improve the output capability of energy harvesters, properties of piezoelectric materials is an influential factor, but the potential of the material is less likely to be fully exploited without an optimized configuration. In this paper, an optimization strategy for PVDF-based cantilever-type energy harvesters is proposed to achieve the highest output power density with the given frequency and acceleration of the vibration source. It is shown that the maximum power output density only depends on the maximum allowable stress of the beam and the working frequency of the device, and these two factors can be obtained by adjusting the geometry of piezoelectric layers. The strategy is validated by coupled finite-element-circuit simulation and a practical device. The fabricated device within a volume of 13.1 mm 3 shows an output power of 112.8 μW which is comparable to that of the best-performing piezoceramic-based energy harvesters within the similar volume reported so far.",
"conclusion": "4 Conclusion An optimization strategy based on an uncoupled point-mass-model analysis on PVDF-based PEH devices has been presented in this paper. It is shown that with given vibration environment, the power density of the PVDF-based PEH is related to the maximum stress in the PVDF beam that is limited by the yield strength of the material. By properly selecting t p L e 2 , the maximum allowed stress in the PVDF beam can be achieved, thus enabling maximum power density output. The resonant frequency also affects proportionally the output power density, but it is subject to the vibration source which generally has low frequency. An example of optimized device has been proposed according to the strategy. The optimal output power density of the device under the maximum allowable stress can be obtained as high as 15.4 mW/cm 3 . A finite element simulation method was adopted and a practical device was fabricated to verify the analytical model. In accordance with the exact design calculated from the model, simulation results and experimental measure fit well in terms of the resonant frequency (34.4 Hz) and the maximum stress (28.5 MPa). The power output and power density were 112.8 μW and 8.61 mW/cm 3 respectively, which was lower than the analytical result, because the influence of the external resistance and the additional capacitance are ignored in the analytical model. Nevertheless, the measured power density is still the highest as far as we have seen in the literature of PVDF-based PEH devices, which proves the efficiency of our optimization approach. It is noted that the impact of additional capacitance can be further diminished by selective patterning of the electrode on the beam and thus a higher output power density can be expected.",
"introduction": "1 Introduction There are many unutilized energy sources in the environment, e.g. thermal energy, electromagnetic waves, and mechanical vibration [1] . The conversion of these ambient energies into electric energy has motivated the development of the energy harvesting devices (EHDs). Meanwhile, the ongoing rapid reduction in the power consumption of electronic devices and micro-electro-mechanical systems (MEMS) make the application of EHDs increasingly promising [ 2 , 3 ]. Mechanical vibration exists broadly in the environment and human body movement, thus EHDs for mechanical vibration show great potential for wide applications, for example, in wireless sensor networks where the battery replacement is a crucial issue [ 4 , 5 ]. EHDs which scavenge energy from human walk can play an important role as a backup power for artificial heart pacemakers in which the power cutoff is fatal [ 6 , 7 ]. Among various routes to collect vibration energy, piezoelectric energy harvesting (PEH) has the advantage of simple device structure, high power density and no need for extra power supply, thus has attracted extensive investigations [ 8 , 9 ]. Most of the previous studies on the PEH devices have chosen piezoceramics as the key material for the mechanical-electrical energy conversion [ 10 , 11 , 12 , 13 , 14 , 15 , 16 ]. Piezoceramics such as lead zirconate titanate (Pb[Zr x Ti 1-x ]O 3 , or PZT) possess high electromechanical coupling coefficient, high mechanical quality factor and large elastic stiffness, which makes them good candidates for the energy-conversion element in PEH devices. However, there has been severe concern on the environmental impact due to the heavy content of Pb in the high-end piezoceramics in recent years [17] . In addition, as piezoceramics are fragile and unable to bear large deformation, a nonpiezoelectric substrate layer is usually necessary in the construction of piezoceramics-based PEH which however degrades the electromechanical coupling coefficient of the whole device. Compared to piezoceramics, piezoelectric polymers such as polyvinylidene fluoride (PVDF) show good environmental compatibility and their flexibility makes it feasible to adopt a full piezoelectric beam without any substrate layers. In addition, the device based on piezoelectric polymers is more resistant to mechanical shock. The off-resonance figure of merit ( d ij g ij ) of PVDF for energy scavenging is as good as that of PZT ceramics [18] . There have been experimental attempts to use PVDF on off-resonance mode for energy harvesting. For instance, Kymissis et al. [19] installed an insole stave made of PVDF film-stacks in shoes to collect the energy due to bending of soles during human walking and obtained an average power output of ∼1 mW. Vatansever et al \n [20] used PVDF films to collect energy from wind and rain drops and showed that piezoelectric polymer materials can generate power more efficiently than piezoceramics. In 2003, Roundy et al \n [21] calculated the power output of cantilever-type PEH devices working on resonant mode with a seismic mass placed on the free end by means of coupled modal analysis. The derived maximum power output for a PVDF-based PEH device fabricated within 1 cm 3 reached over 200 μW at 120 Hz, under an input vibration with an acceleration magnitude of 0.25g (g = 9.8 m/s 2 ), comparable to PZT-based PEH devices. However, the experimental result on the performance of PVDF-based PEH devices was far inferior to what Roundy et al estimated. Jiang et al \n [22] obtained only around 16 μW at 17 Hz resonant frequency under 1.2g sinusoidal vibration and Cao et al \n [23] reported a power output of 3.0 μW in air and 10.6 μW in vacuum, under 1.0g vibration input at ∼100 Hz. The output power density (output power per beam volume) is ∼176 μW/cm 3 \n [22] and ∼1943 μW/cm 3 \n [23] in the two experimental studies while it reaches as large as 30000 μW/cm 3 \n [21] in the calculation of Roundy et al . It should be noted that as far as we know, none of the experimental attempts have fabricated the PVDF-based PEH devices through configuration optimization process. It is hence worthy to develop a facile and reliable optimization approach for PVDF-based PEH devices based on analytical models. In this paper, an uncoupled model is introduced to study how the configuration of PVDF-based PEH devices affects their output power density that is used as a criterion for high-performance PEH devices. Strategies for the configuration optimization have been formulated according to the analytical results and an example of optimized device has been given. A coupled finite-element-circuit simulation (CFECS) approach is conducted to justify the approximations used in the analytical model as well as to prove the structure-performance relation that is deduced from the analytical model. We also fabricated a real device according to the optimized configuration and experimentally characterized its performance under the given vibration condition. Both the finite element simulation and experimental data demonstrate the high output power density of the optimized device, proving the reliability and efficiency of our optimization approach.",
"discussion": "3 Results and discussion 3.1 CFECS approach Recently, the CFECS approach [31] has been proposed to simulate the performance of PZT-based PEH devices connected to external electric loads and shows good reliability. We thus conducted CFECS for PVDF-based PEH devices on the commercial software COMSOL Multiphysics 4.2 to check the two assumptions made in the derivation of the analytical results as well as to simulate the performance of the exemplar device proposed in the previous section. The mesh distribution of the model are shown in Fig. 4 with the boundary condition “prescribed displacement” as excitation and “floating potential” as the surface electrode which connected external resistance. Fig. 4 CFECS model in COMSOL. Fig. 4 Fig. 5 shows the simulation of the frequency-dependent vibration amplitude ( A ) of the proposed optimized device under different circuit conditions, with the vibration acceleration maintained at 0.5g upon varying the frequency. It is seen that the resonant frequency changes only by less than 0.3 Hz (<1%) with external resistance spanning zero, optimum resistive load and infinity, confirming that the load-induced frequency shift can be safely neglected. The vibration amplitude also has a weak dependence on the connected external resistance. It decreases a bit (by <10%) with optimum resistance load compared to that under the open-circuit and short-circuit condition, as the converse piezoelectric effect counter plays with the mechanical excitation. This indicates that slight overestimate of output power might occur in the uncoupled model. Fig. 5 Simulation result of the frequency response to vibration amplitude in different circuit states under constant 0.5g vibration acceleration. Fig. 5 3.2 Experimental results and discussions The experimental setup for the device test is shown in Fig. 6 a and Fig. 6 b. A function generator (RIGOL DG1022U) is used to generate a sinusoidal signal. Amplified by the power amplifier (SINOCERA YE5872A), the signal is sent to the vibrator (SINOCERA JZK-5). A vibration sensor (SINOCERA CA-YD-1181) is attached on the vibrator, noting that the accelerometer is placed as close as possible to the attached beam end. The accelerometer processes a sensitivity of 100 mV/g and a signal amplifier (SINOCERA YE3822A) is used to magnify the output signal. Both the sensor signal and the PVDF-based PEH voltage output are measured by a digital oscilloscope (RIGOL DS1102E). Note that the optimum external resistive load is rather high, a 100MΩ 1.7pF oscilloprobe (RIGOL RP1300H) is used to diminish current shunt effect. Fig. 6 (a) Instruments used for device measurement and (b) their assembling sketch. Fig. 6 The resonant frequency of the fabricated device is measured to be 34.4 Hz from the frequency sweep of the output voltage under open-circuit condition. The performance of the device as a function of external load was then measured at 34.4 Hz under 0.5g excitation acceleration. The output power value is obtained according to P = V p − p 2 8 R where V p-p is peak-to-peak voltage measured with the oscilloscope. The measured output voltage and output power are presented in Fig. 7 . A peak power value of 112.8 μW appears at 6.5–7.0 MΩ, in good agreement with the optimum resistive load R opt of 6.81 MΩ calculated from the overall actual capacitance of the device. The output power density of the fabricated device is 8.61 mW/cm 3 , which is the highest among the reported PVDF-based PEH devices and even comparable to high-performance PZT-based PEH devices (as listed in Table 3 ). Fig. 7 The effects of external loads on V p-p and P at 0.5g excitation and resonant frequency. Fig. 7 Table 3 Comparison of the cantilever PEH devices in our work and reported literature. Table 3 Piezoelectric layers ACC f res (Hz) P max (μW) (P/V) max (mW/cm 3 ) Ref. PVDF 0.5g 34.4 112.8 8.61 This work PZT 0.25g 109.5 335.2 0.296 [11] PZT 0.6g 42 0.114 1.48 [13] PVDF 1.2g 17 16 0.176 [22] PVDF 1g 103.8 10.6 1.94 [23] PZT 1g 29.6 15300 13.5 [29] PMN-PT 0.23g 174 586 0.753 [30] It should be noticed that the measured value 8.61 mW/cm 3 is clearly lower than the expected 15.4 mW/cm 3 from the analytical model. Two factors account for the difference between the analytically predicted output power density P ana /V and that measured in the practical device P pra /V. The first is the impact of the additional capacitance. The PVDF part clamped in the fixed end as well as between the two pieces of stiff mass do not deform to generate the potential. Wiring/electroding of these parts leads to a decrease in the output power compared to the case with only the deformed PVDF part electrode, with a relationship of P pra /P ana = C deform /C total where C deform is the capacitance of the deformed part of the PVDF beam while C total is the capacitance of the overall beam. The overall capacitance of the fabricated device is measured to be 720 pF, while the capacitance of the deformed part is estimated to be 488 pF, leading to P pra /P ana = 0.68. Secondly, as pointed out previously, the overestimation of vibration amplitude of the beam and the corresponding overestimation of the strain, which also leads to overestimation of output power, following a law of P pra /P ana = (A prac /A ana ) 2 . As shown in Fig. 5 , A prac /A ana ≈ 0.91, leading to P pra /P ana = 0.83. Summarizing the impact of the two factors, P pra /P ana ≈ 0.53, well explaining the origin of the difference between measured output power density of 8.61 mW/cm 3 and the predicted value of 15.4 mW/cm 3 in the uncoupled model. The above argument was also confirmed by the CFECS results, which include the effect of both the additional capacitance and the converse piezoelectric in the modeling. As shown by blue curve in Fig. 7 , the simulated output voltage and power as a function of external resistive load are in good agreement with the measured data. The frequency-dependent output power was then measured at R = R opt under 0.5g acceleration, and the result is shown in Fig. 8 . It is seen that the optimum working resonant frequency is 34.4 Hz, closely matching the designed frequency of 35 Hz. The full width at half maximum (FWHM) is about 3.5 Hz. Fig. 8 Simulation and measurement result of the frequency response to output power with optimum working resistance. Fig. 8 The output performance under different vibration accelerations (ACC) were measured at the optimum working frequency with optimum external load, and the result is shown in Fig. 9 . According to Eq. (8) and Eq. (14) , the output power density increases quadratically with the vibration acceleration ACC. The CFECS also indicates that P∝ACC 2 and V p-p ∝ACC, as shown with the blue dash line and black solid line in Fig. 9 . It is seen that the device behaves as predicted when ACC ≤ 0.5g while fails in following the expected output power for ACC ≥ 0.625g. The maximum stress in the PVDF beam occurs at the top or bottom surface of the fixed end of the beam and it reaches 28.5 MPa (close to the analytical result) under 0.5g vibration excitation as simulated by CFECS ( Fig. 10 ). As the stress approaches σ yield , the output power is lower than that simulated from the linear stress-strain relationship, because PVDF begins to yield and some non-linear effects start to be prominent. Therefore, the power under 0.5g vibration excitation has been proved to be the maximum one under the allowable stress and the proposed device configuration is well-optimized. Fig. 9 Influence of vibration acceleration on V p-p and P of the device which is excited at resonant frequency and connected to optimum resistance load. Fig. 9 Fig. 10 The distribution of stress in the x -direction simulated with COMSOL, showing that the maximum stress reaches 28.5 MPa. (a) The distribution of stress in x direction. (b) The distribution of stress on the whole surface. Fig. 10\n\n3.2 Experimental results and discussions The experimental setup for the device test is shown in Fig. 6 a and Fig. 6 b. A function generator (RIGOL DG1022U) is used to generate a sinusoidal signal. Amplified by the power amplifier (SINOCERA YE5872A), the signal is sent to the vibrator (SINOCERA JZK-5). A vibration sensor (SINOCERA CA-YD-1181) is attached on the vibrator, noting that the accelerometer is placed as close as possible to the attached beam end. The accelerometer processes a sensitivity of 100 mV/g and a signal amplifier (SINOCERA YE3822A) is used to magnify the output signal. Both the sensor signal and the PVDF-based PEH voltage output are measured by a digital oscilloscope (RIGOL DS1102E). Note that the optimum external resistive load is rather high, a 100MΩ 1.7pF oscilloprobe (RIGOL RP1300H) is used to diminish current shunt effect. Fig. 6 (a) Instruments used for device measurement and (b) their assembling sketch. Fig. 6 The resonant frequency of the fabricated device is measured to be 34.4 Hz from the frequency sweep of the output voltage under open-circuit condition. The performance of the device as a function of external load was then measured at 34.4 Hz under 0.5g excitation acceleration. The output power value is obtained according to P = V p − p 2 8 R where V p-p is peak-to-peak voltage measured with the oscilloscope. The measured output voltage and output power are presented in Fig. 7 . A peak power value of 112.8 μW appears at 6.5–7.0 MΩ, in good agreement with the optimum resistive load R opt of 6.81 MΩ calculated from the overall actual capacitance of the device. The output power density of the fabricated device is 8.61 mW/cm 3 , which is the highest among the reported PVDF-based PEH devices and even comparable to high-performance PZT-based PEH devices (as listed in Table 3 ). Fig. 7 The effects of external loads on V p-p and P at 0.5g excitation and resonant frequency. Fig. 7 Table 3 Comparison of the cantilever PEH devices in our work and reported literature. Table 3 Piezoelectric layers ACC f res (Hz) P max (μW) (P/V) max (mW/cm 3 ) Ref. PVDF 0.5g 34.4 112.8 8.61 This work PZT 0.25g 109.5 335.2 0.296 [11] PZT 0.6g 42 0.114 1.48 [13] PVDF 1.2g 17 16 0.176 [22] PVDF 1g 103.8 10.6 1.94 [23] PZT 1g 29.6 15300 13.5 [29] PMN-PT 0.23g 174 586 0.753 [30] It should be noticed that the measured value 8.61 mW/cm 3 is clearly lower than the expected 15.4 mW/cm 3 from the analytical model. Two factors account for the difference between the analytically predicted output power density P ana /V and that measured in the practical device P pra /V. The first is the impact of the additional capacitance. The PVDF part clamped in the fixed end as well as between the two pieces of stiff mass do not deform to generate the potential. Wiring/electroding of these parts leads to a decrease in the output power compared to the case with only the deformed PVDF part electrode, with a relationship of P pra /P ana = C deform /C total where C deform is the capacitance of the deformed part of the PVDF beam while C total is the capacitance of the overall beam. The overall capacitance of the fabricated device is measured to be 720 pF, while the capacitance of the deformed part is estimated to be 488 pF, leading to P pra /P ana = 0.68. Secondly, as pointed out previously, the overestimation of vibration amplitude of the beam and the corresponding overestimation of the strain, which also leads to overestimation of output power, following a law of P pra /P ana = (A prac /A ana ) 2 . As shown in Fig. 5 , A prac /A ana ≈ 0.91, leading to P pra /P ana = 0.83. Summarizing the impact of the two factors, P pra /P ana ≈ 0.53, well explaining the origin of the difference between measured output power density of 8.61 mW/cm 3 and the predicted value of 15.4 mW/cm 3 in the uncoupled model. The above argument was also confirmed by the CFECS results, which include the effect of both the additional capacitance and the converse piezoelectric in the modeling. As shown by blue curve in Fig. 7 , the simulated output voltage and power as a function of external resistive load are in good agreement with the measured data. The frequency-dependent output power was then measured at R = R opt under 0.5g acceleration, and the result is shown in Fig. 8 . It is seen that the optimum working resonant frequency is 34.4 Hz, closely matching the designed frequency of 35 Hz. The full width at half maximum (FWHM) is about 3.5 Hz. Fig. 8 Simulation and measurement result of the frequency response to output power with optimum working resistance. Fig. 8 The output performance under different vibration accelerations (ACC) were measured at the optimum working frequency with optimum external load, and the result is shown in Fig. 9 . According to Eq. (8) and Eq. (14) , the output power density increases quadratically with the vibration acceleration ACC. The CFECS also indicates that P∝ACC 2 and V p-p ∝ACC, as shown with the blue dash line and black solid line in Fig. 9 . It is seen that the device behaves as predicted when ACC ≤ 0.5g while fails in following the expected output power for ACC ≥ 0.625g. The maximum stress in the PVDF beam occurs at the top or bottom surface of the fixed end of the beam and it reaches 28.5 MPa (close to the analytical result) under 0.5g vibration excitation as simulated by CFECS ( Fig. 10 ). As the stress approaches σ yield , the output power is lower than that simulated from the linear stress-strain relationship, because PVDF begins to yield and some non-linear effects start to be prominent. Therefore, the power under 0.5g vibration excitation has been proved to be the maximum one under the allowable stress and the proposed device configuration is well-optimized. Fig. 9 Influence of vibration acceleration on V p-p and P of the device which is excited at resonant frequency and connected to optimum resistance load. Fig. 9 Fig. 10 The distribution of stress in the x -direction simulated with COMSOL, showing that the maximum stress reaches 28.5 MPa. (a) The distribution of stress in x direction. (b) The distribution of stress on the whole surface. Fig. 10"
} | 5,596 |
35423966 | PMC8698205 | pmc | 6,863 | {
"abstract": "Self-healing efficiency and mechanical strength are always a pair of mechanical contradictions of a polymer. Herein, a series of novel mussel-inspired modified graphene oxide/polyurethane composites were successfully fabricated via rational molecular design and introducing hyperbranched polymer-modified graphene oxide. The composites exhibit outstanding self-healing performances with a self-healing efficiency of 87.9%. Especially, their self-healing properties possess exceptional water-insensitivity, which presents a high self-healing efficiency of 92.5% under 60 °C water for 2 h and 74.6% under 25 °C water for 6 h. Furthermore, the tensile strength of the composites increased by 107.7% with a high strain of 2170%. In addition, the composites show a remarkable recovery capability of 76.3% and 83.7% under tensile and compression loading, respectively, after 20 cycles. This strategy shows prominent application potential in high-performance solid propellants, protective coating, electronic skin, soft sensors and other water-insensitive devices.",
"conclusion": "4. Conclusions In this work, a series of novel mussel-inspired modified graphene oxide/polyurethane composites were designed and successfully synthesized. Owing to the successful graft of hyperbranched polymer onto GO sheets, MGO can crosslink with polymer chains. As a result, MGO could dramatically improve the mechanical properties of composites. The tensile strength of SPUM-2 increased by 107.7% with a high strain of 2170%. Notably, SPUM-2 also exhibits outstanding deformation recovery capability under cycle tensile and compression loading with recovery degree of 76.3% and 83.7%, respectively, after 20 cycles. Furthermore, owing to the synergistic effect of aromatic disulfide bonds and catechol groups, SPUM-2 shows a high self-healing efficiency of 87.1% at room temperature for only 6 h. More importantly, the self-healing performance of SPUM-2 exhibits excellent water-insensitivity with high self-healing efficiency of 92.5% under 60 °C water for 2 h and 74.6% under 25 °C water for 6 h. In summary, the synthesized HTPB-based PU composites possess excellent mechanical and self-healing properties, simultaneously, which may be potentially applied in solid propellants, protective coating, electronic skin, soft sensors and other water-insensitive devices.",
"introduction": "1. Introduction Polyurethane (PU), a kind of polymeric material with unmatchable properties, has aroused widespread attention in numerous fields, such as insulation, coating and diaphragms. 1,2 Hydroxyl-terminated polybutadiene (HTPB)-based PU is one of the promising materials for composite solid propellant binders and protective coating 3–5 owing to its outstanding elasticity, toughness and water-resistance, as well as sub-ambient glass transition temperature ( T g ). 6–10 However, as a kind of conventional thermosetting polymeric material with irreversible crosslinking, it cannot be repaired, once damaged. Over the past decade, self-healing technology has attracted considerable attention, which provides a feasible route to address the above-mentioned problems. Commonly, there are two strategies to achieve self-healing abilities, including extrinsic and intrinsic self-healing, according to the self-healing mechanisms. 11–13 The extrinsic self-healing materials generally require repairing agents stored in pre-embedded microcapsules or microvascular networks to achieve the self-healing process, showing a drawback of single/few-time healing with the exhaustion of repairing agents. 14–16 In contrast, the intrinsic self-healing materials can achieve repeated repair without external repairing agents by dynamic reversible covalent bonds and non-covalent interactions, such as hydrogen bond, 17,18 supramolecular interaction, 19,20 metal–ligand interaction, 21 π–π interaction, 22 Diels–Alder chemistry, 23 host–guest interactions, 24 and disulfide bond exchange. 25–27 Recently, aromatic disulfide bonds have been determined to be a promising reversible bond to achieve self-healing at room temperature owing to their low bond dissociation energy and stable free radicals. 28,29 Moreover, reports have also pointed out that hydrogen bonds and disulfide bonds can synergistically promote the self-healing process. To be specific, hydrogen bonds might accelerate initial sticking or interfacial adhesion, promoting the diffusion of molecular chains and accelerating the exchange reactions of disulfide bonds. 2,30 Along with this, disulfide bonds will weaken the strength of hydrogen bonds, leading to their easy dissociation under mild conditions. 28 Hence, a polymer with rapid room-temperature self-healing capability may be obtained by introducing both hydrogen bonding network and aromatic disulfide bonds into a polymer network simultaneously. Nature, during a long evolutionary process, has evolved numerous remarkably intelligent biological systems and marine mussel is one of them. Nowadays, marine mussel has aroused considerable attention in various fields, especially in self-healing materials. 31–33 The mussel-inspired self-healing materials originate from the fact that the high content of catechol groups in the mussel adhesive proteins can form strong covalent and non-covalent bonds with various surfaces, 32,34,35 thereby possessing outstanding adhesion ability. Especially, the dihydroxy of catechol groups can form bidentate hydrogen and metal-mediated reversible coordination, which is important for the self-healing process. 32,33 Moreover, mussel adhesive protein exhibits outstanding adhesion properties in an extremely harsh environment of the ocean, 36 and the mussel-inspired catechol groups can be well-bonding and maintain high adhesion underwater. 37 Inspired by the above-mentioned features, it is expected to design an HTPB-based PU with underwater and room-temperature self-healing ability based on disulfide bonds and mussel-mimetic hydrogen bonds, which is significant and consistent with its long-term working environment. 38 Nevertheless, we found that HTPB-based PU based on the above design had a low mechanical strength during our study process, which severely limited its applications in many fields. In fact, mechanical strength and self-healing efficiency are always a pair of mechanical contradictions of the polymer. 39–43 It is difficult and necessary to develop materials that possess both adequate mechanical strength and rapid self-healing speed. Recently, graphene oxide (GO), a specific derivative of graphene, has been considered a promising reinforcer for polymer composites to achieve high strength because of its remarkable properties, 44–46 and some studies have attempted to add GO to develop novel self-healing materials. 47–49 However, it has been reported that GO tends to aggregate in the polymeric matrix and has poor interfacial interaction with the matrix. 50,51 Therefore, the current strategy is to attach some active groups onto the surface of GO sheets to react and improve the compatibility between GO and matrix. 4,52 Consequently, the above-mentioned information prompted us to believe that integrating modified GO with rational molecule design might generate a novel composite that has not only better self-healing properties but also improved mechanical properties. Herein, we prepared a series of composites with excellent mechanical properties and rapid room-temperature self-healing performances. In our design, the incorporation of modified GO mainly enables the composite with excellent mechanical strength, while disulfide bonds and mussel-mimetic hydrogen bonds within the polymer chains provide self-healing ability. Effects of modified GO on the mechanical and self-healing properties were investigated carefully.",
"discussion": "3. Results and discussion 3.1. Design, synthesis and characteristics of materials First of all, we modified GO with amino-terminated hyperbranched polyamide (HBPA) to improve the compatibility and interfacial interactions of GO with the polymer matrix. We chose HBPA as the modification agent mainly based on the following considerations: HBPA is a kind of hyperbranched polymer with three-dimensional branching structures, nonentangled architectures, low solution viscosity and ultrahigh solubility, which have aroused increasing attention in reinforcing polymer matrix recently. 52–54 Meanwhile, the abundant reactive terminal groups on its molecular chains can react with GO and polymer matrix simultaneously, resulting in a stronger interfacial interactions and adhesion between GO and polymer matrix and beneficial to improve the mechanical properties. 55–57 The modification of GO with HBPA was determined using FTIR spectra. As shown in Fig. 1(a) , the characteristic absorption peaks of GO at 1728 cm −1 , 1628 cm −1 and 1053 cm −1 correspond to stretching vibrations of C \n \n\n<svg xmlns=\"http://www.w3.org/2000/svg\" version=\"1.0\" width=\"13.200000pt\" height=\"16.000000pt\" viewBox=\"0 0 13.200000 16.000000\" preserveAspectRatio=\"xMidYMid meet\"><metadata>\nCreated by potrace 1.16, written by Peter Selinger 2001-2019\n</metadata><g transform=\"translate(1.000000,15.000000) scale(0.017500,-0.017500)\" fill=\"currentColor\" stroke=\"none\"><path d=\"M0 440 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z M0 280 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z\"/></g></svg>\n\n O, C C and C–O, respectively, and the wide absorption peak at 3300–3000 cm −1 is ascribed to stretching vibrations of O–H in hydroxyl and carboxylic acid. 58,59 In the case of MGO, after modification, the peak at 1728 cm −1 disappeared while some new absorption peaks at 3250 cm −1 , 2959–2850 cm −1 , 1640 cm −1 and 1555 cm −1 appeared, which are assigned to stretching vibrations of N–H groups, stretching vibrations of –CH 3 and –CH 2 groups, amide C O stretching (amide I), amide N–H bending and C–N stretching (amide II), respectively, implying that GO was successfully modified with HBPA via the reaction between the terminal amine of HBPA and carboxylic acid groups of GO. 51,60 XRD patterns of GO and MGO are shown in Fig. 1(b) . Due to the incorporation of HBPA onto the surface of GO, the 2 θ peak of MGO was shifted down to 9.05°, indicating a larger interlayer spacing of 0.97 nm. It is known that the peak intensity ratio of D and G bands ( I D / I G ) on Raman spectra ( Fig. 1(c) ) is related to disordered structures of carbon, and the lower I D / I G ratio corresponds to fewer defects. 61 After functionalization, the I D / I G value of MGO increases slightly from 0.926 to 1.007. Obviously, more defects are formed, reflecting the successful modification with HBPA. Fig. 1 (a) FTIR spectra of GO, MGO and HBPA. (b) XRD spectra of GO and MGO. (c) Raman spectra of GO and MGO. (d) TGA spectra of GO, MGO and HBPA. TGA was also conducted to confirm the success of functionalization of GO ( Fig. 1(d) ). It can be seen that there are two main loss stages on the TGA curve of GO, which can be ascribed to the loss of absorbed water and the decomposition of labile oxygen-containing groups on the surface of GO, respectively. 62 As for HBPA, there is only a sharp decomposition stage from 200 °C to 460 °C and the residual weight was only 0.54 wt%. In the case of MGO, the weight loss of the second decomposition stage was significantly reduced compared to that of GO. It can be explained in a way that during the modification procedure, HBPA replaced some oxygen-containing groups via the reaction between –COOH of GO and –NH 2 of HBPA. Notably, a new decomposition stage of MGO from 200 °C to 450 °C can be observed, which corresponds to the decomposition of grafted HBPA. The morphological structure was also studied using TEM. As we can see from TEM images ( Fig. 2 ), GO sheets display better transparency than MGO, which are similar to the images reported in previous related studies. 52,63 In addition, elemental mapping of MGO can further demonstrate successful modification with HBPA as presented in its STEM images shown in Fig. 2(c and d) . It can be found that the nitrogen element, originated from HBPA, was evenly distributed on its surface. Additionally, AFM results ( Fig. 2(e and f) ) indicated that the thickness of MGO is ∼4.35 nm, which is obviously thicker than that of GO (∼0.8 nm), owing to the grafting of HBPA. Combined with the results from the above various characterizations, we confirm that the HBPA was successfully grafted onto the surface of GO. Fig. 2 TEM images of (a) GO and (b) MGO. (c and d) STEM image of MGO and the corresponding elemental mapping of nitrogen. AFM images of (e) GO and (f) MGO. Subsequently, MGO-reinforced self-healing polyurethane composites were fabricated via a four-step strategy ( Fig. 3 ). First, a terminal –NCO prepolymer was synthesized using HTPB and IPDI as the raw materials. Second, MGO was added to cross-linking with the prepolymer. Then, APD was used as a chain extender to react with the prepolymer, introducing the dynamic reversible disulfide bonds. Finally, DA was added to consume the remaining –NCO groups, yielding the mussel-mimetic adhesion structure, which can provide hydrogen bonds. Just by changing the contents of MGO, a series of self-healing polyurethane were fabricated. As far as we know, high self-healing efficiency is always at the expense of mechanical properties. 39–43 In our design, the self-healing properties originating from dynamic covalent bonds and mussel-mimetic structure, and the improved mechanical properties are well-combined in the HTPB-based PU. On the one hand, aromatic disulfide bonds can exhibit dynamic properties at room temperature, 29 which is conducive to the self-healing process at room temperature. On the other hand, the catechol groups on DA can form a weak dual hydrogen bond, which plays an initial sticking or interfacial adhesion role to promote the self-healing process, and the presence of the catechol groups is potential to achieve underwater self-healing. 37 In addition, well-dispersed MGO sheets greatly improved mechanical properties on the premise of maintaining high self-healing efficiency. Finally, the superior water-resistance of HTPB gives it the innate advantage of underwater self-healing. 10 Fig. 3 Synthesis of the self-healing polyurethane composites. The chemical structure of composites can be determined by FTIR. Fig. 4(a) exhibits FTIR spectra of DA, PU–S and SPU. The spectrum of DA shows characteristic peaks of O–H, C–N and N–H. 36 For PU–S, there are obvious characteristic peaks at 3331 cm −1 and 1709 cm −1 , corresponding to the stretching vibrations of N–H and C O in urethane linkage, and the peaks at 1520 cm −1 and 1350 cm −1 are attributed to urea N–H bending vibration (amide II) and aromatic C–N stretching vibration, respectively. 64 Additionally, there is an obvious absorption peak at 2266 cm −1 , attributed to the remaining –NCO groups, which disappear absolutely after adding DA in SPU. The presence of DA can be confirmed by utilizing UV-vis spectroscopy as elaborated in Fig. 4(b) . The SPU sample shows the absorption peak at ∼280 nm, which is the characteristic peak of DA. 33,65 Furthermore, the glass transition temperatures ( T g ) of the obtained composites were measured by DSC tests and the curves are displayed in Fig. 4(c) . It can be seen that the T g value increases with the increase of MGO content, which can be assigned to the crosslinking of MGO and polymer chains, limiting the mobility of molecular chains. Notably, the T g values of all samples are far below room temperature (∼−79 °C), implying the polymer chains hold high mobility at room temperature, which is beneficial to the self-healing process. The dispersion degree and interfacial interaction of the nanofiller have a great influence on the mechanical properties of composites. 4 XRD spectra display a sharp peak of GO at the 2 θ position of 9.05° ( Fig. 1(b) ), whereas SPU and SPUM-2 show similar XRD patterns, revealing that GO nanosheets are uniformly dispersed in the PU matrix ( Fig. 4(d) ). 66 Fig. 4 (a) FTIR spectra of DA, PU–S and SPU. (b) UV-vis spectra of SPU. (c) DSC curves of different composites. (d) XRD spectra of SPU and SPUM-2. The dispersion degree of MGO sheets in the matrix shows significant effect on the mechanical and self-healing properties of the composites, 67 as such SEM was used to observe the cryo-fractured surfaces of the obtained composites. As shown in Fig. 5(a) , obvious aggregates of GO can be observed, and interfaces of GO are quite bare ( Fig. 5(c) ), suggesting weak interfaces exist in SPUG matrix. In contrast, uniformly dispersed MGO can be seen in the fracture surface of SPUM-2 ( Fig. 5(b and d) ). Besides, owing to the crosslinking between MGO and matrix, MGO sheets are covered by the matrix and show indistinct interfaces. Fig. 5 SEM images of (a and c) SPUG-1 and (b and d) SPUM-2. Red arrows indicate the MGO sheets. 3.2. Self-healing properties Through the above molecular structure design, we successfully endow the composites with remarkable self-healing properties. To investigate the self-healing ability, all the samples are cut into two parts by a blade, and the cut surfaces were immediately brought together for a certain time at room temperature. As displayed in Fig. 6(a–e) , SPU samples show a high η value after healing for only 6 h. After the addition of MGO, the apparent decline occurred in the η value with increasing content of MGO. Especially, when the MGO content reaches 0.2 wt% (SPUM-4), the η value drastically decreases (62.5% for healing 6 h) even if the healing time was extended to 12 h. By comparison, SPUM-1 and SPUM-2 samples could still maintain a high η value of 87.9% and 87.1%, respectively ( Fig. 6(f) ). The outstanding self-healing capability of SPU samples could be ascribed to the synergistic effect of disulfide bonds and hydrogen bonds formed by catechol groups. Specifically, the disulfide exchange reactions can occur at a mild temperature due to its low bond dissociation energy and stable free radicals. 28,29 Second, the hydrogen bonds formed by catechol groups can accelerate initial sticking or interfacial adhesion, promoting the diffusion of molecular chains and accelerating the exchange reactions of disulfide bonds. 2,30 Third, the disulfide bonds will weaken the strength of hydrogen bonds, leading to their easier dissociation under mild conditions. 28 In turn, the exchange of hydrogen bonds can provide energy to promote disulfide metathesis. Moreover, APD with a zigzag structure could facilitate the molecular chain motion that readily induces hydrogen bond exchange and generates disulfide metathesis. 64 Nevertheless, the crosslinking between MGO and polymer chains will inhibit mobility which is essentially important for the self-healing ability, 68 resulting in a sharp decline in the η value when MGO content reaches up to 0.2 wt%. There may be a critical value for the content of MGO. Under this critical value, composites can still possess a high η value. From the results of self-healing experiments, 0.1% may be the critical value, therefore, we choose SPUM-2 for further study. Fig. 6 The tensile stress–strain curves of (a) SPU, (b) SPUM-1, (c) SPUM-2, (d) SPUM-3 and (e) SPUM-4 at different healing time under room temperature. (f) Self-healing efficiency ( η ) of different samples. To further intuitively study the outstanding self-healing performances, two samples, SPU and SPUM-2, were cut into two pieces and different samples were put together (SPU + SPUM-2) for healing for 6 h at 25 °C. Upon stretching, the healed sample exhibited no breaking ( Fig. 7(a) ). Subsequently, the tensile test was performed on the SPU + SPUM-2 sample to quantitatively describe the self-healing performance ( Fig. 7(b) , Movie S1 † ), and the stress–strain curves are displayed in Fig. 7(c) . It can be seen that even after healing with different samples, the stitched sample can still exhibit self-healing capability. Moreover, it is noteworthy that the two different parts show different elongations due to the difference in their moduli. Fig. 7 (a) The cut SPU (yellow) and SPUM-2 (brown) samples were put together for healing for 6 h at 25 °C (SPU + SPUM-2) and stretched again. (b) Images of stretching healed SPU + SPUM-2 sample after healing for 6 h at 25 °C. (the red dotted line represents the stitching position) (c) tensile stress–strain curves of the original SPU sample and SPU + SPUM-2 sample. Various devices based on PU will be inevitably affected by the complex environment during their service life. Therefore, the self-healing experiments were also conducted under different conditions to confirm the environmental adaptability of the self-healing properties. As shown in Fig. 8(a) , SPUM-2 samples were healed at different temperatures for 1 h, and the η value shows an upward trend with the increase of temperature. This phenomenon can be attributed to faster exchange reactions of disulfide bonds and hydrogen bonds at higher temperatures. Furthermore, the underwater self-healing property of the composites will significantly promote its applications in prospective coating, electrical skin and wearable devices. 69,70 To this end, the SPUM-2 sample was cut into two separate pieces by a blade and soaked in water to wet the fractured surfaces, and then put into a contact. After healing in 60 °C water for 1 h, the healed sample could be stretched to a high strain and without breaking (Movie S2 † ). Then, the tensile tests were conducted on the samples healed under 25 °C and 60 °C water, and the stress–strain curves were measured as displayed in Fig. 8(b) . We can see that the healed samples can be stretched up to 1870% strain with the η value of 92.5% under 60 °C water for 2 h. As for the sample healed under 25 °C water for 6 h, it still shows a high strain of 1750% and the η value of 74.6%, which just shows a slight decline compared with that in air atmosphere for 6 h. These results demonstrated that the obtained composites possess outstanding water-insensitivity and underwater self-healing property, which is meaningful for its applications. Fig. 8 (a) Tensile stress–strain curves of SPUM-2 sample after healing at different temperatures for 1 h. (b) Tensile stress–strain curves of OSPUM-2 sample after healing under 60 °C water for 2 h and 25 °C water for 6 h. (c) Schematic illustration of the self-healing mechanism of the composites. In order to intuitively elaborate the self-healing process, a schematic diagram is illustrated in Fig. 8(c) . Once the damaged surfaces are contacted together, the catechol groups can form hydrogen bonds to achieve initial sticking and interfacial adhesion. Then, the disulfide exchange reactions occurred at surfaces, which will promote the interdiffusion of polymer chains. 2 Finally, the continuous randomization of polymer chains gradually completes the self-healing process. 71 Moreover, a comparison of the η value, self-healing time and self-healing temperature between SPUM elastomers and other self-healing materials in recent studies is summarized in Fig. 9 . It can be found that some self-healing materials exhibit considerable η value but require relatively high temperature, and some can heal at medium temperature but a long time is necessary. On the contrary, SPUM samples not only exhibit a high η value but also show a rapid room-temperature self-healing property. Fig. 9 Comparison of the η value, self-healing time and self-healing temperature of the elastomers reported in this work and those reported in recent studies. 2,18,21,26,41,64,69,72–78 3.3. Mechanical properties Reports have indicated that the self-healing properties are always achieved by sacrificing mechanical properties. 39–43 The aim of our design was to fabricate PU with both rapid room-temperature self-healing properties and excellent mechanical properties by molecular structure design and controlling the content of MGO. It was expected that the mechanical properties of SPUM samples can be significantly improved by MGO. Consequently, the mechanical properties were carefully evaluated as follows. As shown in Fig. 10(a) , the original SPU samples show the elongation at break of nearly 2400% but low tensile strength and Young's modulus. After the incorporation of MGO, the tensile strength of SPUM samples increases with increase in MGO loading, while the elongation at break decreases except for SPUM-2 samples. To be specific, by increasing the content of MGO, the tensile strength of SPUM samples increases by 37.7%, 107.7%, 240.2% and 287.1%, relative to SPU. It is notable that the SPUM-2 samples exhibit no decline in elongation at break (2170%) when their tensile strength was improved, compared to SPUM-1 samples. It is different from the regular phenomenon reported in previous works, which are reporting that the addition of nanofiller into polymer will continuously decrease the elongation at break. 4,79 There is an optimal value of MGO loading, under which the tensile strength of the samples can be significantly improved while maintaining high elongation at break (SPUM-2). Instead, SPUG samples exhibit more apparent decline in the elongation at break and only slight improvement in tensile strength. Combining the SEM images, we infer that these phenomena can be attributed to the following factors: (1) The relative strong interfacial interactions promote stress transfer from matrix to MGO sheets. (2) Well-dispersion of MGO in matrix reduces the stress concentration from the agglomeration of MGO sheets. Subsequently, dynamic mechanical analysis was also conducted on these samples. As shown in Fig. 10(b) , the storage modulus ( G ′) shows an increasing trend with the increase in MGO loading over the entire frequency range, which is in good agreement with tensile tests. Fig. 10 (a) Stress–strain curves and (b) DMA results of various samples. As is known, PU has been widely used in numerous fields, and it always suffers from various exterior loading during applications. 1,2 Therefore, the deformation recovery capability of the composites is important for its long-term working. Given the outstanding self-healing properties and excellent tensile properties of SPUM-2 samples, we chose it for further research. As presented in Fig. 11(a and b) , cycle tensile tests were carried out on SPU and SPUM-2 samples. Both the stress–strain curves of SPU and SPUM-2 show significant and pronounced hysteresis in consecutive cycles without any relaxation time, implying a tremendous energy dissipation from bond breakage of disulfide bonds and hydrogen bonds during the tensile process. But the tensile stress gradually decreases with cycles because the polymer networks cannot be reformed in a short time. 80 Subsequently, they were allowed to relax for 10 min and stretched again. It is obvious that SPUM-2 samples show a similar cycle curve to that of the first cycle, while the SPU samples require more time even for 30 min, indicating the outstanding recovery capability of SPUM-2 samples. To further study the tensile recovery capability, 20 successive cycle tensile tests with 10 min of relaxation time between each cycle was carried out on SPUM-2 samples ( Fig. 11(c) ). To be more quantitative, we define both the recovery degree and dissipated energy by calculating the ratios of stress at the strain of 100% and the hysteresis loop area to those values at the first one. It is noticeable in Fig. 11(d) that, after 5 cycles, the recovery degree and dissipated energy of SPUM-2 recovered by 97.6% and 92.6%, respectively. Even after 20 cycles, they were still maintained at 93.5% and 76.3%, respectively. All the above results confirm that SPUM-2 possesses excellent recovery ability. Due to the outstanding mechanical properties of MGO and crosslinking between MGO and polymer chains, there have formatted a stronger network in polymer matrix. Fig. 11 Cyclic tensile curves of (a) SPU and (b) SPUM-2 with a strain of 100%. The sample was stretched and released 5 cycles continuously, then relaxed for another cycle. (c) Twenty successive cycle tests of SPUM-2 with 10 min of relaxing between each cycle. (d) Dissipated energy and recovery degree of SPUM-2 with 20 cycles. The compressive load is also a typical external load that PU will suffer from during the entire service life. Therefore, the compression cycle tests were also performed to reflect the deformation recovery capability. As shown in Fig. 12(a) and Movie S3, † SPUM-2 samples can recover their initial shape after manual compression in a few seconds. Then, different strains were applied on SPUM-2 ( Fig. 12(b and c) ). Its compressive stress increased with an increase in the strain and it can recover under such a high strain of 70%. Subsequently, the compression recovery of SPUM-2 was further investigated via a successive loading-unloading cycle test at the strain of 60% for 20 cycles with the recovery time of 1 min between each cycle ( Fig. 12(d and e) ). Obviously, the hysteresis loop area and compression stress exhibit a slight decline. Here, we also define both the recovery degree and dissipated energy by calculating the ratios of compression stress to the strain of 60% and the hysteresis loop area to those values at the first one, and the results are summarized in Fig. 12(f) . After 20 cycles, the value of recovery degree still remains at 83.7% and the dissipated energy shows a slight decline but still stayed above 71%. These results further demonstrated the excellent recovery ability of SPUM-2 samples under different external loads. The crosslinking between MGO and polymer chains reforms an elastic network where well stress transfer can effectively disperse the stress during loading–unloading procedures. 81,82 Fig. 12 (a) Images of compressed SPUM-2 samples. (b) Compression curves of SPUM-2 at different compression strains. (c) Corresponding changes in compression stress overtime during cycles. (d) Twenty successive compression cycles of SPUM-2 with 1 min recovery between each cycle. (e) Corresponding compression stress overtime during cycles. (f) Dissipated energy and recovery degree of SPUM-2 with 20 cycles. According to the study in the last section, composites still possessed excellent self-healing properties under the appropriate content of MGO. Hence, we could obtain a PU with both outstanding self-healing properties and superb mechanical properties from molecular structure design and controlling the content of MGO. From the above results, we believe that the SPUM-2 is an ideal PU. In summary, the superb self-healing and mechanical properties of our composites are attributed to the following factors: (1) the low T g (∼−79 °C) value of composites endows high mobility to the molecule chains, which enable self-healing process even at room temperature. (2) The aromatic disulfide exchange reactions can occur at mild temperature due to its low bond dissociation energy and stable free radicals. 28,29 (3) The mussel-inspired catechol groups can form weak dual hydrogen bonds to accelerate initial sticking or interfacial adhesion, which will promote the diffusion of molecular chains and accelerate the exchange reactions of disulfide bonds. 2,30 (4) The water-resistant adhesive characteristics of catechol groups promote initial sticking and interfacial adhesion under a water environment. 83 (5) Owing to the superior water-resistance of HTPB chains, 10,84 composites are not significantly affected by water molecules during the self-healing process. (6) The crosslinking between MGO and polymer chains can promote the stress transfer from the matrix to MGO, resulting in improved mechanical properties."
} | 7,939 |
35196089 | PMC8865770 | pmc | 6,865 | {
"abstract": "Hydrogen bond engineering is widely exploited to impart stretchability, toughness, and self-healing capability to hydrogels. However, the enhancement effect of conventional hydrogen bonds is severely limited by their weak interaction strength. In nature, some organisms tolerate extreme conditions due to the strong hydrogen bond interactions induced by trehalose. Here, we report a trehalose network–repairing strategy achieved by the covalent-like hydrogen bonding interactions to improve the hydrogels’ mechanical properties while simultaneously enabling them to tolerate extreme environmental conditions and retain synthetic simplicity, which proves to be useful for various kinds of hydrogels. The mechanical properties of trehalose-modified hydrogels including strength, stretchability, and fracture toughness are substantially enhanced under a wide range of temperatures. After dehydration, the modified hydrogels maintain their hyperelasticity and functions, while the unmodified hydrogels collapse. This strategy provides a versatile methodology for synthesizing extremotolerant, highly stretchable, and tough hydrogels, which expand their potential applications to various conditions.",
"introduction": "INTRODUCTION Hydrogels are biocompatible, mechanically tunable, conductive, and optically transparent, making them ideal materials for tissue engineering ( 1 , 2 ), microlenses ( 3 ), ionic conductors ( 4 ), ionotronic devices ( 5 ), electroluminescent devices ( 6 ), soft robotic actuators ( 7 – 9 ), etc. These applications rely on the stable and excellent mechanical properties of hydrogels under various conditions such as dehydration and low temperatures. However, network imperfections in hydrogels have a considerable effect on their mechanical properties, which often lead to mechanical properties degradation. For example, the mechanical properties of hydrogels are far below their corresponding theoretical values under the perfect network assumption, which can vary by several orders of magnitude ( 10 ). Moreover, the material properties of hydrogels can be further exacerbated due to water crystallization at low temperatures and water loss caused by air drying since they become damaged, brittle, rigid, and nonconductive, which will result in malfunction of hydrogel devices such as contact lens ( 11 ) and microcarries ( 12 ) under these environmental stresses. The significant gap between theory and experiment suggests a great potential to improve the mechanical properties of hydrogels by repairing network imperfections. Currently, intense efforts have been devoted to improving mechanical properties of hydrogels ( 13 – 17 ), while these strategies suffer from one or more limitations: complicated synthetic process, limited enhancement, lack of generality and biocompatibility, loss of functions after dehydration, and narrow working temperature range. It is very challenging to develop a general strategy to enhance the mechanical properties of hydrogels under various conditions while simultaneously retaining biocompatibility and synthetic simplicity. To this end, one needs to select an effective agent to fulfill the network-repairing function by introducing strong molecular interactions. Besides, a qualified agent should be biocompatible and ensure the sound operation of hydrogels under extreme conditions. In nature, numerous organisms, including bacteria, fungi, yeast, plants, insects, and tardigrades, can survive against adverse environmental conditions such as low temperatures and dehydration ( 18 – 21 ). Especially, water bears can withstand extreme temperatures from −273° to 100°C, almost complete dehydration ( 22 ), and high pressure ( 23 ). Evidence shows that they adapt by synthesizing a large amount of trehalose when exposed to these environmental stresses ( 24 ). Specifically, trehalose molecules can stabilize biomolecules by forming the strong interactions between the ─OH groups in trehalose and polar groups in biomolecules, and they can even replace hydration water molecules with stronger hydrogen bonding to biomolecules ( 19 , 25 , 26 ). Besides, trehalose protects against water crystallization at low temperatures since it deconstructs the tetrahedral hydrogen bond network of water and inhibits the emergence of ice-forming hydrogen bond configurations ( 24 , 26 – 29 ). It is well known that incorporating salts such as CaCl 2 into hydrogels can improve their anti-freezing capabilities; however, this strategy can degrade their mechanical properties. For example, the mechanical properties of CaCl 2 -modified hydrogels, such as strength and stretchability, decrease obviously with the addition of CaCl 2 at various temperatures ( 30 ). In addition, adding organic liquids such as propylene glycol into hydrogels can also improve their operating temperature range ( 31 ); however, the toxicity of organic liquids can result in health hazards and environmental pollution ( 32 ). In contrast, trehalose and its derivatives are environmentally benign and biocompatible. Inspired by its unique properties, trehalose can act as an effective network-repairing agent for hydrogels by introducing the covalent-like hydrogen bonding interactions, which is aimed to impart extra stretchability, high toughness, extremotolerant property, and self-healing capability to hydrogels.",
"discussion": "DISCUSSION We report a versatile strategy to enhance the mechanical properties of various hydrogels including PAAm, PVA, and PAAm-alginate DN hydrogels while simultaneously enabling them to tolerate adverse environmental conditions and to retain synthetic simplicity. Trehalose acts as the network-repairing agent by forming the covalent-like hydrogen bonds between trehalose and polymer chains. As a result, the strength, stretchability, and fracture toughness of the modified hydrogels significantly increase with increasing trehalose content. Since trehalose inhibits the formation of ice crystals by deconstructing the ice-forming hydrogen bond configurations, the modified hydrogels can retain their high stretchability, fracture toughness, and conductivity at low temperatures. Furthermore, trehalose protects the molecular structure of hydrogels after dehydration and consequently maintains their mechanical properties and functions. The protective roles are attributed to the support of trehalose clusters and the water-retaining mechanism. Last, we demonstrated several applications that require both good conductivity and excellent mechanical properties of hydrogels. This strategy provides a versatile methodology for greatly improving the mechanical properties of hydrogels under various conditions, which will expand the scope of hydrogel applications."
} | 1,673 |
23638358 | PMC3628875 | pmc | 6,866 | {
"abstract": "We investigated microbial communities active in methane oxidation in lake sediment at different oxygen tensions and their response to the addition of nitrate, via stable isotope probing combined with deep metagenomic sequencing. Communities from a total of four manipulated microcosms were analyzed, supplied with 13 C-methane in, respectively, ambient air, ambient air with the addition of nitrate, nitrogen atmosphere and nitrogen atmosphere with the addition of nitrate, and these were compared to the community from an unamended sediment sample. We found that the major group involved in methane oxidation in both aerobic and microaerobic conditions were members of the family Methylococcaceae , dominated by species of the genus Methylobacter , and these were stimulated by nitrate in aerobic but not microaerobic conditions. In aerobic conditions, we also noted a pronounced response to both methane and nitrate by members of the family Methylophilaceae that are non-methane-oxidizing methylotrophs, and predominantly by the members of the genus Methylotenera . The relevant abundances of the Methylococcaceae and the Methylophilaceae and their coordinated response to methane and nitrate suggest that these species may be engaged in cooperative behavior, the nature of which remains unknown.",
"conclusion": "Conclusions The well-characterized “aerobic” methanotrophs and most prominently the Methylobacter species are responsible for metabolism of methane in Lake Washington sediment in both aerobic and microaerobic conditions. In aerobic conditions, some type of a cooperative behavior between the Methylococcaceae , and most prominently the Methylobacter species is suggested by our data with the Methylophilaceae species, among which the Methylotenera species are most prominent. The nature of this type of cooperation remains unknown and requires a separate investigation. Both functional groups respond positively to the addition of nitrate. However, their ability to carry out classic respiratory denitrification is unlikely, as is a direct metabolic linkage between methane oxidation and denitrification. It is more likely that nitrate stimulates the methylotroph communities as a nutrient.",
"introduction": "Introduction Methane is recognized as one of the most powerful greenhouse gases, with annual emissions of approximately 600 Tg ( King, 1992 ; Hanson & Hanson, 1996 ; Etiope & Klusman, 2002 ; Keppler et al., 2006 ). Its atmospheric concentration has been steadily increasing over the past 300 years, mostly due to anthropogenic activities ( Singh et al., 2010 ). Until recently, two major modes have been recognized by which methane is removed from the environment: aerobic oxidation conducted by a specialized group of bacteria, known as methanotrophs ( Hanson & Hanson, 1996 ; Chistoserdova & Lidstrom, 2013 ), and anaerobic oxidation linked to sulfate reduction, conducted by a specialized group of archaea, known as anaerobic methanotrophs or ANME ( Valentine, 2002 ; Knittel & Boetius, 2009 ). The former process is important for methane consumption in freshwater sediments and soils, whereas the latter is thought to be the major process in anoxic marine environments. More recently, however, evidence has been accumulating that other metabolic modes for methane consumption must exist, linked to alternative electron donors, such as nitrate/nitrite-dependent anaerobic/microaerobic bacterial methane oxidation in freshwater environments ( Wu et al., 2011 ) and metal-dependent methane oxidation by archaea in marine environments ( Beal, House & Orphan, 2009 ). These new findings point toward novel biogeochemical processes that need elucidation in order to be placed into the context of the global carbon cycle. However, the relative environmental significance of these processes, the identity of the microbes involved, and the details of their metabolism remain poorly characterized. On the other hand, the clear separation between the aerobic and the anaerobic modes of metabolism may represent an artifact originating from experimental data predating environmental microbiology approaches, including culture-independent approaches. This notion is nicely illustrated by anaerobic nitrite-dependent methane oxidation in members of the NC10 phylum occurring by aerobic methane oxidation. This metabolic mode involves canonical methane monooxygenase, the classic sets of methanol and formaldehyde oxidation enzymes, and a strict reliance on the presence of oxygen that, in this case, is produced intracellularly ( Wu et al., 2011 ). At the same time data are available suggesting that at least some of the classic aerobic methanotroph and methylotroph species may be able to thrive in microaerobic environments and potentially utilize alternative electron acceptors, such as nitrate, for methylotrophic metabolism ( Costa et al., 2000 ; Modin, Fukushi & Yamamoto, 2007 ; Kalyuzhnaya et al., 2009 ; Stein & Klotz, 2011 ). We have previously characterized communities involved in methylotrophy in Lake Washington, Seattle, USA, using both culture-reliant and culture-independent approaches, focusing on organisms active in aerobic conditions. These studies identified a diverse functional community and highlighted the potential importance of the Methylococcaceae and the Methylophilaceae species as members of this community ( Nercessian et al., 2005 ; Kalyuzhnaya et al., 2008 ; Chistoserdova, 2011a ). Metagenome-based metabolic reconstruction of these species has indicated that at least some of them are capable of denitrification, suggesting that they may be adaptable to an anaerobic/microaerobic life style ( Kalyuzhnaya et al., 2008 ; Kalyuzhnaya et al., 2009 ). In this study, we have expanded the previous efforts of characterizing functional methylotroph communities by addressing the nature of the communities involved in methane metabolism in both aerobic and microaerobic conditions. In addition, we have addressed the potential role of nitrate in these communities in an attempt to further link carbon and nitrogen cycles in terrestrial environments.",
"discussion": "Discussion The metagenomic approaches, including “functional metagenomics” allow glimpses into the content of natural microbial communities, including uncultivated species, along with understanding their most prominent activities in global elemental cycles ( Chistoserdova, 2010 ; Morales & Holben, 2011 ). We have previously employed a “high-resolution” metagenomics approach to communities inhabiting freshwater sediment using stable isotope probing (SIP), in order to specifically target populations involved in utilization of single carbon compounds with a few notable outcomes ( Kalyuzhnaya et al., 2008 ). In this previous work we uncovered a dominant presence of Methylobacter species as part of the bacterial community actively consuming methane in this environment, in contrast to the results from cultivated methanotroph species ( Auman et al., 2000 ). We also discovered a prominent presence of novel Methylophilaceae species that were classed into a separate, novel genus, Methylotenera ( Kalyuzhnaya et al., 2006 ). These species appeared to be active in consuming a variety of C1 substrates, most notably methylamine, methanol and methane ( Kalyuzhnaya et al., 2008 ). As we were able to cultivate Methylotenera species at the same time ( Kalyuzhnaya et al., 2006 ; Kalyuzhnaya et al., 2012 ), another contradiction arose: as expected for members of the Methylophilaceae ( Anthony, 1982 ), these species contained no genes that would encode methane oxidation functions. No genes for a typical (MxaFI) methanol dehydrogenase were present in these organisms ( Lapidus et al., 2011 ). How then could they successfully compete for carbon from either methane or methanol with other species that possess the traditional enzymes for such types of metabolism? One other notable discovery was the persistent presence of genes for the denitrification pathway in Methylotenera species, suggesting a potential connection between methylotrophy and denitrification and a potential for electron acceptor alternatives to oxygen ( Kalyuzhnaya et al., 2008 ). However, in the laboratory the cultivated Methylotenera species revealed very low potential for methanol metabolism ( Kalyuzhnaya et al., 2006 ; Kalyuzhnaya et al., 2012 ). However, further experiments with in situ populations using labeled methanol, varying tensions of oxygen and varying presence of nitrate have confirmed that the Methylophilaceae , and most prominently the Methylotenera species, must be the major methanol utilizers in Lake Washington ( Kalyuzhnaya et al., 2009 ). XoxF, a homolog of the traditional methanol dehydrogenase (large subunit) was proposed as a gene involved in methanol oxidation ( Beck et al., 2011 ), supported by high expression of these genes in in situ conditions ( Kalyuzhnaya et al., 2010 ). In this work, we pursued three major objectives: determining what, if any, guilds beyond Methylococcaceae and Methylocystaceae were involved in methane oxidation in freshwater lakes, whether Methylophilaceae were involved in this process, and if the presence of nitrate had an effect on methane-oxidizing communities. We demonstrate that the known methanotroph guilds, Methylococcaceae and Methylocystaceae appear to be the major responders to the methane stimulus in aerobic microcosm incubations. More specifically, the Methylococcaceae and species belonging to or related to the genus Methylobacter are both the dominant species in the natural environment as well as the dominant responders to methane and to nitrate in aerobic conditions. While sequences of the recently described methanotroph guild NC10 that carries out methane oxidation anaerobically and links it to nitrite or nitrate ( Wu et al., 2011 ) were detected in all samples, these were minor members of the community, and no response to methane or nitrate was observed. Even though the abundance of Methylococcaceae sequences in microaerobic microcosms was much lower compared to both aerobic microcosms and to the unamended sample, they significantly outnumbered the NC10 sequences. No other phylum revealed a pattern suggesting involvement in methane oxidation, and no novel methane monooxygenase genes were detected, suggesting that in both aerobic and microaerobic conditions methane was metabolized by the methanotrophs traditionally called “aerobic methane oxidizers” ( Chistoserdova & Lidstrom, 2013 ). Methanotrophs of the family Methylococcaceae revealed a pronounced positive response to the addition of nitrate in aerobic conditions. However, these organisms do not appear to encode a complete respiratory denitrification pathway and likely use nitrate and nitrite reductases for assimilating nitrogen. Most if not all of these organisms also encode nitrogenases. The Methylocystaceae that constitute a smaller population in Lake Washington sediment also positively responded to methane but not to nitrate, in aerobic conditions, but they were almost undetectable in microaerobic conditions. The only non-methanotroph guild that responded to methane and nitrate stimuli was the Methylophilaceae , of which Methylotenera species were the most prominent in the datasets analyzed. Moreover, the response pattern of the Methylophilaceae correlated well with the pattern of the Methylococcaceae in aerobic conditions, suggesting a potential cooperation between the two groups at ambient oxygen tension. On the contrary, at low oxygen tension, and especially in the presence of nitrate, high community diversity, including the diversity of the denitrification genes, was observed, suggesting cross-feeding from labeled metabolites originating from the methanotrophs, even though the latter were present at a low population level. The nature of the cooperation between the Methylococcaceae and the Methylophilaceae is not obvious. It could be suggested that the methane oxidizers release methanol as a result of high activity of methane monooxygenase, and that the Methylophilaceae consume this methanol, quickly incorporating it into their biomass. However, the dominant population of the Methylophilaceae enriched in the methane-fed microcosms appears to be most closely related to Methylotenera versatilis , cultivated representatives of which grow poorly if at all on methanol and lack bona fide (MxaFI) methanol dehydrogenase ( Kalyuzhnaya et al., 2012 ). On another hand, multiple guilds that are known to be robust methanol oxidizers, such as Methylobacteriaceae, Hyphomicrobiaceae, Xanthobacteriaceae , as well as methanol-oxidizing Methylophilaceae ( Methylovorus, Methylophilus ) are minor members of the enriched communities. The Methylophilaceae could be involved in detoxification of nitrogen species to some of which, most notably ammonia, Methylococcaceae are known to be sensitive ( Nyerges & Stein, 2009 ). However, in this case it is difficult to explain why Methylophilaceae are more successful than other species active in nitrogen metabolism. The same argument would be appropriate if a non-specific cross-feeding (for example on metabolites resulting from lysis of the Methylococcaceae ) is suggested. The analyses presented here suggest that methanotrophs known as “aerobic” methanotrophs appear to be responsible for metabolizing methane in both aerobic and microaerobic conditions, even though they appear not to be as efficient at low oxygen tension as they are at high oxygen tension. The Methylophilaceae appear to be involved in methane oxidation in the aerobic conditions but not in microaerobic conditions, suggesting that the methanotrophs, dependent on the specific environmental circumstances, may engage in different types of partnerships, involving either a very specialized guild such as methylotrophs of the family Methylophilavceae or a diverse group of heterotrophs with versatile metabolic repertoires."
} | 3,492 |
34050157 | PMC8163789 | pmc | 6,868 | {
"abstract": "In living organisms, proteins are organized prevalently through a self-association mechanism to form dimers and oligomers, which often confer new functions at the intermolecular interfaces. Despite the progress on DNA-assembled artificial systems, endeavors have been largely paid to achieve monomeric nanostructures that mimic motor proteins for a single type of motion. Here, we demonstrate a DNA-assembled building block with rotary and walking modules, which can introduce new motion through dimerization and oligomerization. The building block is a chiral system, comprising two interacting gold nanorods to perform rotation and walking, respectively. Through dimerization, two building blocks can form a dimer to yield coordinated sliding. Further oligomerization leads to higher-order structures, containing alternating rotation and sliding dimer interfaces to impose structural twisting. Our hierarchical assembly scheme offers a design blueprint to construct DNA-assembled advanced architectures with high degrees of freedom to tailor the optical responses and regulate multi-motion on the nanoscale.",
"introduction": "Introduction Cellular life functions through a collection of highly-controlled dynamic processes involving the self-assembly and organization of diverse molecular building blocks. These biological constructs optimized through billions of years of evolution provide us with the inspiration to create artificial systems, which emulate the structural and functional features of their natural counterparts. In particular, oligomerization is essential in many protein-involved cellular processes 1 , 2 . For instance, a GTPase called dynamin is at the heart of endocytic vesicle fission. The dynamin unit is an antiparallel dimer, which can oligomerize into a helical polymer. The ratchet model proposes that GTP hydrolysis powers the relative sliding of the helical turns, giving rise to twisting of the helix and eventually membrane fission 3 , 4 . Such a self-association mechanism through dimerization and oligomerization not only enables new functions at the intermolecular interfaces but also elicits a wealth of structural and functional advantages. This offers a blueprint to create mimics that can collectively operate to define functions in artificial machinery. Among different state-of-the-art techniques, DNA nanotechnology 5 – 17 represents a unique tool to build bio-inspired artificial systems, taking advantage of the precision, addressability, and programmability of DNA on the nanoscale. A variety of DNA-based dynamic devices 18 – 21 that mimic motor proteins 22 in living cells has been accomplished, including walkers 23 – 30 , rotors 31 – 36 , sliders 37 – 40 , and assembly lines 41 . Recently, efforts to constructing bio-inspired DNA-assembled systems are taking a step forward from monomeric to high-order structures 42 – 53 and from simple to complex motion 38 , 49 , 54 – 57 . Here, we demonstrate a DNA-assembled building block with rotary and walking modules, which can form dimeric structures to execute sliding, as well as subsequent higher-order architectures linked through dimer interfaces of the rotary and walking modules for controlled multi-motion. The monomeric building block is a chiral system 58 , 59 , which consists of two interacting gold nanorods (AuNRs) templated by DNA origami 60 – 63 . The two AuNRs can carry out the rotation and walking, respectively. Through dimerization, the walking modules are merged into a sliding module, imposing relative movements between two building blocks. Further oligomerization leads to higher-order structures with alternating dimer interfaces for collective rotation and sliding, imposing structural twisting. Our work delineates one of the rudimentary steps towards self-assembled artificial structures that imitate the collective motion of protein complexes, with the ultimate goal to build and engineer cellular mimics with fully operational structural and functional features de novo. Although it is an incredible adventure, fortunately, we will be able to look at the solutions provided by nature for every problem we encounter.",
"discussion": "Discussion In this work, we have demonstrated dimerization and oligomerization of DNA-assembled building blocks for controlled multi-motion in high-order architectures. The monomeric UNIT contains walking and rotary modules. Activation of the walking dimer interface enables the formation of a DIMER from two UNITs and meanwhile introduces the sliding module. Activation of the dimer interfaces of both the walking and rotary modules leads to the formation of OLIGOMERs, which can undergo collective rotation and sliding processes. The collective motion also enforces relative twisting within the structures and leads to varied lateral dimensions. The hierarchical assembly through activation of different dynamic dimer interfaces provides a pathway to achieve complex and controlled multi-motion in high-order structures."
} | 1,236 |
38573429 | PMC11021290 | pmc | 6,870 | {
"abstract": "Plants can produce complex pharmaceutical and technical proteins. Spider silk proteins are one example of the latter and can be used, for example, as compounds for high-performance textiles or wound dressings. If genetically fused to elastin-like polypeptides (ELPs), the silk proteins can be reversibly precipitated from clarified plant extracts at moderate temperatures of ~ 30 °C together with salt concentrations > 1.5 M, which simplifies purification and thus reduces costs. However, the technologies developed around this mechanism rely on a repeated cycling between soluble and aggregated state to remove plant host cell impurities, which increase process time and buffer consumption. Additionally, ELPs are difficult to detect using conventional staining methods, which hinders the analysis of unit operation performance and process development. Here, we have first developed a surface plasmon resonance (SPR) spectroscopy-based assay to quantity ELP fusion proteins. Then we tested different filters to prepare clarified plant extract with > 50% recovery of spider silk ELP fusion proteins. Finally, we established a membrane-based purification method that does not require cycling between soluble and aggregated ELP state but operates similar to an ultrafiltration/diafiltration device. Using a data-driven design of experiments (DoE) approach to characterize the system of reversible ELP precipitation we found that membranes with pore sizes up to 1.2 µm and concentrations of 2–3 M sodium chloride facilitate step a recovery close to 100% and purities of > 90%. The system can thus be useful for the purification of ELP-tagged proteins produced in plants and other hosts. Supplementary Information The online version contains supplementary material available at 10.1007/s11248-024-00375-z.",
"conclusion": "Conclusions Here we have developed an SPR-based quantitation method for ELP fusion proteins, which are difficult to detect using conventional staining methods. This will accelerate future process development because the performance of individual unit operations can be rapidly assessed, e.g. in terms of purity and recovery. We have also established a fast, membrane-based purification method for ELP fusion proteins that can simplify manufacturing, e.g. for future technical applications of spider silk proteins. Specifically, a repeated cycling of ELP fusion proteins between aggregated and dissolved state can be avoided during purification as impurities are flushed out much like in a regular UF/DF operation. Because the method uses readily available membranes of 0.2 µm pore diameter, an implementation into production processes seems straight forward. Once 1.2 µm membranes become commercially available for UF/DF devices, the throughput of the system may be increased, e.g. due to a reduction in membrane fouling and concentration polarization (Kim 2007 ). Quantifying such unwanted side effects along with typical loadings (i.e. grams of aggregated ELP fusion protein per square meter of membrane area) should be the focus of subsequent scale-up experiments, for example using tangential flow filtration devices. In this context, the impact of pH during wash and aggregation of ELP fusion proteins can be assessed too, but may have little effect due the uncharged amino acids constituting the ELP tag. Additionally, implementing the removal of the ELP (and tag) fusion parts in the process, for example through (self-)cleavage (Lingg et al. 2022 ; Li 2011 ), will be necessary in the future.",
"introduction": "Introduction Plants can have several advantages for the production of biopharmaceutical proteins such as low upstream production costs of ~ 50 € kg −1 wet biomass due to simple cultivation (Ridgley et al. 2023 ) and an inherent safety because they do not support the replication of human pathogens (Schillberg et al. 2019 ; Donini and Marusic 2019 ; Moustafa et al. 2016 ; Tschofen et al. 2016 ; Buyel 2023 ). However, the recovery of recombinant proteins from plant extracts can be challenging due to large quantities of particles and host cell proteins that are typically released during biomass homogenization (Wilken and Nikolov 2012 ; Buyel 2015 ). Various methods have been develop in the last years to address this issue, including flocculation or blanching to simplify clarification and purification respectively (Buyel et al. 2014 ; Buyel and Fischer 2014a ). Apart from process modifications, purification challenges can be circumvented by genetically engineering a product. For example, the product can be extended with short stretches of amino acids, so called tags, that simplify product capture (Costa et al. 2014 ; Pina et al. 2014 ; Young et al. 2012 ). One of the most commonly used tags facilitating affinity chromatographic purification is a stretch of about six histidine residues typically located at the N-terminus or C-terminus of a product. This tag enables selective binding of a product to immobilized divalent metal ions (Debeljak et al. 2006 ). However, the histidine tag may be inefficient for purification for example if chelating agents are present (Gengenbach et al. 2018 ). Also, the tag may be cleaved of in dependence of the plant cultivation conditions such as growth temperatures > 40 °C (Knödler et al. 2019 ) or the corresponding chromatography resin can be too costly for the bulk production of technical proteins. An example of the latter are spider silk proteins that can be used as high-performance textile fibers (Belbéoch et al. 2021 ). Alternatively, elastin-like polypeptide (ELP) based purification tags offer a straightforward and simple chromatography-free separation of product and host cell proteins (HCPs). Also, ELPs can be effectively produced in plants and plant cells (Floss et al. 2010 ) and the tag can be cleaved-off after purification (Lan et al. 2011 ) so that authentic product can be recovered. ELPs comprise 30‒100 repeats of a VPGXG motif, where X is any amino acid except proline (Urry 1988 ). This sequence mediates a reversible aggregation of ELPs as well as ELP-fusion proteins (Christensen et al. 2013 ) at a so called transition temperature ( T t ), which is ≥ 30 °C in the presence of ≥ 1.0 M salt. The T t can be reduced by increasing the salt concentration as well as by increasing number of VPGXG repeats in the ELP tag (Conley et al. 2009 ) as well as by increasing the hydrophobicity of the guest residue X in the ELP motif (Urry et al. 1992 ; Miao et al. 2003 ). Accordingly, the method is compatible with heat-labile fusion proteins (Bischof and He 2005 ). The aggregates formed by ELP fusion proteins are in the micrometer range (Miao et al. 2003 ) so that > 95% of soluble HCPs can be removed with the supernatant after centrifugation (Meyer and Chilkoti 1999 ) or in the flow-through of a membrane filtration step (Phan and Conrad 2011 ). The method is termed ‘inverse transition cycling’ (ITC) and it has been used successfully to purify spider silk proteins (Weichert et al. 2014 ), hemagglutinin (Phan et al. 2014 ) and lectins (Tian and Sun 2011 ). However, both centrifuge (cITC) and membrane-based (mITC) methods are currently operated in a discontinuous mode, requiring several aggregation-disaggregation cycles to achieve high product purity and have limited scalability. We therefore set out to investigate if ITC can be adapted to ultrafiltration/diafiltration and the corresponding good manufacturing practice-compliant equipment, which allows a simple scale-up and a continuous operation with only a single aggregation step because residual HCPs can be separated from the product in diafiltration operation mode (Fig. 1 ). We termed this approach ‘membrane-based inverse transition purification’ (mITP) to discriminate it from the previous methods that require cyclic processing of feeds. Five spider silk-ELP fusion proteins (Fig. 2 A) were used to develop the method and to demonstrate its transferability. Fig. 1 Schematic representation of the membrane-based inverse transition purification (mITP) process. Starting with a clarified plant homogenate ( A ), the temperature is increased in the presence of salt to trigger the precipitation of ELP-fusion proteins ( B , here: fused to spider-silk proteins). The suspension is then applied to a membrane of suitable pore size (e.g. 0.2–2.0 µm), for example in an ultrafiltration/diafiltration device, so that the precipitate is retained whereas the bulk homogenate passes into the flowthrough ( C ). Next, the membrane is flushed with a hot, salt-rich buffer to remove residual impurities whereas the ELP-fusion proteins remain in a precipitated state ( D ). Lastly, a cold buffer (without salt) or plain water is used to re-dissolve the ELP precipitate and to elute the product from the membrane ( E ) Fig. 2 Spider silk elastin-like polypeptides (ELP) fusion proteins and their quantification of with a surface plasmon resonance (SPR) spectroscopy competitive binding assay. A Schematic representation of the five fusion proteins used in this study. The c-myc part of the fusion protein is shown in blue, whereas the spider-silk domain is colored in green and the ELP part is orange. B Competition assay principle. Anti-c-myc antibody (red) pre-incubated with c-myc-tagged ELP fusion protein (domain color code as in A) containing sample or standard is brought in contact with a surface decorated with peptides containing a c-myc epitope or variant thereof. Only antibodies with at least one unoccupied valency can bind to the surface resulting in a response signal. C Response resulting from antibody (green—9E10, n = 1; orange—A00704, n = 3) binding to a surface decorated with peptide 3 in dependence of the concentration of ELP standard the antibody was pre-incubated with. Data were fitted to a site competition model (Eq. 1 ) to derive inflection points",
"discussion": "Results and discussion ELP fusion proteins can be quantified by an SPR competition assay We first set out to establish a reliable quantitation assay for ELP fusion proteins based on a myc-tag present in all constructs (Fig. 2 A). This was important because common detection methods such as densitometric analysis of Coomassie-stained polyacrylamide gels have proven insensitive to ELP-containing proteins, probably because the latter contain < 2% basic amino acids, which are necessary for binding the Coomassie dye (Hassouneh et al. 2010 ). Initially, we immobilized mouse-anti-c-myc antibody 9E10 (Krauss et al. 2008 ; Hilpert et al. 2001 ) on an amine sensor chip with dextran coating (the latter can increase chip capacity) for a direct quantification assay (Fig. S1 A) achieving up to 1500 RU (Fig. S1 B) after coupling, which was equivalent to ~ 1.5 mg m −2 of mAb and good compared to previous reports (Schasfoort and Schasfoort 2017 ; Murphy et al. 2017 ; Opdensteinen et al. 2023 ). Using repeated injections of the same clarified plant extract sample containing VSO1ELP in a 1:20 dilution in SPR running buffer, we found that the RU signal declined in the course of 150 runs from 790 to 82 RU (Fig. S1 C). Because the baseline signal was stable, we excluded leaching of mAb from the chip surface as a reason for this reduction. Instead, we assumed that the required regeneration conditions (30 mM hydrogen chloride, Table S3 ) were too harsh for mAb 9E10 causing its denaturation (Lazar et al. 2010 ) and thus signal reduction. We therefore deemed this direct assay inadequate for ELP quantitation. Next, we tested an indirect SPR assay where a defined amount of anti-c-myc antibody was added to ELP fusion protein samples and the resulting antibody binding to a chip surface covered with peptides containing the c-myc epitope was measured (Fig. 2 B). Because of their small size, these peptides did not have a distinct three-dimensional structure so denaturation was not an issue during the regeneration of the sensor chip surface. However, the small size may also limit the number and steric accessibility of functional groups for coupling to the chip surface. We therefore tested three peptide variants (Table 1 ) that we developed based on previous recommendations (Krauss et al. 2008 ). The highest coupling response and signal stability over repeated sample injections was observed for peptide 3 (Fig. S1 D), which then we used for all subsequent quantifications. We confirmed that the competition assay provided quantitative results by establishing a high-quality standard curve (adj. R 2 > 0.99; Eq. 1 ; Fig. 2 C) using defined mixtures of an 100xELP standard with antibody 9E10 as well as a commercially available anti-c-myc antibody A00704 as reference material. The inflection points of the models were ~ 5 × 10 –8 M (9E10) and ~ 3 × 10 –8 M (A00704) corresponding to ~ 250 and ~ 350 µg L −1 of 100xELP respectively, which marked the most reliable quantification region of the assay. Given TSP levels of ~ 12 g kg −1 in tobacco biomass (Opdensteinen et al. 2018 ) and fusion protein expression levels of up to 0.02–1.00% TSP in previous work (Phan et al. 2013 ), a 1:4 dilution during extraction and a 1:40 dilution during sample preparation, we expected product concentrations of 15–750 µg L −1 during measurement and concluded that this matched well with the quantitation range of the assay. A modified clarification process is necessary to avoid product losses We then individually expressed the five fusion proteins VSO1ELP, MaSp1ELP, MaSp2ELP, FlagELP and 100xELP in transgenic tobacco plants and observed product levels of ~ 0.02 (MaSp2ELP) to 0.60 (100xELP) g kg −1 biomass (VSO1ELP ~ 0.40 g kg −1 ; MaSp1ELP ~ 0.45 g kg −1 ; FlagELP ~ 0.30 g kg −1 ), which was slightly higher compared to previous reports where ELP fusion proteins accumulated up to 0.01–0.12 g kg −1 (Phan and Conrad 2011 ; Phan et al. 2013 ). We extracted the products with a buffer containing salt in a concentration required for the subsequent mITP (1.50 to 3.00 M sodium chloride) to streamline the process. A sequence of bag and depth filtration was used for clarification as described before (Buyel and Fischer 2014b ). Interestingly, the recovery of ELP fusion proteins was low under these conditions, for example < 5% for VSO1ELP (data not shown). We first speculated that even at ~ 22 °C the high salt concentrations may have caused some degree of ELP fusion protein aggregation as reported before (Christensen et al. 2013 ). This could result in a retention of the product on the filters, especially as the latter had nominal pore diameters of < 1.0 µm, which is about the size of ELP aggregates (Hassouneh et al. 2010 , 2012 ; Dreher et al. 2008 ). Therefore, we shifted the salt addition to after depth filtration rather than before extraction for all subsequent experiments. This restored the recovery of VSO1ELP to 56 ± 9% (n = 9; Fig. 3 A), whereas the recovery remained < 5% for all other ELP fusion proteins unless depth filtration was replaced by centrifugation (Fig. S2 ), as was the case for MaSp1ELP (Fig. 3 B). It seemed implausible to us that these proteins would form aggregates due to the base salinity of the extraction buffer (~ 20 mS cm −1 ) at 22 °C. Instead, unspecific binding of proteins to depth filters containing diatomaceous earth has been reported before (Knödler et al. 2023 ; Opdensteinen et al. 2021 ), and we then assumed that electrostatic interactions between the proteins and the charged components of the depth filter caused the product losses as previously observed for other molecules (Menzel et al. 2018 ). We therefore tested a set of depth filters that contained less diatomaceous earth (Fig. 3 C–E, Table S1 ), because this component can absorb proteins (Yigzaw et al. 2006 ; Buyel et al. 2015 ). Filter PDR1 performed best, combining a high filter capacity (> 60 L m −2 ), low turbidity (< 25 NTU) and high product recovery (> 0.75) (Fig. 3 F). This filter was then used for all subsequent experiments also because it was easier to scale up compared to centrifugation which achieved a similar product recovery. Fig. 3 Screening of depth filters and ELP fusion protein recovery during clarification. A Western blot of process samples using depth filter P1 and anti-c-myc for detection of VSO1ELP. Elution fractions originated from 1&3—0.2 µm membrane pore size, 2.0 M sodium chloride during aggregation, 30 °C during wash; 2&4—1.2 µm membrane pore size, 2.4 M sodium chloride during aggregation, 45 °C during wash; primary elution was carried out using 15 mM sodium phosphate buffer pH 7.5 whereas de-ionized water (indicated by “w”) was used for a second elution step. B Western blot of process samples using centrifugation and anti-c-myc for detection of MaSp1ELP. Elution fraction conditions as in A. C ELP fusion protein recovery achieved with different depth filters (Table S1 ). D Filter capacity in dependence of filter layer combinations and ELP fusion protein. E Turbidity observed after clarifying ELP fusion protein containing extract with different depth filters. Error bars in C – E indicate the standard deviation (n ≥ 3). F Western blot of process samples using depth filter PDR1 and anti-c-myc for detection of ELP fusion proteins. ELP elastin-like polypeptide, ITP inverse transition purification. Error bars indicate the standard deviation of replicate runs with n ≥ 2 Different membranes and aggregation conditions can be used for fusion protein purification We used VSO1ELP for an initial statistical screening experiment to identify suitable conditions for mITP, achieving high purity and recovery. Extract clarified by bag filtration and depth filtration with PDR1 was used to test different mITP conditions (Table S2 ). The resulting model for product recovery was of fair quality given the complex sample preparation and small scale of the experiments (adj. R 2 0.56; Table S4 ). The model indicated that the salt concentration during wash was the dominating factor for VSO1ELP recovery and that aggregation salt concentration and membrane pore sizes had a smaller effect. The aggregation temperature did not have a significant contribution in the investigated range (30–45 °C). High VSO1ELP recoveries were identified for two different factor combinations, both performing aggregation and wash at 2.0 M and 3.0 M respectively, but using either a 1.20 µm or 0.20 µm membrane (Table S2 , Fig. 4 A and B). Fig. 4 Screening for mITP conditions ensuring a high recovery and purity of ELP fusion proteins using VSO1ELP as a model protein. A Response surface model for VSO1ELP recovery in dependence of sodium chloride concentration during aggregation and wash using a 0.2-µm membrane for aggregate retention. B Same model as in A but using a 1.2-µm membrane. C Response surface model for VSO1ELP purity increase as a multiple of the starting purity in dependence of sodium chloride concentration during aggregation and wash using a 0.2-µm membrane for aggregate retention. D Same model as in C but using a 1.2-µm membrane. The aggregation temperature did not have a significant effect in the 30–45 °C range and was set to 37.5 °C in panels A – D . Dots indicate actual measurements. E Overlay of Coomassie-stained LDS-PAA gel and corresponding western blot using anti-c-myc antibody for VSO1ELP detection in process samples for the verification of the process optimum using a 0.2-µm membrane. F Same setup as in E but with samples from runs confirming the optimum for a 1.2-µm membrane We also analyzed VSO1ELP purity in mITP elution fractions and found that hardly any protein was detected on Coomassie-stained LDS-PAA gels (Fig. 4 C and D) but that silver-staining revealed a dominant band of the size expected for VSO1ELP (~ 70 kDa) as well as a smaller band at ~ 55 kDa. Because the latter band was also detected by western blotting and had a similar size as the 100xELP protein which did not contain a spider silk fusion part, we assume that it corresponded to a product-related degradation (Fig. 3 E and F). Combining the detection limit of Coomassie brilliant blue-based protein staining of ~ 0.01 g L −1 (Opdensteinen et al. 2018 ) with the VSO1ELP quantification by SPR, we estimated the minimal VSO1ELP purity to be > 90%. This was in good agreement with previous reports using mITC that achieved purities of up to 97% (Phan et al. 2013 ). Interestingly, no significant model was obtained for the product purity. We concluded that the mean purity was the best estimator and that model factors did not have a relevant influence on VSO1ELP purity in the investigated ranges. The two optimal conditions in terms of VSO1ELP recovery were verified in independent runs and yielded recoveries of 114 ± 28% (1.2-µm membrane) and 86 ± 19% (0.2-µm membrane) respectively (n = 3). The high recoveries were in good agreement with previous reports for mITC where up to 90% were reported (Phan and Conrad 2011 ; Floss et al. 2009 ). Effective mITP conditions can be identified for individual ELP fusion proteins We then conducted an additional design of experiments to adapt the two optimal conditions to other fusion proteins MaSp1ELP, MaSp2ELP, FlagELP and 100xELP. Because 1.2-µm membranes were not available from the manufacturer of the ultrafiltration/diafiltration (UF/DF) device for a subsequent scale-up, we limited the investigation to the aggregation temperature and salt concentration but kept the wash salt concentration at 3.0 M, the optimal level for all membranes. The recovery of MaSp1ELP was only affected by the aggregation salt concentration (adj. R 2 = 0.89) with an optimum at ~ 2.7 M sodium chloride (> 90% recovery), whereas there were no significant effects on the recovery of FlagELP or 100xELP which were ~ 95% and 70% respectively. A model for MaSp2 was not established because the expression level and concentration after depth filtration of the recombinant protein were too low to allow a quantification."
} | 5,484 |
26943628 | PMC4989316 | pmc | 6,871 | {
"abstract": "Groundwater ecosystems are conventionally thought to be fueled by surface-derived allochthonous organic matter and dominated by heterotrophic microbes living under often-oligotrophic conditions. However, in a 2-month study of nitrate amendment to a perennially suboxic aquifer in Rifle (CO), strain-resolved metatranscriptomic analysis revealed pervasive and diverse chemolithoautotrophic bacterial activity relevant to C, S, N and Fe cycling. Before nitrate injection, anaerobic ammonia-oxidizing (anammox) bacteria accounted for 16% of overall microbial community gene expression, whereas during the nitrate injection, two other groups of chemolithoautotrophic bacteria collectively accounted for 80% of the metatranscriptome: (1) members of the Fe(II)-oxidizing Gallionellaceae family and (2) strains of the S-oxidizing species, Sulfurimonas denitrificans . Notably, the proportion of the metatranscriptome accounted for by these three groups was considerably greater than the proportion of the metagenome coverage that they represented. Transcriptional analysis revealed some unexpected metabolic couplings, in particular, putative nitrate-dependent Fe(II) and S oxidation among nominally microaerophilic Gallionellaceae strains, including expression of periplasmic (NapAB) and membrane-bound (NarGHI) nitrate reductases. The three most active groups of chemolithoautotrophic bacteria in this study had overlapping metabolisms that allowed them to occupy different yet related metabolic niches throughout the study. Overall, these results highlight the important role that chemolithoautotrophy can have in aquifer biogeochemical cycling, a finding that has broad implications for understanding terrestrial carbon cycling and is supported by recent studies of geochemically diverse aquifers.",
"introduction": "Introduction Microbial communities in nature are seldom exposed to a static environment, and their responses to changes in their physicochemical environment can help us to better understand and predict what organisms and processes will drive biogeochemical cycling under a range of conditions (e.g, Allison and Martiny, 2008 ; Hamilton et al. , 2014 ; Faust et al. , 2015 ). In this study, we characterized the response of an aquifer microbial community to an increase in the flux of nitrate, a naturally occurring electron acceptor, at a field site in Rifle, Colorado (USA). This nitrate perturbation study was motivated in part by observations of dynamic fluctuations in nitrate concentrations within the capillary fringe of this aquifer (ranging from ~50 μ m to more than 6000 μ m over time periods of 2 months and vertical distances of 0.5 m; unpublished data). This study allowed us to observe, in a controlled manner, how the aquifer microbial community could respond to nitrate being transported into the saturated zone from the vadose zone. The response to an influx of a thermodynamically favorable electron acceptor like nitrate could be expected to enhance microbial oxidation of organic electron donors (e.g., via heterotrophic denitrification; Korom, 1992 ; Starr and Gillham, 1993 ; Bengtsson and Bergwall, 1995 ), consistent with current heterotroph-dominant paradigms of groundwater ecology ( Baker et al. , 2000 ; Foulquier et al. , 2010 ), as well as inorganic electron donors (e.g., via chemolithoautotrophic denitrification; Pauwels et al. , 1998 ; Beller et al. , 2004 ; Herrmann et al. , 2015 ). A wide range of factors could modulate the response to a nitrate perturbation, but a predictive and mechanistic understanding of these factors requires highly resolved empirical studies, ideally at genome-scale resolution such that strain-level behavior could be correctly interpreted ( Allen et al. , 2007 ; Sharon et al. , 2013 ; Luo et al. , 2015 ). Here, we used strain-resolved metagenomic reconstructions combined with metatranscriptomic analysis to derive transcriptional evidence of metabolic activity and highlight important biogeochemical processes. Although an increasing number of metatranscriptomic analyses of environmental systems have been reported over the past decade ( Poretsky et al. , 2005 ; Frias-Lopez et al. , 2008 ; Gilbert et al. , 2008 ; Urich et al. , 2008 ; Poretsky et al. , 2009 ; Shrestha et al. , 2009 ; Lesniewski et al. , 2012 ; Quaiser et al. , 2014 ; Kopf et al. , 2015 ), most have focused on marine, freshwater and surface soil systems, and rarely on aquifer environments, where studies have typically targeted selected, diagnostic transcripts rather than the entire metatranscriptome (e.g., Chandler et al. , 2006 ; O'Neil et al. , 2008 ; Alfreider and Vogt, 2012 ; Herrmann et al. , 2015 ).",
"discussion": "Discussion In this study, we characterized the response of an aquifer microbial community to an increase in the flux of nitrate, a naturally occurring electron acceptor at the Rifle field site. This nitrate perturbation study was motivated, in part, by observations of dynamic fluctuations in nitrate concentrations in the Rifle aquifer. We took advantage of synoptically collected metagenomic, metatranscriptomic and geochemical data during the 2-month study to characterize community response, with a focus on transcriptional evidence of metabolic activity for bacteria represented by strain-resolved metagenomic reconstructions. Although influx of a thermodynamically favorable electron acceptor like nitrate could be expected to enhance microbial oxidation of organic electron donors as well as inorganic ones, the metatranscriptomic data unequivocally indicated a strong chemolithoautotrophic response, with coupled reduction of nitrate and other nitrogen oxides, oxidation of Fe(II) and reduced S compounds, and CO 2 fixation. Thus, pervasive and diverse chemolithoautotrophic activities mediated C, S, N and Fe cycling in the aquifer both before and during the nitrate release. Notably, the very high level of chemolithoautotrophic activity indicated by the metatranscriptomic data would generally not have been inferred based on metagenomic data alone ( Figure 2 ). The three most active groups of chemolithoautotrophic bacteria in this study, Gallionellaceae, S. denitrificans and anammox bacteria, have overlapping metabolisms that allowed them to occupy different yet related metabolic niches throughout the study. Although members of all three groups respired nitrate, they had markedly different temporal expression profiles and activity maxima, even among closely related species/strains within the three groups ( Figures 2 , 3 , 4 and 6 ). Temporal and spatial variations in geochemistry, such as nitrate influx and composition of minerals containing Fe(II) and reduced S, and in community structure, likely established the foundations for temporal and spatial niche variation. Anammox bacterial activity was relatively high only on day 0 ( Figure 3 ), whereas Gallionellaceae metabolic activity was highest on days 21 and 47 (bin 22.5) or day 35 (bin 22.6) or day 47 (bin 22.9) ( Figure 4 ), and S. denitrificans activity was greatest on day 21 (bin 93) or day 35 (bin 62) ( Figure 6 ). An example of apparent niche differentiation between and within the major chemolithoautotrophic groups can be illustrated with S-compound oxidation. Different expression patterns were observed for S-compound oxidation pathways, such as sox and reverse dsr , among two S. denitrificans strains ( Figure 6 ) and several of the putative Sideroxydans spp. within the Gallionellaceae group ( Figure 4 ). The observation of S-compound oxidation in Gallionellaceae strains was unusual, as these bacteria are typically characterized by Fe(II) oxidation. Although growth of Sideroxydans ES-1 with a reduced S compound (thiosulfate) has been observed in the laboratory ( Emerson et al. , 2013 ), this is one of the few reports of S oxidation by Sideroxydans spp. in the field ( Purcell et al. , 2014 ). It appears that both nitrate-dependent S oxidation and nitrate-dependent Fe(II) oxidation were metabolic strategies used in situ by members of the Gallionellaceae group, including both putative Gallionella and Sideroxydans species. Neither of these nitrate-dependent activities has been documented previously for isolated members of the Gallionellaceae, although nitrate-dependent Fe(II) oxidation has been inferred for a Sideroxydans member of an enrichment culture ( Blöthe and Roden, 2009 ). The observation that the genomes of various Gallionellaceae strains encoded two different nitrate reductases, the periplasmic NapAB and the membrane-bound NarGHI, is particularly noteworthy because the reported genomes of Gallionellaceae isolates do not contain any genes encoding dissimilatory nitrate reductases. These annotations of nitrate reductases entail a high level of confidence; for example, alignment of the translated napA from bin 22.5 vs structurally characterized versions of NapA reveals conservation of all six active-site residues ( Moura et al. , 2004 ) and a high degree of sequence identity ( Supplementary Figure S3 ). The scaffold containing napA had ribosomal proteins with high sequence identity (89–95%) to the corresponding proteins in Gallionella capsiferriformans . A putative Gallionella genome from a different Rifle aquifer study also included napAB genes (JF Banfield and A Shelton, personal communication), although the gene organization was different than that observed in bin 22.5 ( Figure 5 ). In metagenomic data from the CO 2 -rich Crystal Geyser aquifer (Green River, UT, USA) ( Emerson et al. , 2015 ), nitrate reductase ( narG , narH ) and sox genes were putatively placed in a Gallionellales bin. The genome of the isolate Sideroxydans ES-1 encodes a nitrite reductase, but nitrite-dependent Fe(II) or S oxidation have not been reported in this strain. Overall, the presence and expression of nitrate reductases in Gallionellaceae strains in this study compels us to expand our view of the metabolic diversity of these chemolithoautotrophic bacteria. Anammox metabolic activity appears to be a favorable niche at the study site under background conditions, as anammox gene expression was greatest before the nitrate release and accounted for a substantial portion of the metatranscriptome on day 0. Although a number of aquifer studies have focused on 16S rRNA and functional gene surveys of anammox bacteria, and of their activity using isotopic methods ( López-Archilla et al. , 2007 ; Clark et al. , 2008 ; Smits et al. , 2009 ; Hirsch et al. , 2011 ), this is the first study to document the activity of an anammox population in an aquifer using metatranscriptomic data. The presence of a population with high relative activity under background conditions suggests that the relevance of anammox in such systems may be underrepresented, particularly in studies that focus solely on metagenomic analysis or marker genes. The precipitous decrease in anammox activity after the initiation of the nitrate release is confounding, particularly because nitrite is the electron acceptor for this process. We explored two possibilities that may have contributed to the marked decrease in anammox gene expression after the addition of nitrate: the production of sulfide by sulfate-reducing bacteria, and the introduction of small amounts of oxygen during the nitrate injection. The active sulfate-reducing bacterial population only constituted ca. 1% of the metatranscriptome at day 0, and decreased to ca. 0.3% by day 47, and diagnostic sulfate reduction pathway genes were not highly expressed in any sample (<35 RPKM). Thus, while low levels of sulfide irreversibly inhibit anammox bacteria ( Jin et al. , 2013 ), the amount of sulfide produced under field conditions was probably too low to have markedly inhibited the anammox bacterial population. A more likely cause for the decreased anammox activity is the introduction of low levels of oxygen during the nitrate injection. Anammox bacteria are obligate anaerobes that are reversibly inhibited by molecular oxygen; a wide range of inhibitory O 2 concentrations has been reported ( Strous et al. , 1997 ; Carvajal-Arroyo et al. , 2013 ). Although we did not collect DO data for this study, we did observe transcriptional evidence for low concentrations of DO. Relatively high expression of genes encoding cbb 3 -type cytochrome c oxidase, a high-oxygen-affinity cytochrome oxidase consistent with oxygen reduction in a suboxic environment, was observed both for Gallionellaceae strains ( Figure 4 ) and S. denitrificans strains ( Figure 6 ). This constitutes evidence of DO within the aquifer test zone at concentrations sufficient to elicit a biological response from two different groups of bacteria. Note that DO concentrations may have been sufficient to inhibit anammox bacteria but may not have supported respiration by other prominent chemolithoautotrophs. S. denitrificans DSM 1251 has been characterized as an obligate denitrifier that is sensitive to oxygen, and its function for cbb 3 -type cytochrome c oxidase could be to prevent oxygen poisoning, not to respire O 2 ( Sievert et al. , 2008 ). Further, background DO concentrations are typically very low (<20 μ m ) at the study site (D Pan, KH Williams, MJ Robbins, and KA Webber, personal communication). The preceding discussion raises an important point about the role of transcriptional data in this study: it informed interpretation of biogeochemical processes beyond what could have been inferred based on metagenomic and geochemical data. Transcriptional data clearly highlighted highly active organisms ( Figure 2 ) and metabolic pathways ( Figures 3 , 4 and 6 ) that would not have been highlighted by metagenomic data alone. Further, transcriptional data revealed considerable information on S-compound oxidation in situ , even though no data were collected on any solid-phase S-containing compounds in the aquifer, which were likely the electron donors for S oxidation and potentially also included some of the oxidized products. Such solid-phase compounds would have posed significant analytical challenges if we had attempted to measure them routinely. The only S-containing compound that was routinely measured in this study was sulfate, and it indeed increased markedly in concentration after the nitrate release started ( Supplementary Figure S2 ), consistent with S-compound oxidation, but the transcriptional data is a powerful complementary indicator of this biogeochemical activity. Overall, these results demonstrating pervasive and diverse chemolithoautotrophy at the Rifle site have implications beyond this site and these study conditions. Although groundwater ecosystems were conventionally thought to be fueled by surface-derived allochthonous organic matter and dominated by heterotrophic microbes living under often-oligotrophic conditions, recent studies at a range of field sites have indicated the importance of chemolithoautotrophy in subsurface carbon cycling ( Alfreider et al. , 2003 , 2009 ; Alfreider and Vogt, 2012 ; Herrmann et al. , 2015 ). For example, in a study of two adjacent limestone aquifers, quantitative PCR surveys of form I and II RubisCO genes revealed that up to 17% of the microbial population had the capability to fix CO 2 by the Calvin cycle, and that S-oxidizing Sulfuricella strains and ammonia-oxidizing Nitrosomonas strains accounted for substantial portions of the cbbM and cbbL transcript pools ( Herrmann et al. , 2015 ). Further, PCR surveys of RubisCO genes and transcripts in groundwater samples from a range of pristine and polluted, shallow and deep aquifers indicated that autotrophs were widespread inhabitants of these groundwater systems ( Alfreider et al. , 2003 , 2009 ; Alfreider and Vogt, 2012 ). Finally, in a study of the Rifle aquifer under background (unperturbed) conditions, Handley et al. (2014) documented that an S-oxidizing chemolithoautotrophic bacterium, Candidatus Sulfuricurvum sp. RIFRC-1, composed ~47% of the sediment bacterial community based on metagenomic data. The present work adds to the collective evidence of the importance of chemolithoautotrophy in aquifers while providing a high degree of resolution of the metabolic pathways being expressed and elucidating metabolic lifestyles that expand our appreciation of the metabolic diversity of known chemolithoautotrophic taxa. Ongoing and future studies include attempts to isolate novel Gallionellaceae strains capable of nitrate-dependent Fe(II) and S oxidation to better understand their underlying biochemistry and efforts to quantify the role of chemolithoautotrophy in the subsurface carbon budget via quantitative environmental proteomics."
} | 4,183 |
39380495 | PMC11714156 | pmc | 6,873 | {
"abstract": "Abstract Adhesives that excel in wet or underwater environments are critical for applications ranging from healthcare and underwater robotics to infrastructure repair. However, achieving strong attachment and controlled release on difficult substrates, such as those that are curved, rough, or located in diverse fluid environments, remains a major challenge. Here, an octopus‐inspired adhesive with strong attachment and rapid release in challenging underwater environments is presented. Inspired by the octopus's infundibulum structure, a compliant, curved stalk, and an active deformable membrane for multi‐surface adhesion are utilized. The stalk's curved shape enhances conformal contact on large‐scale curvatures and increases contact stress for adaptability to small‐scale roughness. These synergistic mechanisms improve contact across multiple length scales, resulting in switching ratios of over 1000 within ≈30 ms with consistent attachment strength of over 60 kPa on diverse surfaces and conditions. These adhesives are demonstrated through the robust attachment and precise manipulation of rough underwater objects.",
"introduction": "1 Introduction Robust adhesives that strongly bond while also controllably releasing in wet or underwater environments are required in diverse fields including bio‐medical applications, [ \n \n 1 \n , \n 2 \n \n ] wearable electronics, [ \n \n 3 \n , \n 4 \n , \n 5 \n \n ] and soft robotics. [ \n \n 6 \n , \n 7 \n , \n 8 \n \n ] However, strong underwater attachment is a significant challenge due to the existence of a layer of water at the interface. [ \n \n 9 \n , \n 10 \n , \n 11 \n , \n 12 \n , \n 13 \n \n ] Attachment becomes even more complicated in “real‐world” environments which often display irregular surfaces and challenging conditions. [ \n \n 14 \n , \n 15 \n \n ] For example, rough or curved surfaces reduce the contact area between the attachment device and substrate which can limit adhesion strength. [ \n \n 16 \n , \n 17 \n , \n 18 \n , \n 19 \n , \n 20 \n \n ] The surface energy of the materials in contact is another factor that affects interaction forces, where lower surface energy materials form weaker attraction forces. [ \n \n 21 \n , \n 22 \n , \n 23 \n \n ] Furthermore, the types of fluid at the interface affect the interaction between adhesive and surface. [ \n \n 24 \n \n ] For example, seawater, which includes various ions, can result in lower adhesion compared to deionized water due to Debye screening effects. [ \n \n 25 \n \n ] Overall, these diverse factors can reduce attachment strength and ultimately decrease the effectiveness of underwater attachment and controlled release in relevant conditions. To achieve robust attachment in wet environments, several different methods have been adopted. For example, hydrogels, which have the ability to absorb water at the interface, and contain functional groups such as amino, carboxyl, or hydroxyl groups can create strong adhesion. [ \n \n 25 \n , \n 26 \n , \n 27 \n \n ] Additionally, chemical modification of adhesives can promote strong bonding to the substrate through chemical reactions in a wet environment. [ \n \n 28 \n , \n 29 \n , \n 30 \n \n ] Although hydrogels and chemical modification each provide robust adhesion performance in wet environments, these approaches are often focused on permanent adhesives and are not readily released. To create a strong yet reversible attachment, switchable adhesives can strongly attach while switching to a low force for release through a prescribed stimulus. [ \n \n 31 \n , \n 32 \n , \n 33 \n , \n 34 \n \n ] In wet or submerged environments, switchable attachment can be achieved through several different mechanisms, which includes capillary forces, hydrostatics (i.e., suction), hydrodynamics, and/or surface adhesion. [ \n \n 34 \n , \n 35 \n , \n 36 \n \n ] To achieve these attachment characteristics in wet or submerged environments, organisms such as the octopus provide inspiration with their outstanding ability to manipulate underwater objects. [ \n \n 37 \n , \n 38 \n , \n 39 \n \n ] For underwater environments, octopus‐inspired adhesives have been shown to attach and detach from wet or submerged objects. [ \n \n 35 \n , \n 40 \n , \n 41 \n , \n 42 \n , \n 43 \n , \n 44 \n \n ] This attachment is often attributed to the ability to generate interfacial pressure, [ \n \n 45 \n , \n 46 \n , \n 47 \n \n ] which is the difference between external pressure and pressure within the attachment structure. However, strong attachment and controlled release on non‐ideal substrates is difficult, [ \n \n 48 \n \n ] due to an inability to form, maintain, and rapidly control interfacial pressure in synthetic attachment structures. Therefore, synthetic materials with strong attachment and rapid, controlled release across diverse surfaces remains a major challenge. [ \n \n 49 \n \n ] \n Here, we present an octopus‐inspired adhesive that strongly attaches and rapidly releases in diverse underwater environments and conditions. Inspired by the infundibulum structure of the octopus, we create a compliant, curved stalk coupled with an active, deformable membrane that changes shape for multi‐surface adhesion ( Figure \n 1 a ). The stalk curvature enhances attachment across diverse surfaces by increasing conformal contact on large‐scale curvature and by increasing contact stress around the stalk perimeter for adaptability to small‐scale roughness. These synergistic mechanisms increase contact across several length scales which increases interfacial pressure for robust underwater attachment and controllable release (Figure 1b ). Our octopus‐inspired switchable adhesive (OSA) strongly attaches yet rapidly releases (≈30 ms, Figure S2 , Supporting Information) complex objects from lightweight shells (2.5 g) to large rocks (872 g) that have coupled roughness and curvature across multiple scales (Figure 1c ; Video S1 , Supporting Information). This switching is achieved by deflecting the membrane with pneumatic‐pressure to achieve attachment switching ratios up to 1000× from the activated to deactivated state (Figure 1d ). The attachment strength is consistently high (≈ 60 kPa) across various conditions, including substrate material, substrate curvature and roughness, testing fluid type, and testing fluid viscosity (Figure 1e ). This approach provides a mechanism for strong yet rapid release to diverse underwater objects and surfaces even in challenging and difficult environments. Figure 1 Octopus‐inspired switchable adhesives. a) Schematic of an octopus‐inspired switchable adhesive cross‐section and actuation of the membrane in the activated and deactivated state. b) Effect of curvature on the ability to seal interfacial pressure on an irregular surface. c) Attach‐and‐release octopus schematic and underwater manipulation demonstration of the octopus‐inspired switchable adhesive on irregular surfaces (scale bar = 15 mm). d) Stress versus time for an activated and deactivated OSA during a pull‐off test underwater. e) Underwater attachment strength on various substrates and environmental conditions. Bar represents the mean value ± s.d. All data are acquired with R * = 25 mm.",
"discussion": "3 Discussion This work demonstrates an octopus‐inspired adhesive with strong attachment and rapid release in challenging underwater environments. By tuning stalk curvature at the contacting surface coupled with an active membrane, we show how contact stress can be utilized to systematically control interfacial pressure. This is especially important when the substrate has roughness, curvature, or a combination of both, where the curved stalk shape enhances conformal contact on large‐scale curvatures and small‐scale roughness. This allows our OSAs to rapidly activate interfacial pressure, hold it for strong attachment, and then release it rapidly to controllable release objects on diverse surfaces and conditions. This ability is enabling for underwater manipulation tasks, where we show through the ability to achieve strong attachment and precisely manipulate irregular objects underwater. This work points to the importance of contact geometry for successful underwater attachment and release. This can guide future work on synthetic underwater adhesives, where additional contact geometries such as stalk architecture and membrane shape or tuning actuation schemes may further control attachment. It also provides insight into the contract geometry of natural octopus suckers, which typically show curvature, where future studies could quantify attachment architecture to build a better understanding of the diversity of contacting shapes. Overall, these results and designs can advance underwater and wet attachment which can be useful for diverse applications including robotic manipulation and healthcare."
} | 2,181 |
34630019 | PMC8496503 | pmc | 6,874 | {
"abstract": "The Hodgkin-Huxley (HH) spiking neuron model reproduces the dynamic characteristics of the neuron by mimicking the action potential, ionic channels, and spiking behaviors. The memristor is a nonlinear device with variable resistance. In this paper, the memristor is introduced to the HH spiking model, and the memristive Hodgkin-Huxley spiking neuron model (MHH) is presented. We experimentally compare the HH spiking model and the MHH spiking model by applying different stimuli. First, the individual current pulse is injected into the HH and MHH spiking models. The comparison between action potentials, current densities, and conductances is carried out. Second, the reverse single pulse stimulus and a series of pulse stimuli are applied to the two models. The effects of current density and action time on the production of the action potential are analyzed. Finally, the sinusoidal current stimulus acts on the two models. The various spiking behaviors are realized by adjusting the frequency of the sinusoidal stimulus. We experimentally demonstrate that the MHH spiking model generates more action potential than the HH spiking model and takes a short time to change the memductance. The reverse stimulus cannot activate the action potential in both models. The MHH spiking model performs smoother waveforms and a faster speed to return to the resting potential. The larger the external stimulus, the faster action potential generated, and the more noticeable change in conductances. Meanwhile, the MHH spiking model shows the various spiking patterns of neurons.",
"conclusion": "6. Conclusion The biological neuron is expressed adequately by the classic HH spiking model. It is sensitive to the temperature, the strength of the external stimulus, and the action time of the stimulus. The MHH spiking model successfully simulates the generation of the action potential in a neuron. When the different external stimuli are applied to the HH and MHH spiking models, the action potential is produced, and various spiking patterns are achieved. The MHH spiking model has advantages in generating the action potential through the comparison with the HH spiking model. The waveforms with smaller perturbations formed by the MHH spiking model are smooth. The higher frequency of the external stimulus, the more action potentials generated. The response speed of the MHH spiking model is faster than that of the HH spiking model. The various spiking behaviors are obtained by adjusting the signal frequency in the MHH spiking model. And meanwhile, the combination between neuron models and a memristor provides the possibility to scale down the neuron circuit and gives a novel way to replicate the functions of the biological neuron.",
"introduction": "1. Introduction Neurons with highly nonlinear characteristics act as the basic functional unit of receiving and propagating signals. The whole procedure of processing signals in the nerve system needs the cooperation of neurons. Some theoretical knowledge and research methods are beneficial to unveil the mechanism of information propagation in neurons. Italian scientist Camillo Golgi worked on the nervous system structure and earned the Nobel Prize for physiology and medicine in 1906 (Dröscher, 1998 ). In 1998, Ramon y Cajal pointed out that the neurons without directly connecting each other in the nerve system (Raviola and Mazzarello, 2011 ). To replicate the functions and mechanisms of neurons, we urgently need to construct the biophysical model. A variety of neuron models are emerging, and the Hodgkin-Huxley (HH) spiking neuron model is the original (Hodgkin and Huxley, 1989 ). Stochastic Hodgkin-Huxley Neuron Systems with the NEF is helpful to study neuron sensitivity (Chen and Li, 2010 ). The Hodgkin-Huxley Model with automatic parameter estimation is applied to the neuromimetic chips (Buhry et al., 2011 ). The space-clamped Hodgkin-Huxley model effectively inhibits the production of spikes under the injection of the noisy synaptic input (Tuckwell and Ditlevsen, 2016 ). The Langevin is combined with the Hodgkin-Huxley system performs accurate interspike interval (ISI) and realizes the accuracy minimal loss (Pu and Thomas, 2020 ). The Berger-Levy theory is introduced to the Hodgkin-Huxley model, demonstrate that the information communication between neurons is related to the presynaptic firing rate and the synchronization (Ghavami et al., 2018 ). The memristor with the non-volatility and variable resistance characteristics is regarded as the fourth passive circuit element. Therefore, it becomes a hot topic in neural computing (Le et al., 2015 ), learning and memorizing (Sayyaparaju et al., 2018 ), micro-circuitry design (Berdan et al., 2014 ), biological synapse (Mandal and Saha, 2016 ), and neuron modeling (Maheshwar et al., 2014 ), and so on. The synaptic plasticity of biological neuronal systems can be realized by memristors and memristive crossbar in 3-D architecture to mimic the human brain (Truong et al., 2016 ). The memristor with hysteresis and memory characteristics is the most promising candidate for establishing the brain-like neuromorphic system (Mokhtar et al., 2017 ). The key features of biological neurons and synapses can be mimicked by memristors (Berdan et al., 2016 ; Mandal and Saha, 2016 ). The ion motion in neurons is represented by the electrical conductance change of a memristor (Xia and Yang, 2019 ). A memristor is used as a two-terminal resistor with memory (Chua, 1971 ; Strukov et al., 2008 ) performs well in storing information according to the physical laws (Yang et al., 2013 ). The memristor entirely avoids the data transformation bottleneck between the memory and computation (Li and Wang, 2019 ). The memristor crossbar array can be used to integrate the co-processor chip, which will realize machine learning algorithms and neuromorphic computing (James, 2019 ). This work elaborates on the construction of the memristive Hodgkin-Huxley spiking neuron model. The mathematical expressions and the circuit of the HH spiking model are presented and analyzed in sections 2, 3. Section 4 describes the MHH spiking model and discusses the memristors used to mimic the ion channels. The comparison between two models under the different stimuli is conducted in section 5. Section 6 is the conclusion of the paper."
} | 1,583 |
35684621 | PMC9182789 | pmc | 6,875 | {
"abstract": "Printed electronic devices have demonstrated their applicability in complex electronic circuits. There is recent progress in the realization of neuromorphic computing systems (NCSs) to implement basic synaptic functions using solution-processed materials. However, a fully printed neuron is yet to be realised. We demonstrate a fully printed artificial neuromorphic circuit on flexible polyimide (PI) substrate. Characteristic features of individual components of the printed system were guided by the software training of the NCS. The printing process employs graphene ink for passive structures and In 2 O 3 as active material to print a two-input artificial neuron on PI. To ensure a small area footprint, the thickness of graphene film is tuned to target a resistance and to obtain conductors or resistors. The sheet resistance of the graphene film annealed at 300 °C can be adjusted between 200 Ω and 500 k Ω depending on the number of printed layers. The fully printed devices withstand a minimum of 2% tensile strain for at least 200 cycles of applied stress without any crack formation. The area usage of the printed two-input neuron is 16.25 mm 2 , with a power consumption of 37.7 mW, a propagation delay of 1 s, and a voltage supply of 2 V, which renders the device a promising candidate for future applications in smart wearable sensors.",
"conclusion": "5. Conclusions We successfully fabricated a two-input fully printed neuron on flexible polyimide using drop-on-demand additive-printing techniques. To the best of our knowledge, this is the first time that such a demonstrator is reported that integrates digital printing techniques with flexible substrates in this field of application. The software-pretrained neuron is based on graphene ink for the conductive tracks, resistors, and EGT source, gate, and drain electrodes. The resistance of the printed graphene film was modulated by changing its thickness. The heating process, simultaneously annealing both graphene ink and In 2 O 3 , was optimized, such that there was no crack formation in In 2 O 3 . The transistor-based diode used electrolyte gating; thereby, the two-input neuron operated at only 2 V, rendering it a suitable candidate for self-powered portable printed-computing systems. The printed neuron had propagation delay of 1 s, power consumption of 37.7 mW, and area utilization of 16.25 mm 2 , paving the road for upscaling fully printed two-input neurons. The presented NCS concept can be further improved by extending currently used ANN building blocks with circuits for negative weight operation. Another extension would be to use a nonlinear activation function. Our future work will implement these changes to fabricate an extended neuronal concept while using the fully printed approach on flexible substrate.",
"introduction": "1. Introduction 1.1. Printed Electronics in Neuromorphic Sensing Printed electronics (PE) are an emerging fabrication technology that5 enable new innovative products and application fields to drive the ever-increasing intelligent sensor market through smart front-end devices. It can substantially decrease manufacturing costs and the time to market, and allows for hardware with a flexible form factor. New application domains that require low-cost flexible electronics are targeted, since conventional silicon-based technologies possess form-factor limitations [ 1 , 2 , 3 ]. Typical applications are soft sensors [ 4 , 5 ], soft robotics [ 6 , 7 , 8 ], flexible wearable medical devices [ 9 ], fast-moving smart consumer goods [ 2 , 10 , 11 ], or Internet of Things (IoT) infrastructures [ 12 ]. Similar to conventional silicon technology, sensor preconditioning and processing can also be performed for printed technology [ 13 ], preferably with the deployment of power- and area-efficient neuromorphic computing systems (NCSs). As an example, NCS can be used to complement highly sensitive soft robotic pressure sensors [ 14 ] for optimizing the grabbing and releasing of objects. Printed electronics rely on digital additive processes such as inkjet printing, and could be printed directly on flexible materials without requiring a mask or subtractive processes [ 15 , 16 , 17 , 18 ]. Due to the difficulty in printing dielectric materials, we chose an electrolyte gating approach that was inkjet-printer-friendly. Therefore, electrolyte-gated transistor technology (EGT) [ 19 ] is a promising candidate. It can be fully processed with digital drop-on-demand properties systems without the need for additional postprocessing steps, and thus fulfils the required for flexible, point-of-use, and low-cost electronics. Furthermore, EGTs can be driven by very low supply voltages, as low as 2 V, allowing for using printed batteries or energy harvesters for the power supply for portable applications. Owing to these offered benefits by EGTs, there are significant efforts in utilizing them for neuromorphic sensing, synaptic coupling, and artificial sensory systems [ 20 ]. Conventional electric circuit designs cannot be directly mapped to our EGT technology owing to missing PMOS, larger feature size and lower integration density. Even though there is progress in organic electronics [ 21 ], a PMOS is still missing in printed inorganic electronics and is not yet optimized for CMOS. Hence digital systems in the technology lead to substantial area overhead, high power consumption and low performance. Consequently, the requirements of the targeted applications cannot be met [ 22 ]. In that respect, biology-inspired neuromorphic computing [ 23 ] coupled with analog computing [ 24 , 25 ] can be leveraged to obtain an efficient circuit implementation for PE applications. An NCS is inspired by the topology of the brain and can be built from artificial synapses and neurons. A significant advantage of the NCS is that sensory data can be processed without having to convert into digital signals, which otherwise requires additional analog–digital converters which increase costs and the hardware footprint. The intrinsic fault tolerance of NCS [ 26 , 27 ] can be leveraged to tolerate manufacturing imperfections and defects, which is a benefit for printed systems. The challenge, however, is to fully print the neuromorphic circuits. 1.2. Device Architecture of Fully Printed Two-Input Neuron In general, very common realizations of NCS are based on feed-forward or artificial neural networks (ANNs) [ 28 ], which are adapted in training routines such as back-propagation [ 29 ]. Printed NCSs consist of a network of artificial neurons comprising a network of resistors that are both intra- and interlayer interconnected to perform regression or classification tasks [ 29 ]. ANNs are implemented into the hardware owing to their simplicity and ease of training, which can be achieved by a least-mean-squares model. Most recently, artificial neural networks fabricated by additive technologies were presented, such as basic synaptic functions on flexible substrates [ 30 ] and a multiply–accumulate (MAC) operation on flexible substrates was also presented [ 31 ]. NCSs based on EGTs [ 32 , 33 , 34 ], and screen-printed [ 35 ] and aerosol-jet-printed NCS [ 36 ] were reported. Some works proposed organic functional materials [ 24 ] and suitable training algorithms for printed NCSs [ 33 , 34 , 37 , 38 ]. Although these contributions strongly indicated the relevance of printed NCS for prospective computing systems, most of the works investigating EGT-based NCSs still use rigid substrates such as silicon and glass. Moreover, passive conductive tracks are often not printed but thermally evaporated or sputtered, and hence require additional subtractive processes such as photolithography or laser ablation. Recent works [ 34 ] that reported printed NCS utilized laser ablation to fabricate resistors that encompass a large area overhead owing to the requirement of large resistances. Instead, a fully printed NCS system that includes printed resistors would greatly reduce the resistor area. However, only few works exist on devices fully printed on flexible substrates, and if so, are mainly screen printed [ 30 ]. In this article and, to our knowledge, for the first time, we present a fully printed and flexible two-input neuron for NCSs based on low-voltage EGT technology. The main advantage of the proposed neuron design and its fabrication process is its capability to be embedded as near-sensor processing units for preprocessing raw sensor data, such as matrix of pressure sensors as part of electronic skin. The NCS is first trained in software, and the proposed design is fabricated using easy additive-printing steps. We circumvented previously deployed nonadditive structuring processes and substituted them with additive-printing processes for the EGT-based NCS on flexible polyimide (PI). We used graphene ink for conductive lines to act as resistors or wires depending on their shape, and utilized In 2 O 3 as the active channel material in the EGT owing to its superior performance on flexible substrates such as PI [ 39 ]. 1.3. Proposed Two-Input Neuron As illustrated in Figure 1 , by replicating the neuron intra- and interlayer-wise, a full ANN can be formed for a given topology, which is similar to conventional (software-based) ANNs with few additional constraints that would need to be considered, such as bounded and coupled ANN weights. Thus, each layer of the ANN that is trained in software is fully printed. Given the ease of fabrication and the reproducibility of printed ANNs, a multiple-input neuron can be easily designed and printed with the proposed design.",
"discussion": "4. Discussion 4.1. Measurement Results For the electrical characterization of the printed two-input neuron, we first measured the DC behavior of the printed EGT-based diode, which is part of the pPLU activation function and allows for extracting the transfer function of the pPLU corresponding to Equation ( 4 ). The DC measurement results of the EGT-based diode in Figure 7 a clearly show that the diode had an input voltage regime with a high forward current ( I D S ) and a low current in the reverse direction. The diode’s forward threshold voltage was about 750 mV, which was suitable to observe a nonlinear response of the diode within the operating voltage. Figure 7 a also shows the computed voltage of the pPLU, which validated the nonlinear behavior of this analog activation function circuit. As the resulting resistance was already high enough to obtain a suitable output voltage swing for the pPLU, only one-fourth of the length of the printed graphene-based resistor was used for the pPLU and neuronal measurements (measurement needle for GND came into contact in the middle of the first meander of R o u t ), leading to R o u t = 3.75 MΩ (instead of 4 × 3.75 MΩ = 15 MΩ ). Such high R o u t resistances do not affect the MAC output, which had resistances R i below 1 KΩ, and currents in the MAC circuit were much higher than those in the pPLU. In the second measurement step, we analyzed the MAC circuit and the overall two-input neuronal behavior. Figure 7 b shows the transient measurements of the printed neuron. Output V x of the MAC operation was the sum of inputs V I N 1 and V I N 2 , as expected, i.e., when the inputs were complementary, the MAC output V x was close to 0 V, as resistors R 1 and R 2 were of equal size, and the resulting weights were w 1 = w 2 = 0.5 , respectively. When both inputs were pulled up to 2 V or −2 V, MAC output voltage V x was likewise. Second, the nonlinear response of the pPLU could also be obtained with the measured waveforms. When pPLU input V x was negative, output V o u t was suppressed; inversely, for positive input voltages, V o u t was pulled up to a higher voltage level. In other words, the pPLU response was nonlinear with respect to the MAC output, as desired. In summary, we successfully demonstrated the overall function of the fully printed neuron prototype, and validated that both fundamental neuron operations operated correctly. 4.2. Outlook From the transient measurements ( Figure 7 ), an input–output delay of 1 s was obtained with an average power consumption of 37.7 mW, the calculation of which is shown in Supplementary Information . The consumed area of the hardware prototypewas as small as 16.25 mm 2 . However, a substantial reduction in power consumption could be achieved by choosing smaller resistances of R i , and with a smaller area of printed resistors and contact pads. This is feasible, as a predefined set of ANN weights can be realized by differently sized resistors, e.g., choosing a common scale factor for all resistors R i in Equation ( 3 ) [ 34 ]. Predefining resistors through the training model also allows for the usage of printed resistors, which are positive weights that cannot be dynamically tuned. We, therefore, assumed that our neuron was trained beforehand in software and fabricated accordingly, which basically rendered our ANN as a “post-trained” network that could perform near-sensor processing. Following this approach, our NN is not run-time-adaptive, but targets applications that require low-cost fabrication on flexible substrates. In order to tune the weights, conductive materials such as Pedot:PSS can be printed on top, and this is being investigated. Furthermore, the number of neuronal inputs is usually not limited to two; by adding additional resistors R i to the MAC, inputs can be scaled up in accordance to taking into account the maximal voltage headroom defined by the input signal and power supply. As the interconnects of the circuit, such as measurement pads for drain, source, gate, GND or R o u t , parasitic resistances are introduced to the circuit of the neuronal hardware prototype; their area and consequently their conductance were enlarged compared to passive components R 1 , R 2 and R o u t . Although input contact pads V I N 1 and V I N 2 influenced the effective resistance of input resistors R 1 and R 2 when measuring the hardware prototype, they did not impact ANN weights. Only the ratio of the effective resistance determines the ANN weights, not their absolute value; see Equation ( 3 ). In our experiment, both weights were set to 0.5, and input pads and resistors R 1 and R 2 were of equal size. However, when a neuron is fabricated as part of a commercial product, no measurement pads are required, and only the output impedance of the connected input device (e.g., a sensor) must be lower than neuron input resistors R i ."
} | 3,633 |
29343761 | PMC5772552 | pmc | 6,877 | {
"abstract": "In contrast to the nearly error-free self-assembly of protein architectures in nature, artificial assembly of protein complexes with pre-defined structure and function in vitro is still challenging. To mimic nature’s strategy to construct pre-defined three-dimensional protein architectures, highly specific protein-protein interacting pairs are needed. Here we report an effort to create an orthogonally interacting protein pair from its parental pair using a bacteria-based in vivo directed evolution strategy. This high throughput approach features a combination of a negative and a positive selection. The newly developed negative selection from this work was used to remove any protein mutants that retain effective interaction with their parents. The positive selection was used to identify mutant pairs that can engage in effective mutual interaction. By using the cohesin-dockerin protein pair that is responsible for the self-assembly of cellulosome as a model system, we demonstrated that a protein pair that is orthogonal to its parent pair could be readily generated using our strategy. This approach could open new avenues to a wide range of protein-based assembly, such as biocatalysis or nanomaterials, with pre-determined architecture and potentially novel functions and properties.",
"conclusion": "Conclusion In summary, we have developed an approach to generate an interacting protein pair that is derived from but orthogonal to the parent protein pair. This is achieved by engineering the protein-protein interacting interface and facilitated by a combination of positive and negative selections in bacteria. To the best of our knowledge, our negative selection is the first example of applying URA3/5-FOA selection in a bacterial two-hybrid system. Presumably, more than one orthogonal protein pair can be generated with multiple cycles of positive and negative selections. The approach and tools that were developed in the current work can also potentially be applied to the generation of other orthogonally interacting proteins pairs of one’s interest. These orthogonal protein pairs can potentially be applied to the assembly of artificial protein complexes both in vitro and in vivo . The precise control of relative contents and positions of building blocks within a protein assembly will likely facilitate the construction of protein complexes for protein-based nanomaterials or for efficient catalytic syntheses of bio-based chemicals through co-localization of enzymes",
"introduction": "Introduction Although significant progress has been made in recent years 1 – 14 , the precise manipulation of artificial self-assembly of protein complexes in vitro remains a great challenge. In contrast, highly ordered permanent or transient protein complexes widely exist in nature and participate in virtually every type of cellular function, including catalysis, structural support, bodily movement, signal transduction, transport, etc. Nature’s error-free self-assembly of protein architectures, such as virus capsids 15 , bacterial carboxysomes 16 , and cellulosomes (Fig. 1 ) 17 – 19 is driven by many weak, noncovalent interactions at protein-protein interfaces 20 . The geometry of subunits in a protein complex is precisely defined by those specific noncovalent interactions 13 . In order to mimic nature’s strategy to construct highly defined three-dimensional protein architectures, we need to have highly specific protein-protein interacting pairs, analogous to G-C and A-T base-pairing interactions in DNA. One potential solution is to explore naturally occurring protein pairs, such as the barnase and barstar pair 21 . The other potential approach is to artificially generate mutually orthogonal protein pairs from a known parent protein pair. This could further expand the repertoire of highly specific protein pairs that are available for the assembly of protein complexes. In addition, such evolved protein pairs are orthogonal but consist of high sequence homology to the parent protein pair and, therefore, have similar physical/chemical properties. This may minimize certain complications when protein pairs with very different properties are used in the assembly of a protein complex. Figure 1 Random and controlled assembly of cellulosome. In nature, the assembly of cellulosomes is mediated by a random attachment of a catalytic module (enzyme), through its dockerin domain, to any cohesin positions on the scaffold protein (scaffoldin). The generation of mutant cohesin-dockerin pairs that are orthogonal to the naturally occurring one may allow a controlled assembly of cellulosomes. To demonstrate the feasibility of the aforementioned approach, we selected a type-I cohesin-dockerin pair from Clostridium thermocellum as our model system. High affinity cohesin-dockerin interactions are the basis of self-assembly of cellulosomes 17 , 18 , which are multi-protein complexes from certain anaerobic bacteria and fungi for a highly efficient degradation of cellulosic material (Fig. 1 ). A cellulosome consists of a core structural protein (scaffoldin) that serves as a scaffold to connect multiple catalytic enzymes through the interaction between the type I cohesin domains on itself and the type I dockerin domains of catalytic enzymes. Due to the indiscriminatory nature of cohesin-dockerin recognition within a microorganism species, the assembled cellulosomes have diverse molecular composition and structure, which corresponds to heterogeneous catalytic activities for cellulosic material degradation. In this work, we seek to generate a mutant cohesin-dockerin pair (Fig. 2A ) that is derived from but orthogonal to the naturally occurring (wild-type) one. The generation of orthogonal cohesin-dockerin pairs will allow controlled assembly of cellulosomes (Fig. 1 ), which will facilitate current studies of synergistic actions among cellulosomal enzymes 17 – 19 . The second cohesin domain from the scaffoldin protein (CipA; residues 182–328) and the dockerin domain from a glycoside hydrolase (xylanase 10B; residues 733–791) were used in this study. The crystal structure of the protein complex of these two domains has been reported 22 . Figure 2 Generation of orthogonal protein pairs from a parent protein pair. ( A ) Generation of mutant dockerin-cohesin (Doc 1 -Coh 1 ) pair that is orthogonal to the parental Doc wt -Coh wt pair; ( B ) Structures of Doc wt -Coh wt pair from C. thermocellum . Residues from dockerin (yellow) are labeled in italic and underlined. Residues from cohesin (green) are labeled in regular bold. A few methods have been developed for a high-throughput engineering of protein-protein interactions. Phage 23 , yeast 24 , and bacterial 25 displayed protein libraries are generally screened with either panning or fluorescence-activated cell sorting (FACS). In contrast, the yeast two-hybrid system 26 links protein-protein interactions to a phenotype (e.g., cell growth) that confers a selective advantage to the host, which simplifies the selection process. Bacterial two-hybrid system 27 – 29 has also been developed. In comparison to the yeast system, the bacterial system has the advantage of higher transformation efficiency and faster cell growth rate. However, the current bacterial system lacks a negative (counter) selection scheme. In the present study, we devised a bacterial negative selection scheme (Fig. 3B ) that is analogous to the yeast one 30 , 31 . We subsequently demonstrated that an orthogonal mutant pair could be readily obtained through a combination of positive and negative selections (Fig. 3 ). Throughout the article, we will use Coh wt -Doc wt and Coh 1 -Doc 1 to represent the parent (wild-type) and mutant (evolved) cohesin-dockerin pairs, respectively. Figure 3 Selection scheme. ( A ) positive selection; ( B ) negative selection; ( C ) selection scheme. The positive selection selects mutant cohesin-dockerin pairs that can engage in effective interaction. The negative selection removes any dockerin or cohesin mutants that retain effective interaction with the parent cohesin or dockerin. The combination of negative and positive selection should yield interacting protein pairs that are orthogonal to their parent. Abbreviations: λcI, bacteriophage λ repressor protein; RNAP, α-subunit of RNA polymerase; P lacZ , the lac operon promoter; 3-AT, 3-amino-1,2,4-triazole; 5-FOA, 5-fluoroorotic acid.",
"discussion": "Results and Discussion General approach In order to generate a mutant cohesin-dockerin pair that is orthogonal to its parent pair, we explored a structure-guided, semi-rational protein engineering approach (Fig. 2 ). This approach consists of two essential steps: (1) mutagenesis. The important amino acid residues at the protein-protein interface of the parent pair are randomized. Presumably, such modification would completely abolish or significantly weaken the interaction between mutants and their parents; and (2) selection. This process identifies mutant protein pairs that interact to each other, but do not have significant cross interaction with the parent protein pair. Our selection system consists of both a positive selection and a negative selection (Fig. 3 ). The positive selection selects mutant cohesin-dockerin pairs that can engage in effective interaction (Fig. 3A ). The negative selection removes any dockerin or cohesin mutants that retain effective interaction with the parent cohesin or dockerin (Fig. 3B ). The combination of negative and positive selection should yield interacting protein pairs that are orthogonal to their parent (Fig. 3C ). This selection scheme can likely be generalized and used to create orthogonal pairs for other proteins as well. Positive selection system The BacterioMatch® II two-hybrid system 27 – 29 was used as the molecular biology platform for the positive selection (Fig. 3A ). Two reasons promoted us to choose bacteria ( E. coli ) two-hybrid over yeast two-hybrid as the positive selection system: (1) E. coli grows much faster than yeast; (2) E. coli is transformed with higher efficiency so larger libraries can be readily constructed and selected/screened. BacterioMatch II selection is built upon the genetic complementation of the chromosomal hisB gene deletion by the episomal expression of the S. serevisiae HIS3 gene in an E. coli host strain. Both genes encode imidazoleglycerol-phosphate dehydratase, which is an essential enzyme in the L-histidine biosynthesis. To study the interaction between a cohesin and a dockerin protein, the cohesin is expressed as a C-terminal fusion protein to the full-length bacteriophage λ repressor protein (λcI), and the dockerin is fused to the N-terminal domain of the α-subunit of RNA polymerase (RNAPα). When both fusion proteins are co-expressed in E. coli selection host, if the cohesin and the dockerin variants interact, they recruit and stabilize the binding of RNA polymerase at the promoter and activate the transcription of the HIS3 reporter gene, which allows cells to grow in the presence of 3-amino-1,2,4-triazole (3-AT), a competitive inhibitor of HIS3 gene product. In general, a stronger interaction confers the cells resistant to higher concentrations of 3-AT, while lack of interaction only permits cells to survive on media without 3-AT. It should be noted that other factors, such as the protein expression level, may affect the cell growth as well. For example, a protein pair with higher expression level can likely survive higher concentrations of 3-AT than a protein pair with lower expression level. If one would like to compare the interaction strength between two different protein pairs by using the positive selection system, the expression levels of the two pairs need to be adjusted to a similar level (e.g., to manipulate gene transcription level using different concentrations of inducer such as IPTG). To test if the positive selection works, we examined the interaction between the Coh wt -Doc wt pair. To this end, two plasmids were constructed, including pBT-Coh wt (containing the gene that encodes the λcI-Coh wt fusion protein) and pTRG-Doc wt (containing the gene that encodes the RNAPα-Doc wt fusion protein). We initially examined a construct in which Doc wt was directly fused to RNAPα. However, we observed poor cell growth in a two-hybrid study of the Coh wt -Doc wt interaction. We hypothesized that such poor cell growth was resulted from the degradation of the RNAPα-Doc wt fusion protein since it was known that dockerin domain is prone to degradation in Escherichia coli 22 . According to literature, dockerin-containing enzymes could be expressed as full-length proteins in E. coli 32 – 35 , we decided to improve the stability of the dockerin protein by inserting the X6b carbohydrate-binding domain between RNAPα and the dockerin domain. The X6b domain is naturally fused to the N-terminus of the type I dockerin domain from C. thermocellum and does not interact with the cohesin domain 18 , 19 . The X6b domain was included in all dockerin constructs in this work. As shown in Table 1 , cell growth was observed in the presence of 5 mM of 3-AT when both pBT-Coh wt and pTRG-Doc wt were co-transformed into the E. coli selection host (entry 1; Table 1 ). As negative controls, no cell growth was detected when either pBT-Coh wt was co-transformed with the empty pTRG vector or the pTRG-Doc wt was co-transformed with the empty pBT vector (entry 2, 3; Table 1 ). The results confirmed that cells only grew when both Coh wt and Doc wt proteins were present. In comparison to the positive control provided by the BacterioMatch® II kit, the interaction between Coh wt and Doc wt (entry 1; Table 1 ) led to similar level of cell growth as the interaction between Gal11P and LGF2 (entry 4; Table 1 ), which were co-expressed from the pTRG-Gal11P and the pBT-LGF2 positive control plasmids. Table 1 Examination of Coh wt and Doc wt interaction using The BacterioMatch® II two-hybrid system as the positive selection. entry plasmids in the selection host cell growth (cfu a ) 0 mM 3-AT 5 mM 3-AT 1 pBT-Coh wt and pTRG-Doc wt ~5 × 10 3 ~1 × 10 3 2 pBT and pTRG-Doc wt ~6 × 10 3 0 3 pBT-Coh wt and pTRG ~5 × 10 3 0 4 pTRG-Gal11P and pBT-LGF2 ~6 × 10 3 ~1 × 10 3 5 pBT-Coh wt and pTRG-Doc AL ~5 × 10 3 0 a cfu, colony-forming unit. Negative selection system As a critical component to enable the generation of orthogonally interacting protein pairs, we developed a negative selection method (Fig. 2B ). We modified the yeast URA3/5-FOA counter selection system 30 , 31 into the bacterial two-hybrid system. URA3 encodes orotidine 5′-phosphate decarboxylase, which catalyzes the transformation of 5-fluoroorotic acid (5-FOA) into a highly toxic compound (5-fluorouracil) and causes cell death. A similar approach was demonstrated in a bacterial one-hybrid system to select for Zn finger proteins 36 . To enable the selection, we first deleted the pyrF gene (encodes orotidine-5′-phosphate decarboxylase) on the chromosome of the BacterioMatch II reporter strain to generate strain WNPPI7. As a result, E. coli WNPPI7 lost the ability to convert 5-fluoroorotic acid (5-FOA) into a cytotoxic compound, and therefore survives on solid minimal media containing 5-FOA and uracil supplementation (Entry 3, 4; Table 2 ). We then modified the F’ plasmid of WNPPI7 to replace the HIS3 reporter gene with a copy of the URA3 gene, resulting in strain WNPPI5. However, the 5-FOA tolerance test showed that the basal expression level of URA3 protein in strain WNPPI5 is high enough to lead to cell death of the host strain itself on plates containing 0.5 mM of 5-FOA (Entry 5, 6; Table 2 ). To solve the problem, we constructed strain WNPPI8, in which the URA3 gene was inserted behind the HIS3 reporter gene. As the second gene in an operon, the reduced basal expression level of URA3 in strain WNPPI8 allowed the cells to survive on plates containing 2.5 mM of 5-FOA (Entry 7; Table 2 ). When an interacting protein pair (LGF2 and Gal11P) was expressed in strain WNPPI8, the increased transcription level of URA3 resulted in cell death in the presence of as low as 0.5 mM of 5-FOA (Entry 8; Table 2 ). The negative selection system was further evaluated when pBT-Coh wt and pTRG-Doc wt were co-transformed into WNPPI8. 5-FOA concentrations ranging from 0 mM to 2.5 mM were included in the experiment. In comparison to cell growth on plates without 5-FOA, a significant decrease in colony formation unit was observed when the cells were cultured on plates containing 0.2 mM 5-FOA (data not shown). Further increasing the 5-FOA concentration to either 0.5 or 2.5 mM completely eliminated cell growth (Entry 9, Table 2 ). As controls, no growth defect was observed when WNPPI8 was transformed with either pBT-Coh wt plus pTRG or pTRG-Doc wt plus pBT. The results showed that the URA3/5-FOA negative selection system could efficiently eliminate interacting cohesin and dockerin variants that are generated in the mutagenesis step. Table 2 Examination of negative selection host strains a . host protein pair cell growth (cfu a ) with or without 5-FOA (mM) 0 0.5 2.5 1 BacterioMatch II none ~1 × 10 3 0 0 2 BacterioMatch II Gal11P and LGF2 ~1 × 10 3 0 0 3 WNPPI7 none ~1 × 10 3 ~1 × 10 3 ~1 × 10 3 4 WNPPI7 Gal11P and LGF2 ~1 × 10 3 ~1 × 10 3 ~1 × 10 3 5 WNPPI5 none ~1 × 10 3 ~10 2 (tiny b ) 0 6 WNPPI5 Gal11P and LGF2 ~1 × 10 3 0 0 7 WNPPI8 none ~1 × 10 3 ~1 × 10 3 ~1 × 10 3 8 WNPPI8 Gal11P and LGF2 ~1 × 10 3 0 0 9 WNPPI8 Coh wt and Doc wt ~1 × 10 3 0 0 10 WNPPI8 Coh wt and Doc AL ~1 × 10 3 ~1 × 10 3 ~1 × 10 3 a cfu, colony-forming unit. b In comparison to the regular colony size (~1 mm) from other tests, these colonies (<0.1 mm) were barely seen by eye. Construction of dockerin and cohesin libraries Structural data (PDB code, 1OHZ; Fig. 1B ) 22 revealed that the recognition mechanism of the cohesin-dockerin pair from C. thermocellum is mainly mediated by polar interactions 22 , 37 . The two highly conserved serine-threonine motifs (Ser11/Thr12 and Ser45/Thr46) in dockerin serve as key recognition codes for binding to the cohesin domain 38 , 39 . Due to a near perfect internal two-fold symmetry of dockerin structure (Fig. 2B ), the two serine-threonine motifs interact with cohesin domains in a similar manner and only one motif interacts with cohesin at one time. Both literature data 40 and our experimental observations showed that dockerin mutants with mutations in only one of the two motifs still recognize wild-type cohesin with no apparent decrease in affinity. However, it was reported that mutations in both motifs caused a significant reduction in affinity of the dockerin mutant toward the wild-type cohesin 40 . To further verify this notion, we generated a dockerin mutant (named as Doc AL ), which contained four mutations, Ser11Ala, Thr12Leu, Ser45Ala, and Thr46Leu, in the serine-threonine motifs. We showed that Doc AL did not have strong interaction with Coh wt (entry 5; Table 1 and entry 10 Table 2 ). In order to obtain dockerin mutants that abolish interaction with wild-type cohesin but can potentially recognize a cohesin mutant, we generated a dockerin mutant library in which four residues (Ser11, Thr12, Ser45, and Thr46) in the two Ser/Thr motifs were fully randomized. We also generated a cohesin mutant library in which four residues (Asn37, Asp39, Gly123, and Ala125) that could potentially affect the recognition of the Ser/Thr motif of dockerin were randomized. NNK codons (N = A, C, T, or G, K = T or G; 32 variants at nucleotide level) were used to cover all 20 amino acids at each mutation site. While mutations into certain amino acid residue(s) may affect the stability and/or expression level of the resulting mutants, such effect is in general hard to predict, and therefore is not taken into consideration in the library construction process. Since most of the amino acid residues are encoded by more than one codon, this design leverages concerns of codon bias-caused difference in protein expression and experimental challenges of constructing large mutant libraries. Both the dockerin and the cohesin library had a diversity of 1.05 × 10 6 at the nucleotide level. The dockerin mutant library was cloned into the pTRG vector to generate pTRG-Doc lib in which dockerin mutants were fused to the RNAPα protein. The cohesin mutant library was cloned into the pBT vector to generate pBT-Coh lib in which cohesin mutants were fused to the λcI protein. Identification of an orthogonal cohesin-dockerin pair through library selections The dockerin library was first subjected to one round of negative selection against Coh wt in order to eliminate dockerin mutants that retained effective interaction with Coh wt . Surviving cells should contain dockerin mutants that were either non-functional or lost the ability to recognize Coh wt . Similarly, the cohesin library was subjected to one round of negative selection against Doc wt in order to eliminate cohesin mutants that retained effective interaction with Doc wt . Surviving cells should contain cohesin mutants that were either non-functional or lost the ability to recognize Doc wt . It should also be noted that some mutants that can engage in effective interaction with Coh wt or Doc wt might survive the negative selection if their expression levels were too low to induce a sufficient level of URA3 expression. While such possibility exists, these mutants will likely be eliminated in the positive selection due to their low expression levels. By using plates containing 2.5 mM FOA, the survival rate of the dockerin and cohesin mutants was estimated to be 10–20%. This number was based on the comparison between the control plate (no FOA) and the selection plate (with 2.5 mM FOA; Figure S3 ). We subsequently examined if we could identify cohesin mutants from the reduced cohesin library after negative selection to engage functional interaction with the aforementioned Doc AL mutant. To our delight, a large number of colonies were obtained after one round of positive selection between Doc AL and the reduced cohesin library. We arbitrarily picked eight colonies with different sizes for DNA sequencing analysis. Seven distinct sequences were obtained, while cohesin mutants 3 and 7 converged to the same sequence (Table S2 ). To estimate the effectiveness of the negative selection and to eliminate false positives (e.g., beneficial host mutations) identified in the positive selection, the pBT-Coh plasmids of the seven distinctive mutants were isolated and reintroduced into the positive selection hosts that harbored either pTRG-Doc wt or pTRG-Doc AL . We observed that all seven mutants engaged in strong interactions with Doc AL and supported cell growth in the presence of 7.5 mM 3-AT. On the other hand, six out of the seven mutants were not able to support cell growth when Doc wt was co-expressed. Apparently, these cohesin mutants did not interact with Doc wt , which indicated that our established negative selection protocol was highly effective. We arbitrarily picked cohesin mutant 1 (Coh 1 ; Asn37Leu, Asp39Thr, Gly123Leu, and Ala125Leu; Table S2 ) for the subsequent selection against the reduced dockerin library. Among a few hundred survived colonies, five were picked and the corresponding pTRG-Doc plasmids were isolated and reintroduced into the positive selection hosts that harbored either pBT-Coh wt or pBT-Coh 1 . Cell growth test confirmed that all five dockerin mutants engaged in strong interactions with Coh 1 and four out of five did not interact with Coh wt . One dockerin mutant displayed moderate interaction with Coh wt . Based on the colony size and growth rate on the positive selection plate, we chose a dockerin mutant (named as Doc 1 ; Ser11Arg, Thr12Pro, Ser45Pro, and Thr46Ala) for the following in vitro characterization. In vitro characterization Previous reports suggested that dockerin domain could not be produced as a discrete entity due to its degradation in Escherichia coli 22 . On the other hand, large quantity of dockerin domain could be obtained when it was co-expressed with cohesin 22 , 41 . In addition, many reports showed that good expression of dockerin could be achieved when it was fused to well-folded proteins 32 – 35 . To this end, dockerin domain variants (including its N-terminal X6b domain) were expressed as a C-terminal fusion to the maltose binding protein (MBP). A His 6 tag was added to the C-terminus of the fusion protein to facilitate the purification and ELISA experiments. Cohesin domain variants were purified as a C-terminal fusion to the glutathione S-transferase (GST). The GST tag improved expression of cohesin domains, facilitated protein purification, and did not interfere with the ELISA experiments using anti-His 6 antibody. We have also verified that GST does not interact with MBP. To estimate the strength of interaction between cohesin and dockerin, we conducted semi-quantitative ELISA experiments. Briefly, wells of microtiter plates were coated with a GST-tagged cohesin. Different concentrations of the His 6 -tagged dockerin of interest were then applied into each well. Following washing steps, the amounts of interacting dockerin were determined immunochemically using anti-His 6 antibody and HRP-labeled secondary antibody. As shown in Table 3 , Doc 1 and Coh 1 displayed a strong mutual interaction with a K d value of 4.57 ± 1.61 nM, which is comparable to that of the parent Doc wt -Coh wt pair ( K d = 0.77 ± 0.10 nM). On the other hand, the Doc 1 -Coh 1 pair did not show obvious cross-interaction with the Doc wt -Coh wt pair. The K d values of the Doc 1 -Coh wt and Doc wt -Coh 1 cross pairs were too large to be accurately measured, which were estimated to be larger than 500 nM (Table 3 and Figure S4 ). The ELISA experiments confirmed that the Doc 1 -Coh 1 pair is orthogonal to the parental Doc wt -Coh wt pair. To verify that the evolved Doc 1 mutant does not have increased level of non-specific interaction with other proteins due to a few hydrophilic-to-hydrophobic mutations, we conducted ELISA experiments between Doc 1 and a control protein, BSA. The K d value of the Doc 1 -BSA interaction was too large to be accurately measured. It is estimated to be larger than 1,500 nM, which is similar to that of the Doc wt -BSA pair ( K d > 1000 nM; Figure S4 ). We therefore concluded that mutations in Doc 1 do not promote non-specific binding. Table 3 Kd values of cohesin-dockerin pairs. Doc wt -Coh wt Doc 1 -Coh 1 Doc wt -Coh 1 Doc 1 -Coh wt K d (nM) 0.77 ± 0.10 4.57 ± 1.61 >500 >500 Conclusion In summary, we have developed an approach to generate an interacting protein pair that is derived from but orthogonal to the parent protein pair. This is achieved by engineering the protein-protein interacting interface and facilitated by a combination of positive and negative selections in bacteria. To the best of our knowledge, our negative selection is the first example of applying URA3/5-FOA selection in a bacterial two-hybrid system. Presumably, more than one orthogonal protein pair can be generated with multiple cycles of positive and negative selections. The approach and tools that were developed in the current work can also potentially be applied to the generation of other orthogonally interacting proteins pairs of one’s interest. These orthogonal protein pairs can potentially be applied to the assembly of artificial protein complexes both in vitro and in vivo . The precise control of relative contents and positions of building blocks within a protein assembly will likely facilitate the construction of protein complexes for protein-based nanomaterials or for efficient catalytic syntheses of bio-based chemicals through co-localization of enzymes"
} | 6,973 |
37950419 | PMC10866074 | pmc | 6,878 | {
"abstract": "Abstract The MBES04 strain of Novosphingobium accumulates phenylpropanone monomers as end‐products of the etherase system, which specifically and reductively cleaves the β‐O‐4 ether bond (a major bond in lignin molecules). However, it does not utilise phenylpropanone monomers as an energy source. Here, we studied the response to the lignin‐related perturbation to clarify the physiological significance of its etherase system. Transcriptome analysis revealed two gene clusters, each consisting of four tandemly linked genes, specifically induced by a lignin preparation extracted from hardwood ( Eucalyptus globulus ) and a β‐O‐4‐type lignin model biaryl compound, but not by vanillin. The most strongly induced gene was a 2,4′‐dihydroxyacetophenone dioxygenase‐like protein, which leads to energy production through oxidative degradation. The other cluster was related to multidrug resistance. The former cluster was transcriptionally regulated by a common promoter, where a phenylpropanone monomer acted as one of the effectors responsible for gene induction. These results indicate that the physiological significance of the etherase system of the strain lies in its function as a sensor for lignin fragments. This may be a survival strategy to detect nutrients and gain tolerance to recalcitrant toxic compounds, while the strain preferentially utilises easily degradable aromatic compounds with lower energy demands for catabolism.",
"introduction": "INTRODUCTION Lignin is a complex aromatic polymer and a major component of plant cell walls (Vanholme et al., 2010 ). In nature, peroxidases, oxidases and other accessory enzymes play major roles in breaking down lignin into small molecules (Abdel‐Hamid et al., 2013 ; Bugg et al., 2011 ; Martinez et al., 2005 ). These processes are mainly carried out by fungi, such as white rot fungi, whereas bacterial contributions are limited, especially in terms of the depolymerisation of high‐molecular‐weight lignin (Abdel‐Hamid et al., 2013 ). However, the metabolic capabilities of bacteria are versatile and crucial for the mineralisation of low‐molecular‐weight aromatic compounds, including partially degraded lignin fragments (Bugg et al., 2011 ). The degradation pathways of lignin‐derived aromatic compounds in bacteria have been studied in various bacterial phyla, including Actinobacteriota, Proteobacteria (Psedomonadota) (Bugg et al., 2011 ; Silva et al., 2021 ), Bacillota (Khan et al., 2022 ; Mesle et al., 2022 ) and Bacteroidota (Taylor et al., 2012 ; Wu & He, 2013 ). Metagenomic approaches (Silva et al., 2021 ) have further expanded the known scope of bacterial metabolic capacities during lignin degradation. The mineralisation of low‐molecular‐weight aromatic compounds generally involves oxidative cleavage of aromatic rings followed by degradation in the β‐ketoadipate pathway, which is connected to the tricarboxylic acid cycle (Granja‐Travez et al., 2020 ). Therefore, such mineralisation processes are generally recognised as energy‐acquisition strategies (Granja‐Travez et al., 2020 ; Liu et al., 2019 ). Among the aforementioned bacterial groups, most intensive studies have been conducted using the strain SYK‐6 affiliated with the genus Sphingobium . The enzymatic system that degrades the various lignin‐derived biaryl and monoaryl compounds is composed of various redox enzymes (Kamimura et al., 2017 ). A particularly distinctive feature of the strain SYK‐6 is that it has a cascade enzyme system, called the etherase system, which specifically catalyses the reductive cleavage of the β‐O‐4 ether bond, the major intramolecular bond of lignin. Etherase systems have been detected exclusively in bacteria belonging to the genera Sphingobium , Novosphingobium and Altererythrobacter (Kumagawa et al., 2023 ; Voss et al., 2020 ) within the order Sphingomonadales. In our earlier study, we isolated a Novosphingobium sp. strain MBES04 from deep‐sea sunken wood (Ohta et al., 2015 ). The genome of the strain MBES04 encodes enzymes involved in the etherase system (Figure 1 ), including Cα‐dehydrogenases (SDR3 and SDR5), etherases (GST4 and GST5) and glutathione lyases (GST3 and GST6) (Ohta et al., 2015 ). This combination of recombinant enzymes produced phenylpropanone monomers from the model β‐O‐4‐type lignin biaryl compounds, guaiacylglycerol‐β‐guaiacyl ether (GGGE) and lignin‐containing fractions extracted from wood (Ohta et al., 2017 ). When we cultured the strain in the presence of wood extract, we detected the accumulation of phenylpropanone monomers in the medium. However, the strain was unable to assimilate phenylpropanone monomers produced from the wood extract (Ohta et al., 2015 ). As the reductive cleavage reaction of the β‐O‐4 bond proceeds via net hydrogen transfer (Reiter et al., 2013 ), bacteria cannot gain energy from the reaction itself. The energy balance turns positive once the aromatic low‐molecular‐weight compounds are mineralised. These facts suggest that the etherase system of strain MBES04 is active and has a physiological significance other than energy acquisition from downstream metabolites. FIGURE 1 Proposed metabolic pathways of the MBES04 strain that are involved in glycolysis, the degradation of aromatic compounds and the etherase system. Embden–Meyerhof, Entner–Doudoroff and pentose phosphate pathways are shown in purple, green and blue, respectively. The etherase system and aromatics degradation pathways are shown in brown and magenta, respectively. The substrates that the strain MBES04 uses as a carbon source are indicated in yellow boxes (tested in a previous study, by Ohta et al., 2015 ). DHA, 2,4′‐dihydroxyacetophenone; GGGE, guaiacylglycerol‐β‐guaiacyl ether; GHP, guaiacylhydroxypropanone; GSH, reduced glutathione; GSSG, oxidised glutathione; GST, glutathione S‐transferase; SDR, short chain dehydrogenase reductase (Cα‐dehydrogenase). In our previous study (Ohta et al., 2015 ), we performed gene expression analysis using MBES04 cells in the early logarithmic growth phase under supplementation of GGGE and its oxidised form, (2‐methoxyphenoxy) hydroxypropiovanillone (MPHPV), to identify candidate genes involved in the metabolism of lignin‐derived molecules. We observed weak induction of several enzymes that degrade aromatic monomers in response to GGGE supplementation. However, we did not observe activation of the etherase system or its downstream catabolic pathways of the key intermediate metabolites from partially degraded lignin, such as vanillin and protocatechuate. Furthermore, the addition of MPHPV appeared to suppress cellular energy production, including a decrease in cytochrome activity; however, the biological response to low‐molecular‐weight lignin and its physiological significance could not be elucidated. The present study aimed to elucidate the physiological significance of the functional etherase system of strain MBES04, especially the reason why the products (phenylpropanone monomers) are not utilised as energy sources. We first reanalysed the genome sequence after circulation of the MBES04 genome to identify the full repertoire of genes responsible for aromatic‐compound metabolism and the energy‐acquisition pathways. Subsequently, the responses to lignin‐related aromatic small molecules were analysed using comparative transcriptome analysis. Furthermore, we investigated effector molecules to decipher associations between the etherase system and transcriptional responses to lignin fragments. Using these research approaches, we will understand how the bacterial species isolated from the wood sense lignin and acclimate to lignin‐existing environments, and we believe that such knowledge will be instrumental in the metabolic engineering for the production of valuable chemicals from lignin.",
"discussion": "RESULTS AND DISCUSSION \nGene sets important for energy acquisition\n Previously, we found that strain MBES04 possesses an active etherase system (Ohta et al., 2015 ), producing phenylpropane monomers from lignin model biaryls linked by β‐O‐4 ether bonds, the major bond in lignin molecules (GGGE and MPHPV), and lignin fragments (Figure 1 ). However, the strain was unable to assimilate the phenylpropanone monomer and discharged outside the cell (Ohta et al., 2015 ). In this study, we found that the addition of glucose or alanine to nutritionally limited media enhanced the production of GHP from GGGE during the growth of strain MBES04 (Figure 2 ). Considering that the etherase system consumes reducing power from 2 mol of reduced glutathione for every 1 mol of phenylpropanone monomer production (Figure 1 ), this finding indicates that an energy supply from other carbon sources helped activate the etherase system. To interpret the phenomena, we performed additional sequencing to fill the gap in the chromosome and searched for gene sets associated with pathways that acquire energy from other carbon sources. FIGURE 2 GHP production and growth of strain MBES04 in the presence of glucose or alanine. GHP production (red) from GGGE in the culture medium and growth of the strain (dark blue) in the mineral medium in the presence of glucose (20 mM) or alanine (40 mM) were monitored using HPLC quantification and optical density of the culture at 600 nm, respectively. Sampling was performed every 24 h for 72 h, including the time of inoculation. Abbreviations are listed in Figure 1 . Among the glycolytic pathways, the Embden–Meyerhof pathway (EMP) was defective because the strain lacks 6‐phosphofructokinase [EC: 2.7.1.11]. In addition, the strain lacks 6‐phosphogluconate dehydrogenase [EC: 1.1.1.44, EC: 1.1.1.343], which is required for the pentose phosphate pathway (PPP) (Spaans et al., 2015 ). In contrast, all enzymes required for the Entner–Doudoroff pathway (EDP) were present, suggesting that the EDP plays a central role in glucose metabolism in the strain MBES04. The strain possessed a complete set of genes that encode all five complexes required for oxidative phosphorylation. Thus, oxidative phosphorylation is important for energy acquisition in the strain. The MBES04 genome validated the lack of orthologs of hpvZ that was previously identified as an essential enzyme for GHP (synonym for β‐hydroxypropiovanillone) catabolism in strain SYK‐6 (Higuchi et al., 2018 ). Higuchi et al. ( 2018 ) reported that the degradation of the propanone (C3) side chain of GHP proceeded via a coenzyme A (CoA)‐ and ATP‐demanding reaction. In contrast, we found that strain MBES04 has a gene repertoire enabling the complete catabolism of lignin‐derived aromatic monomers with a C1 side chain, such as vanillin and 4‐hydroxybenzoate (Figure 1 ), consistent with the results of previous culture experiments obtained using a single carbon source (Ohta et al., 2015 ). Strain MBES04 utilises benzoate as a sole carbon source (Ohta et al., 2015 ); however, 3‐oxoadipate CoA‐transferase (PcaIJ) orthologs were not identified in benzoate and catechol ortho‐cleavage pathways, suggesting the presence of alternative genes to PcaIJ. Considering the gene repertory and capability for carbon utilisation, the complex metabolic pattern of MBES04 enhances its robustness in terms of energy acquisition. To compare the energy acquisition pathways with those of other strains in the family Sphingomonadaceae that possess etherase systems, we studied the Novosphingobium sp. PP1Y, N. aromaticvorans DSM12444, Sphingobium sp.SYK‐6 and Altereythrobacter sp. B11. Whole‐genome sequence similarities between the five strains were estimated using digital DNA–DNA hybridization (dDDH) (Meier‐Kolthoff et al., 2013 ) and improved versions of average nucleotide identity (OrthoANI) (Lee et al., 2016 ; Yoon et al., 2017 ) analyses. The dDDH and OrthoANI values of strain MBES04 versus the four strains ranged from 13.4 to 18.9 and from 70.3% to 76.6%, respectively (Table 1 ). The results showed that the genomic similarities amongthe strains are below the cut‐off for the species boundary (dDDH value >70%; OrthoANIu value >95%). TABLE 1 Values of digital DNA–DNA hybridization (dDDH) and improved versions of average nucleotide identity (OrthoANI) using Novosphingobium sp. MBES04 as control. Strain \n Novosphingobium sp. PP1Y \n Novosphingobium aromaticvorans DSM12444 \n Sphingobium sp. SYK‐6 \n Altereythrobacter sp. B11 orthoANI (%) 76.5 73.3 70.3 72.1 dDDH (%) 18.9 14.9 13.4 13.7 TABLE 2 Strain or plasmid used in this study. Strain or plasmid Relevant genotype and main characteristics Reference or source Strain \n Novosphingobium sp. strain MBES04 Wild‐type strain (NBRC114556) Ohta et al. ( 2015 ) MBE lacZ \n MBES04 transformed by pQF‐ lacZ :: P ClusterG‐II, \n Tcr \n This study \n Escherichia coli DH5 α \n \n F′, Φ80dlacZΔM15, Δ(lacZYA \n \n − \n \n argF)U169, deoR, recA1, endA1, hsdR17(rK \n \n − \n \n , mK \n \n + \n \n ), phoA, supE44, λ \n \n − \n \n , thi‐1, gyrA96, relA1 \n TOYOBO Co., LTD (Osaka, Japan) \n Escherichia coli S17‐1 λpir Tp r Sm r \n recA thi hsdRM \n + \n RP42::.Tc::Mu::Km Tn7 λpir phage lysogen de Lorenzo and Timmis ( 1994 ) Plasmids pQF pCM62 with cymR*, PQ5 and MCS for N‐ and C‐terminal fusions to 3 × FLAG tag; Tcr (Addgene plasmid #48095) Kaczmarczyk et al. ( 2013 ) pQF‐ lacZ \n pQF with lacZ (Addgene plasmid #48094) Kaczmarczyk et al. ( 2013 ) pQF‐ lacZ :: P ClusterG‐II \n pQF‐ lacZ with the 600 bp PCR amplicon carrying cluster G‐II promoter region (the map and sequence are shown in Figure S2 ) This study Strains PP1Y, SYK‐6 and B11 shared defects in the PPP with strain MBES04. Strain DSM12444 shared defects in the EMP. As shown in Figure 3 , with respect to glycolysis, all strains lacked at least one of the three pathways. Among Sphingomonadaceae strains possessing the etherase system, strain SYK‐6 was unable to grow sufficiently on most sugars and organic acids (Kamimura et al., 2017 ; Masai et al., 2007 ), which was caused by a lack of genes involved in the phosphotransferase system, resulting in insufficient sugar transport and phosphorylation (Varman et al., 2016 ). FIGURE 3 Comparison of the genome strain MBES04 with that of related strains possessing an etherase system. The phylogenetic tree of the strains possessing etherase systems was constructed using the neighbour‐joining method with a ribosomal protein gene ( rpsC ). The tiles show presence/absence of genes involved in glycolysis (Embden–Meyerhof pathway [purple], Entner–Doudoroff pathway [green] and pentose phosphate pathway [blue]), and aromatics degradation pathways (magenta) for vanillin (left panel) and benzoate/catechol (right panel). Annotation was primarily based on the classification of genes into Kyoto Encyclopedia of Genes and Genomes orthologous groups. Vanillin dehydrogenase was considered present if LigV orthologs (≥70% identity) were found using a BLASTP search. Abbreviations of the referenced enzymes are listed in Table S4 . Many gram‐negative bacteria preferentially use the EDP for glycolysis owing to its lower cost in terms of protein production and more favourable thermodynamic characteristics (Flamholz et al., 2013 ). The EDP efficiently yields NADPH, which confers a high tolerance to oxidative stress (Chavarria et al., 2013 ). In general, gram‐negative bacteria can obtain the energy necessary for survival via oxidative phosphorylation (Stettner & Segr é, 2013 ). Recently, Linz et al. ( 2022 ) constructed a genome‐scale metabolic model of N . aromaticivorans to estimate the efficiency with which aromatic lignin was converted into valuable chemicals. The findings of their investigation suggest that energy derived from non‐aromatic carbon sources is crucial for the demethylation of lignin‐derived aromatic compounds. This is because the demethylation of methoxylated lignin‐derived aromatic compounds is not only an essential step (Abe et al., 2005 ) but also a rate‐limiting step that requires energy input (Venkatesagowda & Dekker, 2021 ) prior to energy production via β‐oxidation, which generates energy in the form of ATP after ring opening. Thus, diverse catabolic pathways for both aromatic and non‐aromatic compounds can compensate for the inefficient energy production from sugar metabolism. \nGene expression in the presence of GGGE \n To clarify the physiological significance of the β‐etherase system of strain MBES04, other than energy acquisition from downstream metabolites, transcriptomic analysis was conducted in the presence of GGGE. GGGE was used as a lignin‐mimicking small substrate cleaved by the β‐etherase system. Here, resting cells, which were prepared from cells aerobically cultured in nutritionally rich media at the growth phase, were used for RNA preparation to obtain sufficient biomass. These experimental settings were designed to avoid the adaptive responses for growth unrelated to β‐etherase reactions. To verify the viability of strain MBES04 cells in the presence of the substrates used in the subsequent transcriptomic experiments, the differences in the colony‐forming units (CFUs) between 0 and 24 h of cultivation were assessed in the presence or absence of the substrates in the nutritionally limited medium (one‐tenth strength of Luria‐Bertani (LB) medium containing 5 mM MgSO 4 ) (Figure 4 ). The increase in CFU after inoculation was 2.6 ± 0.4 × 10 8 CFU/mL in the presence of 2.5 mM GGGE, whereas 2.4 ± 0.1 × 10 8 CFU/mL in the absence of GGGE (control). The growth of strain occurred at the same level in both conditions. Additionally, the reaction of 2.5 mM GGGE by the cell suspension was analysed using LC–MS (Figure 4A ). At 4 h after exposure to GGGE, the production of GHP from GGGE reached a detectable level and was completed within 24 h, indicating the duration of the active metabolic capacity of the cells for subsequent transcriptomic experiments. FIGURE 4 Analysis of reaction of MBES04 cells with GGGE, vanillin and APA. Overnight‐cultured MBES04 cells were collected and incubated in the basal medium with 2.5 mM GGGE (A), 2.5 mM vanillin (B), or 0.1% APA‐lignin (C) for 24 h at 30°C. The aliquots of reaction mixtures were extracted with ethyl acetate triplicate and analysed by LC–MS. For the quantification of the substrate, a calibration curve with standards was used, and extraction errors were adjusted by internal standards. The error bars represent the standard error of the mean of triplicate experiments. In the presence of 2.5 mM GGGE, a significant increase in the overall gene expression levels was observed. Specifically, the expression of 82 and 21 genes increased and decreased by >2‐ and 0.5‐fold, respectively (Figure 5A ; Table S1A ). Notably, the expression levels of three genes in clusters G‐I to G‐III (Figure 6 ; Table S1 ) were 2.2–44.6‐fold higher in the presence of GGGE than under control conditions (Figure 5A ). FIGURE 5 Comparison of global genomic expression in the presence of GGGE, vanillin and APA‐lignin. Red, light blue and dark blue circles represent expression values, RPKMs, of genes in clusters G‐II, L‐III and G‐III/L‐V, and genes of the previously identified etherase system, respectively. Orange, light green and dark green circles show expression values of genes involved in metabolic pathways for benzoate/aminobenzoate degradation, oxidative phosphorylation and ribosomal proteins, respectively. Grey “plus” symbols represent expression values of other genes (CDS) of strain MBES04 with expression (RPKM ≧ 1). The genes showing significant upregulation (2‐fold) and downregulation (0.5‐fold) are depicted above and below the dotted lines. (A) Expression values of GGGE treatments versus control. (B) Expression values of vanillin treatments versus control. (C) Expression values of APA‐treatments versus control. GGGE, guaiacylglycerol‐β‐guaiacyl ether; RPKM, mapped reads per length of transcript/1000 per total million reads. FIGURE 6 Overview of gene clusters of strain MBES04 differentially expressed in the presence of APA‐lignin, GGGE and vanillin. The expression level, RPKM, mapped to the MBES04 genome (grey ring) is represented as a line of colour depending on the tested substrate (None [control]; orange, APA‐lignin; magenta, GGGE; green, vanillin; light blue) within a threshold of 20,000 (Table S1A ). The RPKMs over the threshold are indicated by blue lines. The distribution of differentially expressed gene clusters consisted of four or more genes (V‐I to V‐III, G‐I to G‐IV and L‐I to L‐IV) and are shown outside the rings. The heatmaps display the relative RPKMs to the control condition in the log 2 function presenting the highest (red) and lowest (dark blue) induction. Possible functions based on the KO assignment and BLASTP search are labelled above the heatmap in black and grey, respectively. Each description of the assigned KO is described in Table S1B . Abbreviations are listed in Figure 5 . Cluster G‐I (gene loci MBENS4_0132 to MBENS4_0136) contains pchF and pchC genes; 4‐cresol dehydrogenase (hydroxylating) [EC:1.17.9.1 (transferred from EC:1.17.99.1)] flavoprotein and cytochrome subunits (Figure 6 ; Table S1 ), respectively. The enzyme plays a key role in the degradation of methoxylated aromatic compounds by oxidising the methyl group of the aromatic ring side chain in 4‐cresol and 2,4‐xylenol to a hydroxy moiety (Chen et al., 2014 ; Cunane et al., 2000 ). The formation of 4‐hydroxybenzoate from 4‐cresol through hydroxylation and subsequent oxidation by inducible enzymes enables further energy acquisition via the aromatic degradation pathway (Figure 1 ). Of note, four genes in cluster G‐II (gene loci MBENS4_1158 to MBENS4_1161) were remarkably induced despite their low expression levels under control conditions (Figures 5A and 6 ), with expression changes of 2.87‐, 7.84‐, 16.46‐ and 44.58‐fold, respectively (Table S1 ). The most highly induced gene (MBENS4_1161) in the cluster G‐II is estimated as 2,4′‐dihydroxyacetophenone (DHA) dioxygenase, which produced 4‐hydroxybenzoate (C1 side chain) from DHA (C2 side chain) (Figure 1 ). The products of the other three genes remain hypothetical proteins. Further study is required at the protein level to understand the gene functions. A previous transcriptomics study of strain SYK‐6 (Varman et al., 2016 ) showed that GGGE repressed the overall expression levels of genes required for NAD(P)H production (except for genes that control the interconversion of NADH and NADPH), and strong induction of different genes was not reported. Cluster G‐III encodes a transcriptional regulator, TetR and two sets of transporters of the resistance‐nodulation‐division (RND) family and the EmrAB complex. The RND family transporters are responsible for the efflux of various compounds, such as heavy metals, hydrophobic compounds, amphiphiles and nodulation factors in several bacteria such as strains of Rizobium (Alphaproteobacteria), Ralstnia (Betaproteobacteria) Pseudomonas (Gammaproteobacteria) and other genera (Nies, 2003 ; Zgurskaya et al., 2021 ). The EmrAB complex confers multidrug resistance and stable cellular stability in E. coli (Lomovskaya & Lewis, 1992 ). Cluster G‐IV encodes three multi‐drug efflux membrane proteins and a TetR/AcrR family transcriptional regulator involving multidrug resistance (Table S1B ). Moreover, genes encoding the TetR family of transcriptional regulators were widely distributed in gram‐negative bacteria. Notably, the role of these genes in multidrug resistance has been demonstrated in Salmonella and other pathogenic bacteria (Colclough et al., 2019 ). However, further research is required to better understand the function of the TetR‐like regulators in Cluster G‐III and G‐IV of strain MBES04. \nGene expression in the presence of vanillin\n In some related bacteria, vanillin and vanillic acid are key intermediate metabolites directly involved in downstream pathways of various lignin‐related aromatic compounds, including various biaryls and monoaryls; GGGE, biphenyls, ferulic acid and etc (Bugg et al., 2011 ; Kamimura et al., 2017 ). In addition, vanillin and vanillic acid are major aromatic monomers nonbiologically produced from lignin by chemical oxidation (Ragauskas et al., 2014 ). Strain MBES04 has complete gene sets for vanillin assimilation via vanillic acid (Figure 1 ), which is consistent with previous results of our culture experiment using vanillin (1 mM) as a sole carbon source (Ohta et al., 2015 ). To confirm the effect of a lignin‐derived aromatic monomer, which is not a substrate for the etherase system, on gene expression, we analysed the gene expression profiles in the presence of vanillin. To verify the viability of strain MBES04 cells in the presence of vanillin (2.5 mM), the difference in CFUs was assessed using the same method described above. In the presence of vanillin, the CFU showed an increase of 1.4 ± 0.2 × 10 8 CFU/mL. The growth of the strain was within the same order of magnitude as that of the control (2.4 ± 0.1 × 10 8 CFU/mL), despite partial inhibition. Additionally, the reaction of 2.5 mM vanillin with the cell suspension was analysed using LC–MS (Figure 4B ). Vanillin was completely degraded within 2 h from the onset of exposure, indicating the high metabolic capability of the cells. The presence of vanillin at 2.5 mM resulted in the overall suppression of gene expression, indicating toxicity to the strain. Specifically, the expression of 19 and 64 genes was upregulated and downregulated by >2‐ and 0.5‐fold, respectively (Figure 5B ; Table S1 ). Notably, the expression levels of three genes in clusters V‐I to V‐III (Figure 6 ; Table S1 ) were 0.11–0.49‐fold lower in the presence of vanillin than under control conditions (Figure 5B ; Table S1 ). Among the suppressed genes, severe suppression (0.11‐ to 0.13‐fold) was observed in the cluster V‐II (gene loci MBENS4_1124 to MBENS4_1133) (Figure 6 ) that includes genes for cytochrome c subunits ( ccoP , ccoQ , ccoO and ccoN ) in an oxidative phosphorylation complex compared to that in the control condition (Table S1 ). Additionally, the expression levels of the 30S and 50S ribosomal protein genes in cluster V‐III were greatly suppressed compared to thosein other conditions (gene loci MBENS4_0778 and MBENS4_0779, 025‐ and 0.28‐fold lower, respectively; MBENS4_4412 to MBES04g14421, 0.32‐ to 0.40‐fold lower) (Table S1 ; Figure 6 ). In contrast, regarding genes associated with vanillin catabolism, two sets of out of three ligAB genes, which are aromatic ring‐opening dioxygenases, were significantly repressed (gene loci MBENS4_1991 and MBENS4_1992; 0.12‐ and 0.13‐fold, MBENS4_2242 and MBENS4_2243; 0.41‐ and 0.49‐fold lower expression, respectively). No significant changes in expression levels were observed for tandemly located ligC (gene locus 1g1953), and other ligAB genes (gene loci MBENS4_1954 and MBENS4_1955) (Table S1 ). The ligC gene, which acts downstream of vanillin metabolism, and the NAD(P) + transhydrogenase and phosphoglycerate kinase genes were strongly activated in the presence of vanillin, which is shown in a metabolomic study of strain SYK‐6 (Varman et al., 2016 ). This response was hypothesised to produce sufficient intracellular reducing power to support highly active gluconeogenesis. However, microarray analysis of strain SYK‐6 did not reveal detectable induction of the NAD transhydrogenase and phosphoglycerate kinase genes (Kamimura et al., 2017 ). Genes for aromatic ring and side chain processing enzymes were upregulated in the transcriptome analysis of N. aromaticivorans DSM12444 with a deletion of the gene sacB (SARO_RS09410, Saro_1879) supplemented with glucose and vanillin during its growth phase (Linz et al., 2021 ). These gene expression differences may be influenced by complex stress factors (Święciło & Zych‐Wężyk, 2013 ; Vargas‐Blanco & Shell, 2020 ) and growth phases (Orellana et al., 2017 ). Further research is required to fully understand the underlying mechanisms. In this study, the severe repression of genes that play important roles in energy production (such as oxidative phosphorylation complex genes and ligAB ) (Figure 1 ) indicated that energy production was suppressed with reduced production of ribosomal proteins in response to vanillin at a toxic concentration. Taken together with the rapid and complete degradation of vanillin, these results indicate that the strain entered a dormant state by the time of RNA preparation after catabolising and detoxifying vanillin. \nGene expression in the presence of plant‐derived lignin\n To compare the transcriptional responses to GGGE and vanillin with that to plant‐derived lignin, a lignin fraction, namely APA‐lignin (Nishimura et al., 2022 ), was used for the substrate. APA‐lignin is a lignin preparation holding the β‐O‐4 ether bond as a major intramolecular bond with minor contamination of plant polysaccharides and unidentified small lignin‐derived aromatic molecules. In this study, APA‐lignin was prepared from wood chips of Eucalyptus globulus . The primary constituents were the syringyl and guaiacyl units, with the syringyl unit being more prevalent. The p ‐hydroxyphenyl unit was present in trace amounts. This was consistent with the earlier studies for the composition of E. globulus lignin (Nunes et al., 2010 ; Rencoret et al., 2011 ). To verify the viability of strain MBES04 cells in the presence of 0.1% APA‐lignin, the difference of CFUs was assessed using the same method as above. The increase in CFU was 2.0 ± 0.2 × 10 8 CFU/mL in the presence of APA‐lignin. The growth of strain occurred in the same order of magnitude as in the control (2.4 ± 0.1 × 10 8 CFU/mL). To enhance our understanding of the microbial conversion of lignin fragments by strain MBES04, we analysed the reaction products from APA‐lignin by the cell suspension for 24 h under the same experimental conditions as those for the transcriptomic analysis using LC–MS (Figure 4C ). By the time of RNA preparation, 4 h after exposure to APA‐lignin, GHP and syringylhydroxylpropanone (SHP) were produced at a detectable level. A dynamic change in the overall gene expression levels was observed. Specifically, the expression of 59 genes increased and decreased by >2‐ and 0.5‐fold, respectively (Figure 5C ; Table S1 ). The overall profile of gene expression is likely a combination of differentially expressed genes in the presence of GGGE and vanillin. Severe suppression (0.03‐ to 0.21‐fold) was observed in cluster L‐II (gene loci MBENS4_1124 to MBENS4_1133) (Figure 6 ), which was the same gene set for cluster V‐II, compared to that under the control condition (Table S1 ). Cluster L‐III, the same gene set for cluster G‐II, showed tremendous expression (7.59‐ to 88.55‐fold) similar to the GGGE stimulus. Cluster L‐IV was upregulated by 2.11‐ to 2.79‐fold. Four out of six genes in the cluster were upregulated by vanillin as well. Cluster L‐IV contains three Fe‐S cluster assembly‐related proteins (SufD, SufC and SufB), a SUF system Fe‐S cluster assembly regulator, and two hypothetical proteins (Figure 6 ). Fe‐S cluster assembly proteins were known to be oxidative response proteins conserved in bacteria (Chiang & Schellhorn, 2012 ). The redox state and subunit structure of Fe–S clusters modulate environmental adaptation by regulating numerous biological processes, including respiration, photosynthesis, nitrogen fixation, DNA replication and repair, RNA modification and gene regulation (Mettert & Kiley, 2015 ). Cluster L‐V was detected at the same gene loci as cluster G‐III, which functions in detoxification by driving proton influx and drug efflux (Zgurskaya et al., 2021 ). To obtain direct evidence of the involvement of the molecules produced during the degradation of APA‐lignin by the etherase system, we analysed the enzymatic reactions for APA‐lignin using LC–MS according to the method described in our previous report (Ohta et al., 2017 ) (Figure 7 ). The main reaction products generated by the etherase enzymes were GHP and SHP, consistent with those observed in strain MBES04 cells. Small amount of p ‐hydroxyphenylhydroxylpropanone (HHP) was also detected in LC–MS extracted chromatograms based on the calculated monoisotopic mass of HHP. The peak areas of HHP under the chromatograms were 7.3 ± 0.6% by enzyme reactions and 5.4 ± 2.6% by MBES04 cells when those of GHP were considered as 100%. The results indicate that GGGE and β‐O‐4‐containing small molecules present in APA‐lignin are responsible for inducing changes in gene expression after being transported into the cells. Notably, GHP was a common reaction product generated from GGGE and APA‐lignin. This observation suggests the potential of GHP as an effector molecule responsible for the induction of common genes upon the addition of GGGE and APA‐lignin. Consequently, the etherase system confers a typical phenotypic feature to strain MBES04 when encountering the lignin fragments. FIGURE 7 Total ion current chromatograms and mass spectra of enzyme and microbial reactions with APA‐lignin. Five enzymes (SDR3, SDR5 and GST3‐5) and 0.2% APA‐lignin were incubated with 20 mM reduced glutathione and 10 mM NAD sodium salt at pH 8.5 and 15°C for 24 h. Strain MBES04 was incubated in a basal medium containing 0.1% APA‐lignin at 30°C for 24 h. Samples were extracted by ethyl acetate and analysed by liquid chromatography‐mass spectrometry (LC–MS). IS, internal standard. \nIdentification of effector molecules that induce the expression of cluster G‐II genes\n The transcription of clusters G‐II and L‐III initiated from the same starting point, which was estimated to be approximately 100 base pairs upstream of the cluster, based on the redundancy of RNA‐sequencing reads (Figure S1 ). We constructed a promoter assay plasmid (pQF‐ lacZ :: P ClusterG‐II ) by inserting the 1–600 bp upstream of MBENS4_1161 (the starting gene of cluster G‐II and L‐III, MBENS4_1158 to MBENS4_1161) directly upstream of the galactosidase gene start codon (Figure S2 ). Galactosidase activity of strain MBES04 transformed by pQF‐ lacZ :: P ClusterG‐II , namely MBE lacZ , was measured in cultures supplemented with GGGE, VGGE, MPHPV, GVG, GHP, vanillin, VA, 2,6‐DMP, DHA, or guaiacol (Figure 8 ). The strongest induction was observed with GHP, and induction was also observed with GGGE, MPHPV and DHA. In contrast, no induction was observed with vanillin or methoxyphenol treatment. Our time‐course observations with GGGE, MPHPV and GHP showed that induction occurred earliest with GHP (after 6 h), whereas delayed induction was observed with GGGE and MPHPV (after 12 h), as shown in Figure 9 . These results suggest that GGGE and MPHPV function as effector molecules after they are converted to GHP—the final reaction product from GGGE and MPHPV by the intrinsic etherase system of the strain. These results indicate that GHP acts as an effector molecule that induces the expression of genes in clusters G‐II and L‐III. FIGURE 8 Screening of effector molecules that induce the expression of cluster G‐II genes. Gene expression activity from 600 bp of putative promoter of cluster G‐II (P ClusterG‐II ) was assessed using a lacZ ‐based promoter assay plasmid (pQF‐ lacZ :: P ClusterG‐II ). The β‐galactosidase activity of strain MBES04 holding pQF‐ lacZ :: P ClusterG‐II , namely MBE lacZ , was measured after 16 h cultivation in basal medium supplemented with 1 mM each of lignin‐related compounds, GGGE, MPHPV, VGGE, GVG, GHP, DHA, vanillin, VA, 2,6‐DMP, or guaiacol, to screen for effector molecules. β‐Galactosidase activity was measured using o ‐nitrophenyl β‐D‐galactopyranoside and expressed as Miller Units. The error bars represent the standard error of the mean of quadruplicate experiments. DHA, 2,4′‐dihydroxyacetophenone; DMP, dimethoxyphenol; GGGE, guaiacylglycerol‐β‐guaiacyl ether; GHP, guaiacylhydroxypropanone; GVG, β‐guaiacyl‐α‐veratrylglycerone; MPHPV, (2‐methoxyphenoxy) hydroxypropiovanillone; VA, veratryl alcohol; VGGE, veratrylglycerol‐β‐guaiacyl ether. FIGURE 9 Determination of the effector molecule based on the time course of gene expression activity from P ClusterG‐II . Changes in β‐galactosidase activities of MBE lacZ cells cultured in basal medium supplemented with 1 mM GGGE, MPHPV, GHP, GGGE, MPHPV, or GHP were measured immediately after sampling every 2 h. Both GGGE and MPHPV were converted to GHP and guaiacol by the intrinsic etherase system of MBE lacZ during cultivation. β‐Galactosidase activity was assayed and expressed as Miller Units (see Figure 8 ). The error bars represent the standard error of the mean of quadruplicate experiments. Abbreviations are listed in Figure 8 . Effector molecules for genes related to the catabolism of lignin‐derived aromatic molecules have been reported for several bacteria that possess etherase systems (Kamimura et al., 2010 , 2017 ; Uchendu et al., 2021 ). Vanillic acid, gallic acid, protocatechuic acid (Araki et al., 2019 , 2020 ; Kamimura et al., 2017 ) and hydroxycinnamoyl‐CoAs (Kasai et al., 2012 ) are known to act as effectors of cis‐elements that regulate catabolism. A recent study of transcriptional regulation by the TetR‐type transcriptional repressor, PprR, in Aromatoleum aromaticum EbN1 T (classified as a Betaproteobacteria) indicated that environmental bacteria can sense very low concentrations of lignin‐derived 3‐phenylpropanoates by forming their coenzyme A thioesters as effectors (Vagts et al., 2021 ). This study demonstrated how bacteria regulate their metabolism to adapt to different nutritional conditions. To the best of our knowledge, this is the first study to show that phenylpropanone monomers act as effector molecules, but without promoting the degradation of phenylpropanone, they promote the degradation of other lignin derivative compounds. \nMetabolic implications of the etherase system\n Degradation of phenylpropanone monomers generated from the etherase system requires activation of the C3 side chain with CoA and ATP, which is energetically demanding. Instead, strain MBES04 prefers to assimilate enzymatically degradable substances, such as glucose, DHA, vanillin, cresol, benzoate, with low energy cost. Such substrates are available in the environment because they are released via chemical and biological degradation of organic materials, such as plant biomass, especially from the lignin, and petroleum‐based chemicals. At the case, we further considered what is the role of the etherase system for strain MBES04. As a sensing system for nutritional plant biomass, strain MBES04 constantly expresses genes for an etherase system that produces phenylpropanone from lignin fragments. From our comparative transcriptomic results, the phenylpropanone induced multiple enzymes, including dioxygenase for aromatic monomers with a C2 side chain and detoxifying transporters for a variety of recalcitrant toxic compounds. These metabolic traits are likely correlated with a survival strategy of the strain MBES04 to detect the presence of nutrients, and utilise more easily assimilated aromatic compounds rather than phenylpropanone, and also to gain the tolerance to the toxic aromatic compounds. In order to clarify the metabolic traits, the cellular response to lignin fragments should be further investigated by means of proteomic and metabolic analyses or using living cells with knockouts of the respective genes. In the future, we will conduct further study on the functions of induced genes, more specifically, enzymatic properties of DHA dioxygenase and other enzymes, regulatory proteins, and binding domains in cluster G‐II/L‐III."
} | 9,912 |
34183815 | PMC8764749 | pmc | 6,879 | {
"abstract": "Bacterial species have diverse cell shapes that enable motility, colonization, and virulence. The cell wall defines bacterial shape and is primarily built by two cytoskeleton-guided synthesis machines, the elongasome and the divisome. However, the mechanisms producing complex shapes, like the curved-rod shape of Vibrio cholerae, are incompletely defined. Previous studies have reported that species-specific regulation of cytoskeleton-guided machines enables formation of complex bacterial shapes such as cell curvature and cellular appendages. In contrast, we report that CrvA and CrvB are sufficient to induce complex cell shape autonomously of the cytoskeleton in V. cholerae . The autonomy of the CrvAB module also enables it to induce curvature in the Gram-negative species Escherichia coli, Pseudomonas aeruginosa, Caulobacter crescentus , and Agrobacterium tumefaciens . Using inducible gene expression, quantitative microscopy, and biochemistry we show that CrvA and CrvB circumvent the need for patterning via cytoskeletal elements by regulating each other to form an asymmetrically-localized, periplasmic structure that directly binds to the cell wall. The assembly and disassembly of this periplasmic structure enables dynamic changes in cell shape. Bioinformatics indicate that CrvA and CrvB may have diverged from a single ancestral hybrid protein. Using fusion experiments in V. cholerae, we find that a synthetic CrvA/B hybrid protein is sufficient to induce curvature on its own, but that expression of two distinct proteins, CrvA and CrvB, promotes more rapid curvature induction. We conclude that morphological complexity can arise independently of cell shape specification by the core cytoskeleton-guided synthesis machines.",
"discussion": "Discussion CrvA and CrvB have specialized roles in generating cell curvature In addition to the genetic evidence suggesting unique roles for CrvA and CrvB, our cell biological description of their roles in filament assembly provides a mechanistic basis for their asymmetry-generating activity. The function of CrvA to assemble small, CrvB-independent structures is likely due to assembly-promoting coiled-coil domains in the N-terminal region of the protein where CrvA shares greater homology with CrvY relative to CrvB 13 . Rather than forming CrvA-independent structures, CrvB promotes higher order CrvA assembly through its CBS domain that can even be functional when directly attached to CrvA, as in the CrvA CBS chimera that readily forms large filaments ( Extended Data Figure 5A ). However, our titration of CrvB levels suggests that for optimal curvature induction, CrvA must be present in excess of CrvB ( Figures 5B - D ). The stoichiometry of the N-terminal and CBS domains is fixed at 1:1 in a CrvA/CrvB hybrid, suggesting that splitting these domains among two genes may have allowed the evolution of a more favorable stoichiometry of N-terminal domains to CBS domains. This may also explain why merely increasing the dosage of CrvA CBS was not sufficient to recover wild-type curvature dynamics, which could have significant functional implications for a rapidly-growing pathogen like V. cholerae ( Figure 5C ). PG binding could enable CrvAB to regulate cell shape independently of the core shape machinery While the enzymes bacteria use to build PG are numerous, functionally redundant, and may differ between species, the cell wall itself is largely similar in composition among Gram-negative bacteria 33 . It is impossible to rule out interaction with every cell wall enzyme, but we found that CrvAB can induce curvature independently of the MreB-dependent elongasome, the FtsZ-dependent divisome, PBP1a, and PBP1b, which include all known spatially-regulated PG synthesis complexes and the enzymes that produce the bulk of total PG. In contrast to previously-characterized shape determinants, the combination of periplasmic localization, filament formation, and direct PG binding suggest a unique mechanism wherein CrvAB filaments directly interact with the cell wall to locally slow its expansion. The compressive force from a PG-bound filament would locally resist the expansion of the cell wall by growth and may explain our previous finding that CrvA locally slows PG insertion 13 . Alternatively, CrvAB could locally inhibit the activity of an as-yet-unidentified cell wall synthesis enzyme. Autonomous modules: a distinct class of cell-shape determinant The surprising ability of the CrvAB module to function in heterologous species separated by 2.5 billion years of evolution 34 suggests a previously-undescribed route for the evolution of cell shape complexity. Previous studies of crescentin and SpmX identified MreB co-option as a strategy for producing complex shapes, but these mechanisms only function in species that possess MreB 17 , 35 . In contrast, we show that CrvAB can heterologously add new geometric features by both generating asymmetry and inhibiting cell wall growth through direct PG binding. As a result, the CrvAB module can even curve species that lack MreB such as A. tumefaciens . There are multiple complex bacterial morphologies for which no determinants have been identified. Our work suggests that these diverse forms may be built not by co-option of core cell biology, but by autonomous cell shape modules like the one that makes V. cholerae curved."
} | 1,341 |
36554943 | PMC9779662 | pmc | 6,880 | {
"abstract": "For the agricultural development of dumps, increase in land use efficiency and protection of food security, to verify the safety, efficacy and sustainability of field-applied arbuscular mycorrhizal fungi (AMF) inoculum, and to exclude the risk of potential biological invasion, in this study, we determined the effect of AMF inoculation and intercropping patterns (maize–soybean) on the temporal dynamics of soil parameters, native AMF communities and crop yields. AMF communities were analyzed using Illumina MiSeq. A total of 448 AMF operational taxonomic units (OTUs) belonging to six genera and nine families were identified. AMF inoculation treatment significantly improved the yield of intercropping maize and increased the content of available phosphorus. AMF diversity was significantly influenced by cropping pattern and growth stage, but not by the inoculation treatment. Inoculation altered the AMF community composition in the early growth stage and facilitated a more complex AMF network in the early and late growth stages. These results indicate that AMF inoculation affects native AMF only in the early stage, and its impact on yield may be the consequence of cumulative effects due to the advantages of plant growth and nutrient uptake in the early stage.",
"conclusion": "5. Conclusions To explore the feasibility and sustainability of agricultural development on coal mine dump land, we determined whether AMF inoculation and cropping patterns could affect native AMF communities and the yield of maize and soybean. Our results demonstrated that AMF diversity was significantly influenced by cropping pattern and growth stage, but not by AMF inoculation. Notably, AMF inoculation altered the native AMF community composition in the early growth stage, leading to a more complex AMF network structure in the early and late growth stages. These results indicate that the effect of AMF inoculation on native AMF may only exist in the early stage, and its impact on crop yield may be the consequence of cumulative effects due to the advantages of plant growth and nutrient uptake in the early stage. Future research will focus on the impact of biofertilizers on the nutritional quality of crops and their microbiochemical processes in the soil.",
"introduction": "1. Introduction The Anthropocene, the present geological epoch defined by human footprints, has wreaked havoc on the ecosystem and jeopardized the global food supply [ 1 ]. Coal mining results in an abundance of ecological and environmental problems. Due to excessive coal mining, vast areas of land and flora have been destroyed, resulting in changes to the soil structure and physicochemical qualities of the soil, as well as the construction of large dump sites [ 2 , 3 ]. However, during the process of dump formation, the original topography was reshaped to form a flat platform, e.g., a “terraced field”, which is suitable for agricultural development and increases in grain production. In our previous studies, we showed that soil nutrients and microbial communities were restored to a greater extent in dumps with more than 10 years of reclamation [ 4 , 5 ]. Therefore, these areas may meet the conditions for agricultural development. However, extensive use of chemical fertilizers and water in traditional agriculture is likely to cause secondary damage to the reclaimed land. Thus, it is important to use more efficient fertilization programs and agricultural management models for agricultural development in these areas. Given the economic and environmental expenses associated with irrigation and chemical fertilization, contemporary agriculture should use microbiomes to optimize production while minimizing input in the case of future environmental upheavals [ 6 , 7 ]. Traditional large-scale agricultural operations place a heavy emphasis on water and chemical fertilizer, while microbiomes play an important role in boosting plant absorption of soil water and nutrients [ 6 , 8 ]. As well as contributing to increased crop yields, Arbuscular mycorrhizal fungi (AMF) play a significant role in ecosystems (e.g., soil structure, nutrient conservation, and plant stability in changing environments) and may lower the quantity of fertilizer necessary for cost effectiveness [ 9 ]. AMF are a significant group of root-associated mutualists in the plant microbiome, and they may develop mutualistic partnerships with over 80% of terrestrial plant species [ 10 ]. AMF exchange soil-derived nutrients for photosynthates from the host plant in this relationship [ 10 ]. AMF relationships have been found in both field and laboratory studies to improve soil nutrient status and plant growth in post-mining environments [ 11 , 12 ]. Because mycorrhizal associations can range from mutualism to parasitism depending on environmental and species-specific factors [ 13 ], AMF could help to encourage more mutualistic or parasitic partnerships. Currently, there is considerable controversy among researchers about how native AMF communities respond to the addition of exogenous AMF inoculum [ 14 , 15 ] and whether the application of AMF inoculum poses a bioinvasive risk to the regional ecology [ 16 , 17 ]. Some studies have shown that AMF inoculation has little or no effect on native fungal communities [ 18 , 19 ], whereas others have shown that exogenous AMF inoculum can displace dominant native microbial taxa and cause disturbance to native microbial communities [ 20 , 21 ]. Scholars are more concerned about the negative impact on plant productivity due to the reduction of soil biodiversity caused by the spread of inoculated AMF to non-target areas [ 22 , 23 ]. Therefore, it is necessary to assess the changes in native AMF communities after the application of AMF inoculum. Intercropping, often known as polyculture or mixed cropping, is a common farming strategy used across the world to prevent soil-borne plant pathogens from accumulating [ 24 ]. Intercropping strategies can improve the ecosystem’s agro-quality while also assisting in the management of diseases, weeds, and pests [ 25 , 26 ]. This is mostly accomplished by antagonistic secondary metabolites produced by one plant root that successfully inhibits the pathogen of another plant [ 27 ]. Increased yield, production sustainability, ecosystem development, and environmental safety are all benefits of the intercropping system. In an intercropping system, two or more crop species are cultivated at the same time; they cohabit and interact with one another and with the agro-ecosystems [ 26 ]. Microbes’ temporal dynamics have been exploited to identify elements that influence community organization and ecological functions [ 28 , 29 , 30 ]. Shifts in the diversity and community structure of AMF assemblages across time and space are linked to plant community succession, anthropogenic activities, and changes in environmental conditions [ 31 , 32 , 33 , 34 , 35 ]. Current understanding of the effects of AMF inoculation is mainly based on short-term greenhouse experiments, with more research on plant performance before and after AMF inoculation [ 16 ], and less research on the sustainability of AMF inoculants under field application conditions. To determine the safety, sustainability and effectiveness of using AMF as biofertilizer in the agricultural development of open-pit coal mine dumps and to exclude potential environmental risks and hazards, this study investigated the effects of AMF inoculum on the community of native AMF and its coupling with soil factors under different cropping patterns (monocrop and intercrop) during three periods of crop growth. This study aims to improve crop yields and soil fertility by using biofertilizers that are beneficial to the environment. The economic, ecological, and social benefits of coal mine dumps can be achieved through the production of green, organic, and healthy products for human consumption.",
"discussion": "4. Discussion In this study, we showed that AMF treatment increased the yield of maize and soybean, which is consistent with previous results showing that seed inoculation with selected isolates can increase plant root colonization and crop productivity [ 52 , 53 , 54 ]. The impact of AMF inoculation on soil microbial biodiversity has recently become a hot issue of discussion [ 15 , 55 ]. We showed that AMF treatment changed the native AMF community composition in the early stage of growth but had no significant effect in the middle and late stages of growth and did not affect AMF alpha diversity. These results indicate that the exogenous inoculants used in this study did not significantly affect the native AMF community. The interactions of various AMF isolates have been shown to be synergistic, neutral, or antagonistic [ 56 , 57 , 58 ]. Moreover, indigenous AMF, which is better adapted to the local conditions, can outcompete some of the inoculated fungi [ 59 ]. The dynamic link between plant dependency on mycorrhizal associations and nutrient availability, eutrophication, or growth-limiting circumstances is largely responsible for the influence of edaphic variables on AMF community structure [ 60 ]. Physical and chemical characteristics of the soil may also have a significant effect on the symbiotic connection between plants and fungus [ 61 ]. We found that the AMF inoculation significantly affected the soil total organic carbon, available P and available K, especially the content of available P. The establishment of mycorrhizal symbioses is helpful for the mobilization and absorption of phosphorus by plants in AP-limited soils [ 10 ]. In general, the synergy and complementarity of diverse modes of action might give additional advantages when soil microorganisms with different properties are used together [ 62 ]. It is also conceivable that the observed favorable benefits are attributable to the expansion of the mycorrhizal niche in the environment [ 56 , 63 ]. The temporal patterns of AMF alpha diversity and community composition were not affected by inoculation treatment or cropping patterns. This might be due to AMF’s reproductive phenology being strongly constrained by evolutionary limitations [ 64 ]. Additionally, AMF diversity increased significantly in the second stage, which could be due to the soil nutrient depletion caused by plant consumption during this period. Traits that allow for early colonization of host plants, such as the generation of more AMF spores, may have significant tradeoffs with the reserved competitive ability [ 65 ]. Alternatively, host plants may preferentially transfer incentives (in the form of increased photosynthate allocation) to more advantageous partners, resulting in changes in AMF community temporal dynamics over time [ 66 ]. We found that neither cropping pattern nor AMF inoculation disrupts the temporal dynamics of AMF community composition. The findings indicate that AMF inoculation has no effect on the native AMF community composition and that the manner of regulation of soil AMF community composition may be influenced by complex environmental conditions that promote orderly succession through several growth stages. In the network analysis, we observed that the nodes of some AMF taxa, e.g., Diversispora and Claroideoglomus in the inoculation and control treatments, respectively, increased throughout the growth stage. Glomus species’ dominance might be attributed to their environmental adaptability, host specificity, functional significance, or ease of reproduction in the soil environment [ 31 , 67 , 68 ]. Intercropping is a practice for improving soil fertility and crop yield. When maize and soybeans are intercropped, maize has a substantial competitive advantage for soil nutrition over soybeans, resulting in changes in soil parameters [ 69 ]. Subterranean root–root interactions between intercropped crops might result in a heterogeneous distribution of nitrogen in the soil profile, hence increasing N input into the cropping system through symbiotic N fixation [ 70 ]. However, our investigation found that cropping pattern had a substantial effect on the NH 4 + –N and NO 3 − –N content, which is consistent with earlier research [ 71 ]. Changes in aboveground plant diversity can modify soil characteristics, hence affecting microbial diversity in intercropping systems [ 72 , 73 ]. Numerous studies have shown that intercropping may enhance the N, K, and TOC levels of the soil [ 74 , 75 ]. Furthermore, several variables, such as soil type, soil condition, plant species and nutrition, have been observed to influence soil fungus diversity [ 76 ]. As a result, intercropping systems have an effect on soil fungal diversity, which may alter in response to variations in plant variety and soil fertility [ 77 ]."
} | 3,181 |
38148858 | PMC10749938 | pmc | 6,882 | {
"abstract": "Addressing the pressing issues of increased food demand, declining crop productivity under varying agroclimatic conditions, and the deteriorating soil health resulting from the overuse of agricultural chemicals, requires innovative and effective strategies for the present era. Microbial bioformulation technology is a revolutionary, and eco-friendly alternative to agrochemicals that paves the way for sustainable agriculture. This technology harnesses the power of potential microbial strains and their cell-free filtrate possessing specific properties, such as phosphorus, potassium, and zinc solubilization, nitrogen fixation, siderophore production, and pathogen protection. The application of microbial bioformulations offers several remarkable advantages, including its sustainable nature, plant probiotic properties, and long-term viability, positioning it as a promising technology for the future of agriculture. To maintain the survival and viability of microbial strains, diverse carrier materials are employed to provide essential nourishment and support. Various carrier materials with their unique pros and cons are available, and choosing the most appropriate one is a key consideration, as it substantially extends the shelf life of microbial cells and maintains the overall quality of the bioinoculants. An exemplary modern bioformulation technology involves immobilizing microbial cells and utilizing cell-free filters to preserve the efficacy of bioinoculants, showcasing cutting-edge progress in this field. Moreover, the effective delivery of bioformulations in agricultural fields is another critical aspect to improve their overall efficiency. Proper and suitable application of microbial formulations is essential to boost soil fertility, preserve the soil’s microbial ecology, enhance soil nutrition, and support crop physiological and biochemical processes, leading to increased yields in a sustainable manner while reducing reliance on expensive and toxic agrochemicals. This manuscript centers on exploring microbial bioformulations and their carrier materials, providing insights into the selection criteria, the development process of bioformulations, precautions, and best practices for various agricultural lands. The potential of bioformulations in promoting plant growth and defense against pathogens and diseases, while addressing biosafety concerns, is also a focal point of this study.",
"conclusion": "10 Conclusion and future prospects The primary focus in advancing agricultural productivity to meet the needs of our growing global population lies in investing in the development of microbial formulations. This greener approach supports plant growth and environmental sustainability. While bacterial strains often perform well in laboratory settings, their efficacy in field conditions is hindered by factors such as poor survivability, inappropriate carrier selection, or ineffective delivery methods. To ensure the success of bioformulations, the process begins with the critical task of selecting microbial strains carefully. These chosen strains must possess a competitive edge against native microflora while demonstrating beneficial functions even under stressful conditions, all the while maintaining their bio-efficacy once released. Creating an effective bioformulation demands several essential steps, including proper isolation and characterization of the microbial strains for their plant growth-promoting traits. Additionally, rigorous testing for pathogenicity is necessary to ensure bio-safety. Moreover, the selection of an ideal carrier is crucial to enhance the shelf life of the bioformulation and preserve its efficacy. Field conditions play a vital role in determining the success of a bioformulation. Therefore, it is imperative to assess the survival of the formulated product in real-world agricultural settings. The overall cost of developing and implementing the formulated product should be considered to ensure its feasibility and practicality on a larger scale. Shifting the research focus towards the development of broad temperature and elevation ranged bioinoculants based bioformulation, harnessing their potential metabolites, holds the key to advancing sustainable and safe practices. Rather than solely concentrating on the isolation and characterization of new bacterial bioformulation, this approach offers several benefits by utilizing bioinoculants bioformulation that relies on potential metabolites, we can significantly enhance field efficacy while simultaneously addressing biosafety concerns. These bioformulations can be tailored to deliver targeted benefits, promoting plant growth, disease resistance, and nutrient uptake without the risk associated with introducing entirely new bacteria into the environment. Moreover, there is a pressing need to explore ways to stabilize these bioformulations and increase their shelf life. By doing so, we ensure their long-term viability and practicality for widespread agricultural adoption, promoting cost-effectiveness and convenience. To achieve this, research efforts should be directed toward identifying numerous inexpensive and non-toxic carrier materials. These materials can play a crucial role in preserving the bioformulations’ effectiveness and longevity, allowing farmers easy access to sustainable solutions without imposing harmful consequences on the environment or human health. Lastly, to truly replace agricultural chemicals and make agriculture more sustainable and productive, it is essential to investigate effective delivery methods. Implementing innovative delivery techniques can ensure that bioinoculant bioformulation reaches their target areas efficiently, maximizing their beneficial impact on crops and reducing the need for conventional chemical interventions. By emphasizing these research areas—developing specific bioinoculants bioformulation based on potential metabolites, stabilizing formulations, exploring eco-friendly carrier materials, and optimizing delivery methods—we pave the way for a more sustainable, productive, and environmentally friendly approach to agriculture.",
"introduction": "1 Introduction In the last few decades, rampant chemical fertilization and biomagnification of hazardous chemicals in the food chain has posed a threat to human health and destroyed the health of the soil. The deterioration of soil fertility and decline in the indigenous beneficial soil microbial population led to decreased crop production. Hence, an alternative and green approach is needed to maintain agricultural productivity without reliance on chemical fertilization. The use of microbial bio-formulations offers an alternative approach for utilizing beneficial plant microorganisms to achieve good plant growth and productivity. The use of bio-formulated products, especially biofertilizers, has been widely popularized as an alternative to the agrochemicals ( Khan et al., 2020a ; Pathak et al., 2022 ; Ayilara et al., 2023 ). Therefore, the term bio-formulation can be represented as the ‘development of material containing living but valuable microbial strains, using suitable carrier materials for their productive use in agriculture, industry, bioremediation, etc ( Balla et al., 2022 ). The key ingredients of a bio-formulated product/bioformulation are potential microbes, possessing plant growth promoting properties including nutrient solubilizers, nitrogen fixers, biocontrol agents, and bioremediation ( Pirttila et al., 2021 ). The major goals of microbial formulations preparation are: (i) to create an appropriate environment for the bioinoculants functioning, ii) to provide physical and chemical protection for an extended period of time to circumvent a rapid reduction in cell viability during storage, (ii) to support the competition of inoculants with the indigenous soil microbiota, and (iii) to reduce losses engendered from depredation by the local micro-fauna. Another goal, however, is to provide a sufficient source of live bioinoculant cells that are accessible for interaction with plants and the soil microbiome ( Vassilev et al., 2020 ). It has been observed that direct use of plant beneficial microorganisms in the green house or small scale is fine but on field or large scale, viability issue of the microorganisms gets enhanced. Indeed, it is necessary to obtain a significant number of microbial cells (at least 10 6 -10 7 ) in order to obtain a positive response of the formulated product ( Bashan et al., 2014 ; Vassilev et al., 2020 ). The abiotic substrates, which have the ability to provide a safer environment for microbial cells and can accommodate viable and physiologically active cells, are called as carrier substances. Solid or liquid materials are used as ‘carriers’ for the development of various microbial formulations, depending on the product type ( Naik et al., 2020 ). The solid formulations are produced in solid, powdery, or granular form and are based on either inorganic or organic carriers. Various carrier materials such as peat, vermiculite, coal, compost, perlite, agro-industrial waste, polysaccharides, etc. are used to produce the most important solid formulations. In contrast, liquid-based formulations also contain microbial cultures with desirable properties, modified with additives that improve the viscosity, constancy, and dispersibility of the cell suspension ( Mishra and Arora, 2016 ). In recent years, formulation technologies have paid more attention to the immobilization of cells, since the tactic of gel-cell immobilization is the technological solution that can better ensure the quality and standardization of the formulated product. In addition, particular attention has recently been paid to cell-free formulations ( Tewari et al., 2020 ). These formulations resemble fermentation broth and encompass various metabolic products, including metal chelators (siderophores), antibiotics, enzymes, notably those with lytic capabilities, toxins, and soluble phosphate. Collectively, these components have the potential to exert a beneficial influence on plant growth. Delivery of bioformulations is a mandatory step, done either by inoculating the soil directly or by treating plants/seeds ( Rocha et al., 2019a ). The escalating concern over the inadequate uptake of chemical fertilizers by plants and their detrimental impact on ecosystems, alongside a global rise in apprehension regarding pollution, greenhouse gas accumulation, and an increased emphasis on plant-based food production, has led to a surging demand for biofertilizer agents. Farmers are increasingly embracing biofertilizers to sustainably and organically cultivate their crops. To date, numerous biofertilizers have been successfully commercialized for various environmental conditions and crops. However, a significant obstacle to the widespread success of biofertilizers in agroecosystems is the lack of knowledge in selecting and correctly applying them. This knowledge gap erodes the confidence of farmers in biofertilizers. Hence, there is a critical need to disseminate knowledge within farming communities about the scientifically sound methods of selecting and applying correct microbial bioformulations according to their native environment and crops."
} | 2,810 |
39710696 | PMC11665205 | pmc | 6,884 | {
"abstract": "Background Currently, efficient technologies producing useful chemicals from alternative carbon resources, such as methanol, to replace petroleum are in demand. The methanol-utilizing yeast, Komagataella phaffii , is a promising microorganism to produce chemicals from methanol using environment-friendly microbial processes. In this study, to achieve efficient D-lactic acid production from methanol, we investigated a combination of D-lactate dehydrogenase ( D-LDH ) genes and promoters in K. phaffii . The yeast strain was constructed by integrating a gene cassette containing the identified gene and promoter into the rDNA locus of K. phaffii , followed by post-transformational gene amplification. Subsequently, D-lactic acid production from methanol was evaluated. Results Among the five D-LDH genes and eight promoters tested, the combination of LlDLDH derived from Leuconostoc lactis and CAT1 and FLD1 promoters was suitable for expression in K. phaffii . GS115_CFL/Z3/04, the best-engineered strain constructed via integration of LlDLDH linked to CAT1 and FLD1 promoters into the rDNA locus and post-transformational gene amplification, produced 5.18 g/L D-lactic acid from methanol. To the best of our knowledge, the amount of D-lactic acid from methanol produced by this engineered yeast is the highest reported value to date when utilizing methanol as the sole carbon source. Conclusions This study demonstrated the effectiveness of combining different enzyme genes and promoters using multiple promoters with different induction and repression conditions, integrating the genes into the rDNA locus, and further amplifying the genes after transformation in K. phaffii . Using our established method, other K. phaffii strains can be engineered to produce various useful chemicals in the future. Supplementary Information The online version contains supplementary material available at 10.1186/s13068-024-02596-0.",
"conclusion": "Conclusions In this study, LlDLDH gene from L. lactis was linked to pCAT1 and pFLD1, integrated into the rDNA locus of K. phaffii , and subjected to post-transformational gene amplification to construct an engineered yeast strain, GS115_CFL/Z3/04, capable of producing D-lactic acid from methanol. To the best of our knowledge, the amount of D-lactic acid produced by this engineered yeast is the highest reported to date when methanol is used as the sole carbon source. In lactic acid production from methanol, strategies such as overexpression of transcriptional activators [ 56 ] and inhibition of the reverse reaction from lactic acid to pyruvate [ 47 ] have been employed to increase lactic acid production. Integrating these genetic modification strategies with the strategy in this study likely further increases the production of D-lactic acid. On the other hand, it has been reported that the methanol assimilation ability of K. phaffii decreases when it is cultured for a long period of time using methanol as the sole carbon source [ 60 ]. Addressing this issue is also important for efficient production of D-lactic acid from methanol. This study demonstrated the effectiveness of combining different enzyme genes and promoters using multiple promoters with different induction and repression conditions, integrating the genes into the rDNA locus, and amplifying the genes after transformation in K. phaffii . This study outlines a method to engineer other K. phaffii strains capable of producing various useful chemicals in the future.",
"discussion": "Discussion In this study, we investigated the type of D-LDH gene expressed and optimal combination of promoters to improve D-lactic acid production from methanol in K. phaffii . The highest D-lactic acid production was achieved by integrating the LlDLDH gene from L. lactis linked to pCAT1 and pFLD1 into the rDNA locus of the K. phaffii genome and performing post-transformational gene amplification. The results indicate that the combination of expressed genes and promoters, use of multiple promoters, and multicopy integration of genes are effective for the production of D-lactic acid from methanol by engineered K. phaffii . In this study, two types of D-LDH genes, LlDLDH and LpDLDH , were expressed in K. phaffii , and they achieved higher D-lactic acid production than conventional LmDLDH [ 24 ]; the highest D-lactic acid production was achieved with LlDLDH (Fig. 1 ). Watcharawipas et al. also expressed these three D-LDH genes in S. cerevisiae and reported that the highest D-lactic acid production is achieved with LpDLDH , followed by LlDLDH and LmDLDH [ 28 ]. In contrast, in a study on D-LDH expression in O. polymorpha , almost no D-lactic acid was produced with LmDLDH [ 44 ]. This indicates that the effective D-LDH gene varies depending on the host microorganism. Therefore, type of D-LDH gene expressed in each host should be considered for high D-lactic acid production. In this study, high D-lactic acid production was achieved when pCAT1 or pFLD1 was used to express D-LDH (Fig. 2 A). Of the eight promoters used in this study, two promoters that improved D-lactic acid production over conventional pAOX1, pCAT1, and pFLD1 are known to be strong methanol-inducible promoters, along with pAOX1 [ 45 ]. Vogl et al. reported that pFLD1 is comparable to pAOX1 and that pCAT1 reaches expression levels up to 1.8-fold higher than those of pAOX1 in K. phaffii [ 46 ]. When pCAT1 or pFLD1 was used, the D-lactic acid concentration nearly reached a plateau at 48 h (Fig. 2 A). This is likely due to a decrease in the pH of the medium or a lack of nitrogen sources in the medium. Similarly, Wu et al. reported that the pH decrease due to lactic acid accumulation reduced lactic acid productivity from methanol [ 47 ]. In addition, the nitrogen source in the minimal medium was limited [ 48 ], and the D-lactic acid production likely reached a plateau due to competition between the nitrogen source for cell growth and D-lactic acid production. In terms of cell proliferation, optimal methanol concentration for culturing K. phaffii is approximately 4.0 g/L [ 49 ]. In addition, methanol concentrations < 10 g/L have been used to produce heterologous proteins and useful compounds using methanol-inducible promoters in K. phaffii [ 35 , 50 – 54 ]. In contrast, methanol concentration used in this study (30 g/L) was higher than that used in other studies. Only a little is known about the expression intensity of methanol-inducible promoters at methanol concentrations > 10 g/L. However, similar to the galactose-inducible promoter, pGAL1, in S. cerevisiae [ 55 ], the level of gene expression driven by the methanol-inducible promoter in K. phaffii may be dependent on the methanol concentration in the medium. pCAT1 and pFLD1 may have higher expression levels than pAOX1 in the presence of high concentrations of methanol. Therefore, exploration of useful promoters at high methanol concentrations is important for the efficient production of various useful chemicals from methanol. When LlDLDH was expressed in K. phaffii using two mutant promoters, pCAT1m and pGAP1m, almost no D-lactic acid production was observed with either promoter (Fig. 2 A). pCAT1m and pGAP1m are mutants of pCAT1 and pGAP1, respectively, with a partial modification of the promoter sequence to improve GFP fluorescence [ 35 , 38 ]. Gene expression levels vary depending on the gene and promoter combination [ 29 , 30 ]. Therefore, mutant promoters effective for the expression of GFP were not suitable for the expression of LlDLDH , leading to low D-lactic acid production. In the future, identification of suitable promoters for target gene expression will be important for the production of chemicals from methanol. GS115_CFL/Z3/04 strain, constructed by integrating the LlDLDH gene linked to two promoters (pCAT1 and pFLD1) and amplifying it after transformation, produced a maximum of 5.18 g/L of D-lactic acid after 168 h of cultivation (Fig. 5 B). There have been several reports of lactic acid production by methylotrophic yeasts using methanol as a carbon source. Bachleitner et al. achieved L-lactic acid production of 17 g/L using methanol and glycerol as co-substrates [ 56 ]. On the other hand, lactic acid production using methanol as the sole carbon source has been reported in three studies, one for D-lactic acid production [ 24 ] and two for L-lactic acid production [ 44 , 47 ]. The final lactic acid production levels in these studies were 3.48 g/L [ 24 ], 3.8 g/L [ 44 ], and 4.2 g/L [ 47 ], respectively. Therefore, to the best of our knowledge, this study achieved the highest lactic acid production when methanol was used as the sole carbon source. Previous studies have reported that the combined use of the two promoters, pAOX1 and pGAP1, enhances the production of heterologous proteins in K. phaffii [ 57 ]. This is possibly because, even under conditions of repression of one promoter, gene expression is maintained at a high level by the action of the other promoter, thereby increasing the production of the target protein. Strengths of the promoters of some methanol metabolism-related genes in K. phaffii are reduced by amino acids in the culture medium [ 58 ], and the repression conditions of pCAT1 and pFLD1 used in this study were different [ 35 , 59 ]. Combination of the two promoters with different repression conditions, pCAT1 and pFLD1, in this study enabled D-LDH to be constantly expressed at a high level and to continue to stably produce D-lactic acid over a long period, thereby achieving the high D-lactic acid production. Therefore, simultaneous expression of key metabolic enzymes using multiple promoters is a useful strategy for chemical production in K. phaffii ."
} | 2,444 |
35113532 | null | s2 | 6,885 | {
"abstract": "There are many strategies to actuate and control genetic circuits, including providing stimuli like exogenous chemical inducers, light, magnetic fields, and even applied voltage, that are orthogonal to metabolic activity. Their use enables actuation of gene expression for the production of small molecules and proteins in many contexts. Additionally, there are a growing number of reports wherein cocultures, consortia, or even complex microbiomes are employed for the production of biologics, taking advantage of an expanded array of biological function. Combining stimuli-responsive engineered cell populations enhances design space but increases complexity. In this work, we co-opt nature's redox networks and electrogenetically route control signals into a consortium of microbial cells engineered to produce a model small molecule, tyrosine. In particular, we show how electronically programmed short-lived signals (i.e., hydrogen peroxide) can be transformed by one population and propagated into sustained longer-distance signals that, in turn, guide tyrosine production in a second population building on bacterial quorum sensing that coordinates their collective behavior. Two design methodologies are demonstrated. First, we use electrogenetics to transform redox signals into the quorum sensing autoinducer, AI-1, that, in turn, induces a tyrosine biosynthesis pathway transformed into a second population. Second, we use the electrogenetically stimulated AI-1 to actuate expression of "
} | 374 |
35527862 | PMC9069710 | pmc | 6,886 | {
"abstract": "The prevention of excessive water uptake in wood in order to avert discoloration, swelling and decay is a major challenge for wood-based applications. We developed a facile surface treatment to protect wood from liquid water uptake that does not require harsh process conditions or toxic solvents. Water-based and surfactant-free dispersions of sub-micron alkyl ketene dimer wax particles were prepared and sprayed onto wood substrates. After the evaporation of water, the wax particles self-assembled into distinctive platelet structures. Depending on the specific conditions of application, water contact angles as high as 166° were measured on treated wood surfaces. The implementation of sub-micro structures clearly reduced surface gloss but transparency and color remained largely unaffected. The method is comparably cost-effective and scalable, overcoming dimensional limitations crucial for many applications of wood.",
"conclusion": "Conclusion It was shown that surfactant-free dispersions of sub-micron AKD particles in water, which remain stable for at least one week, can be produced in a simple fashion. Upon application to wood surfaces and subsequent drying, AKD particles self-assemble to nanometer-thin platelets which endow the treated surface with superhydrophobicity. Up to the melting range, increased drying temperature favors platelet development, same as prolonged drying time. With the exception of reduced surface gloss, no unwanted alterations in the optical appearance of treated wood surfaces were observed, making the method a promising approach for facile and cost-effective hydrophobisation of woody substrates.",
"introduction": "Introduction Wood is a renewable, easily accessible and inexpensive resource that has been used by humankind for thousands of years. As a building or furniture material it has numerous favorable properties such as high stiffness and strength at comparably low weight, 1 appealing visual appearance, and low embodied energy and carbon impact. 2 Today, due to the growing awareness for the environmental impact of different materials, but also because of demand for its advantageous properties, the use of wood is increasing. 3 Major drawbacks of wood as a structural material lie in its susceptibility to biological degradation, and in dimensional instabilities that can result in cracking or warping. Both can reduce the service life of wood-derived products dramatically. These issues are caused by fundamentally different processes but share one common origin, i.e. the water uptake of wood due to its hydrophilic nature. During the past decades, extensive research was carried out on measures preventing decay and dimensional instabilities of wood. Conventional wood modification comprises various approaches that target optimized bulk behavior during water contact. 4–6 Thermal and chemical modification – essentially acetylation and furfurylation – turned out to be the most promising methods and resulted in successful market implementations. 7 Nevertheless, bulk wood modification faces fundamental drawbacks such as the necessity for elaborate processes, unsuitability of certain wood species, or dependence on high chemical loadings. Often, modification results in reduced mechanical properties or unwanted color changes. Such disadvantages might increase the prices or limit possible applications of modified wood products. To overcome the wetting of wood with water and ensuing dimensional change or biodegradation, recent research aimed for the development of highly water-repelling surfaces. This approach follows an interdisciplinary trend of superhydrophobicity research that started in the beginning of this century and has now reached most diverse industries and materials. 8 Superhydrophobic surfaces are characterized by water contact angles (WCA) higher than 150° and can show other interesting features such as self-cleaning, anti-icing or anti-fogging ability. 9,10 Initiated by insight on the surface of the Lotus ( Nelumbo nucifera ) leaf, 11 theoretical principles for such exceptional water-repellency were discussed frequently, 12,13 focusing on hierarchical structures created by low surface free energy materials. The first approaches for the preparation of superhydrophobic wood surfaces were based on the deposition of zinc oxide onto the wood cell wall, which was then modified with stearic acid. 14,15 The natural roughness of wood provides a primary surface pattern on the microscale. Implementation of a second, nanoscale roughness can then lead to superhydrophobic hierarchical surface structures. Following this principle, several approaches were based on the deposition and subsequent modification of inorganic nanostructures such as silica, 16–18 TiO 2 , 19,20 WO 3 , 21 or ZnO. 22 Others applied small-scale structure onto a hydrophobic PDMS coating. 23–25 These modified surfaces of intrinsically hydrophilic wood substrate reached WCA above 150° but were prepared using complex or time-consuming processes. Most of these procedures consist of two or more production steps, including at least one immersion or autoclave treatment hardly suitable for continuous and efficient large scale production. Few single-step applications have been reported. Examples are immersion in PMS solution for 18 hours, 26 or a six-hour hydrothermal treatment in an autoclave, 27 resulting in similar drawbacks as mentioned above. Further, many of the above-named approaches are based on the use of hazardous raw materials or solvents. In contrast, a more environmentally-friendly process was published by Lozhechnikova et al. , 28 who developed surfactant-free carnauba wax dispersions for layer-by-layer deposition with ZnO nanoparticles on wood surfaces. Still, 8 bilayers had to be developed via 16 immersion repetitions to reach superhydrophobicity. Similarly to wood, also paper requires a certain degree of water-repellency in many applications, which is why decades of experience in wax-based hydrophobisation of cellulosic material can be found in paper industry. Alkyl ketene dimer (AKD) emulsion patented in 1949 29 is the most widely used paper sizing agent beside alkenyl succinic anhydride. 30,31 AKD consists of a 2-oxetanone ring system with an alkyl group and an alkylidene group attached on the 3- and 4-position, respectively ( Fig. 1 ). For fully AKD sized paper, maximum (advancing) WCA of 110° are reported. 32 Significantly higher WCA has been shown for AKD-coated glass substrate, 33,34 where the drying of an AKD melt under nitrogen gas atmosphere resulted in the growth of fractal structures exhibiting WCA as high as 174°. Later, a superhydrophobic AKD coating of similar structure was developed on paper using rapid expansion of supercritical CO 2 . 35 High pre-expansion pressure from 100–300 bar but no toxic solvents were used in this process. The resulting surface consisted of thin, randomly oriented but mainly upright AKD platelets of few micrometers in length. Fig. 1 Chemical structure of alkyl ketene dimers with R 1 and R 2 representing straight carbon chains of usually 16–18 carbon atoms in length. The present study reports the utilization of AKD for generating highly water repellant wood surfaces. Based on the spraying of water-based dispersions of sub-micron AKD particles, we aim at a simple and straightforward process involving neither high temperature or pressure, nor sophisticated infrastructure or the extensive use of solvents. The influence of particle size, the number of spraying repetitions, and the drying conditions on the ultimate wetting behavior will be investigated together with the effect of AKD coating on the visual appearance of wood surfaces.",
"discussion": "Results and discussion Properties of AKD dispersions The preparation parameters of dispersions together with the particle properties obtained are listed in Table 1 . Particle size was influenced by the intrinsic wax concentration and, even more clearly, by the preparation method. The use of high energy ultrasonication (dispersions V–VIII) resulted in mean particle sizes below 300 nm. Target wax concentration of 10 g L −1 led to opaque-whitish dispersions, indicating a higher final concentration than in transparent dispersions prepared using only 1 g L −1 of wax. The cooling method had no influence on ultrasonicated variants. The properties of the dispersions I–IV, prepared by means of blending, showed higher deviations. Ice-cooling yielded smaller mean particle size and PDI than slow cooling at room temperature. Dispersion III was the only specimen with a mean particle size above 1 μm and had a high DPI of 0.56. Apparently, a high number of large particles was lost during filtration. Based on the requirement of high particle concentration in dispersion, two variants with distinctly different mean particle size were selected for further experiments. The variants IV and VII, with mean particle sizes of 841 nm and 291 nm, respectively, fulfilled these requirements. In the following paragraphs, they will be referred to as “AKD-841 nm” and “AKD-291 nm”, respectively. In the production of AKD dispersions, surfactants are commonly used to provide stability. 31 Such stabilizers can affect the surface energy of the dispersed particles and, consequently, lower the hydrophobisation effect achieved on a treated surface. The stability of colloidal suspensions depends on the sum of van der Waals attractive forces and electrostatic repulsive forces. 37 Zeta potential provides information on the latter and is therefore often related to colloidal stability. Nanoparticle dispersions with absolute zeta potential values above 30 mV are usually presumed highly stable. In this work, fairly stable surfactant-free AKD dispersions were obtained, which was supported by their zeta potential values. For the variant AKD-841 nm, a zeta potential value of −31.6 mV was measured, whereas a slightly lower (but, more importantly, higher absolute) value of −38.1 mV was determined for AKD-291 nm. We did not observe any changes of the stability of dispersions within the first week after their preparation. Dispersions were used for spraying applications without any difficulty. During longer storage, some particles started to form agglomerations that floated on the surface. Still, the dispersions remained opaque and whitish, indicating a rather high remaining concentration of homogeneously dispersed particles. It is proposed that surface charge inherent to the AKD particles was responsible for the temporary stability of the dispersions. Representative particle size distribution curves obtained by DLS measurements of both variants are shown in Fig. 2 . Samples of AKD-841 nm were mainly bidisperse. Small size particles showed an average intensity peak at approximately 170 nm, and a larger particle peak at approximately 1270 nm, with a polydispersity index (PDI) of 0.44. For AKD-291 nm, comparable peaks were located at approximately 90 nm and 540 nm with a mean PDI of 0.42. Some measurements of that variant (such as the example shown in Fig. 2 ) revealed a third fraction at approximately 5 μm particle size. The intensity of Rayleigh scattering, which is the underlying concept of DLS measurements, is proportional to the sixth power of the radius. 38 Thus, large particles are generally overrepresented in DLS intensity measurements and these peaks correspond to a rather small number of particles. The preparation of AKD-291 nm by means of high-energy ultrasonication resulted in very small average particle size and high absolute zeta potential values. Nonetheless, the straightforward blending method used for AKD-841 nm also resulted in stable dispersions in spite of the higher average particle size obtained. Fig. 2 Exemplary particle size distribution measurements of AKD dispersions. AFM images of particles from highly diluted and then dried dispersions confirmed bidispersity. Fig. 3a shows a high number of particles with a mean particle size of approximately 200 nm. The majority of single particles reaches heights of only 20–50 nm and thus can be described as rather flat spheroids. Fig. 3b shows particles of the same dispersion, which correspond to the second DLS intensity peak. Diameters of 1–2 μm and heights up to 1.3 μm indicate a rather spherical shape, as is evident from the three-dimensional depiction of a particle in the inset in Fig. 3b . Fig. 3 AFM images of dried AKD-841 nm suspension showing a small (a) and a large particle fraction (b). The z -axis value scale of the 3D inset was set 1 to prevent height distortion in the depiction of the particle. Influence of the spray treatment on microscale roughness and optical properties of wood surfaces The surface characteristics of wood are determined by various factors such as the material's fibrous anatomical structure, density deviations, annual ring variation, and processing parameters such as the type of machining used. 39 Consequently, sanded uncoated wood surfaces possess microscale roughness. To provide water-repellency, solid wood substrate was sprayed with AKD dispersions between 2 and 12 times. Fig. 4 shows SEM images of sanded wood that appeared comparatively even ( Fig. 4a ), and wood after different numbers of spraying repetitions with aqueous AKD dispersion ( Fig. 4b and c ). The water-based spray treatment resulted in the swelling of fibers exposed to the surface. After four repetitions, fibers of millimeters in length were clearly protruding from the surface, increasing the surface roughness to hundreds of micrometers. Nevertheless, little influence of the deposited wax particles on roughness was visible at this level of magnification ( Fig. 4b ). After 10 spraying cycles, the swollen fibers appeared puttied with a layer of wax. Further, agglomerated particles with spherical or spheroidal shapes of diameters up to 150 μm introduced an additional level of roughness at this high number of spraying cycles (arrows, Fig. 4c ). Fig. 4 SEM images of sanded wood (a) and wood sprayed 4× (b) and 10× (c) with AKD-291 nm reveal changes of the surface roughness during the spraying treatment. Scale bar corresponds to 500 μm. Increasing amounts of deposited AKD wax did not only influence the micro-level surface structure, but also the optical properties. Fig. 5 shows how color change, expressed by Δ E *, increased with an increasing number of spraying cycles. The corresponding photograph of the samples shows a gradual lightening from the native reddish beech wood color to a whitish surface. On variants sprayed 12 times, not only obvious color change, but even large wax agglomerations were visible. Notably, coatings obtained by means of only two spraying repetitions were transparent, preserving the natural appearance of such variants. The gloss values diminished from approximately 6 GU for uncoated wood to almost 0 GU after only two spraying repetitions, indicating a strong effect of wax coating on surface gloss. Several studies on optics have shown, that increasing surface roughness increases diffuse light scattering and therefore reduces gloss. 40,41 Hence, almost completely mat surfaces indicated distinctive surface structures after the spraying of wax dispersions. Fig. 5 Color change, gloss after the spraying treatment, and a photograph of samples sprayed 2–12 times with AKD-841 nm or AKD-291 nm dispersion. Error bars represent standard deviations. Influence of wax particle size and spraying repetitions on surface structure and wetting behavior of wood surfaces Previous studies have shown that water-repellency is limited even for the chemically lowest surface free energy materials, as long as insufficient surface structure is created. 42 To investigate the success of our intention of equipping wood substrate with small-scale roughness by means of AKD particles, we performed SEM and WCA measurements. Fig. 6 and 7 show surfaces of samples sprayed 2–10 times (a–e) with AKD-871 nm and AKD-291 nm dispersions, respectively. The SEM images revealed that sub-micron particles sprayed onto the wood surface assembled into numerous flat, upright wax platelets with lengths of about 1–5 μm and widths of approximately 100–300 nm. Collectively, the assembled platelets form branched, porous wax layers on the substrate surfaces. Fig. 6 SEM images of samples sprayed 2–10 times with AKD-841 nm, scale bars correspond to 50 μm and 5 μm (inset), respectively (a–e). Influence of spraying repetitions on the WCA after 60 s of exposure, exemplary WCA behavior over 60 s, and an image of the maximum WCA measurement of a sample sprayed 10×. Error bars represent standard deviation (f). Fig. 7 SEM images of samples sprayed 2–10 times with AKD-291 nm, scale bars correspond to 50 μm and 5 μm (inset), respectively (a–e). Influence of spraying repetitions on the WCA after 60 s of exposure, exemplary behavior over 60 s, and an image of the maximum WCA measurement of a sample sprayed 10×. Error bars represent standard deviation (f). The coverage of the surface area with thin platelets is visible in Fig. 6a , but the structures of some regions seem underdeveloped when only two spraying repetitions of AKD-841 nm dispersion were performed (boxes in Fig. 6a ). Small, hardly connected platelets rather than a branched and interconnected layer, and a relatively high number of spherical particles, that had not (yet) transformed into platelets (arrows, inset) are present. Four spraying repetitions ( Fig. 6b ) resulted in a more homogeneous surface structure, but large particles still remain in spherical shape (arrows, inset). Some small regions lack wax particles, resulting in uncovered parts or very small platelets (box). By means of six spraying repetitions, a sufficient amount of wax was deposited inducing the development of a smooth, rather homogeneous AKD coating ( Fig. 6c ). No obvious surface defects and a decreasing number of spheres were detected. This tendency persisted also for surfaces sprayed 8–10 times. Consequently, hardly any large particles originating from the original dispersions are present in Fig. 6d and e . It was assumed, that the extended exposure to temperature during the iterated drying process promoted the transformation from large wax spheres into thin platelets on repeatedly sprayed substrates. Excessively deposited AKD formed comparatively large wax agglomerations in the diameter range of tens of micrometers (box Fig. 6e ), already visible in Fig. 4c . The wetting behavior of the coated wood substrates is shown in Fig. 6f , expressed by mean WCA measured 1 min after droplet deposition. Overall, the initially hydrophilic surface of beech wood was endowed with excellent water-repellency. A mean WCA of approximately 139° was measured after only two spraying repetitions with AKD-841 nm. Remarkably, even the imperfect surface structure of this variant resulted in an incipient WCA of 152° but values decreased over the total observation time. All other samples reached a superhydrophobic state expressed by WCA ranging from approximately 158–162°. These contact angles remained constant throughout the full measurement period, as depicted in the insert graph of a variant sprayed four times. An exemplary image extracted from the automatic WCA calculation is also shown in Fig. 6f , demonstrating the accuracy of the Young–Laplace fit for modelling a large volume droplet shape. Coatings prepared using the AKD-291 nm variant are displayed in Fig. 7 . They reveal a similar wax layer as observed for the AKD-841 nm dispersion ( Fig. 6 ). Interestingly, little influence of the intrinsic particle size on the resulting shape of single platelets was observed. Even so, it appears that the usage of smaller particles favors a more homogeneous coverage of the surface already at a low number of spraying repetitions ( Fig. 7a and b ). Considerable formation of large agglomerations due to excess AKD was observed after the sixth spraying repetition ( Fig. 7c ). The inset of Fig. 7e shows that such wax agglomerations are covered with a layer of thin platelets. Overall, few large spherical shapes were found. Same as with AKD-841 nm the sophisticated surface structure lead to stable superhydrophobic behavior for all AKD-291 nm variants, starting with as few as two spraying-drying cycles. The mean WCA ranged from approximately 156–165° for variants sprayed four and ten times, respectively. Apparently, for both dispersions used there existed a critical number of spraying repetitions necessary to reach superhydrophobic water-repellency. For AKD-841 nm it took four iterations, whereas a stable WCA of 159° was already after spraying twice with AKD-291 nm. Any additional deposition of wax did not seem to have significant influence on the wetting behavior. Influence of drying conditions on surface structure and wetting behavior DSC measurements of AKD revealed a specific melting behavior shown in Fig. 8 . Considering the melt range shown, different samples of one variant (4× AKD-291 nm) were repeatedly sprayed and dried for 5 min at temperatures from 20–80 °C in order to investigate the impact of the drying temperature on surface morphology and wettability. Fig. 8 Differential scanning calorimetry of AKD giving insight on the material's melting range. SEM images ( Fig. 9 ) reveal wax platelets which are not as well-developed as the ones present on surfaces shown in Fig. 6 and 7 . Since the variants of Fig. 9 were dried for a shorter period (5 instead of 10 min) and stored for a shorter time before the SEM investigation (3 days instead of 1 week), it is assumed that a longer drying time favors platelet growth. Nonetheless, WCA of the samples dried between 20–50 °C for only 5 min are clearly found in the superhydrophobic region ( Fig. 9h ). A slight increase of water-repellency with increasing drying temperature was observed. From a comparison of the insets in Fig. 9a–d it can be inferred that increasing temperature favors self-assembly, resulting in a more distinctive surface structure. A drying temperature of 60 °C, located at the onset of the materials melting range, resulted in partial melting and therefore increased distance between the remaining platelets. WCA was slightly reduced but still in the superhydrophobic range. Drying at 70 °C, a temperature beyond the DSC-derived melting range, led to full flattening of most surface parts. Only few single platelets of small size remained on the surface after exposure to 80 °C. Comparatively large wax agglomerations in the micro-range but little remaining sub-micro roughness resulted in a mean WCA of 137°. Fig. 9 SEM images of samples sprayed 4× with AKD-291 nm and dried at 20–80 °C, scale bars correspond to 50 μm and 5 μm (inset), respectively (a–g). Influence of drying temperature on the WCA after 60 s of exposure, error bars represent standard deviation (h). Surface curing A follow-up experiment to investigate effects of the time of exposure to temperature was carried out using a variant treated with four spraying cycles of AKD-291 nm. The variant was iteratively dried at 40 °C, kept in the oven for 24 h at the same temperature and thereafter stored for two days at 20 °C. SEM images presented in Fig. 10 revealed a very homogeneous, well-developed porous wax layer. No trace of large, spherical particles as was the case in earlier experiments applying shorter drying times ( Fig. 6 , 7 , and 9 ) was observed. The WCA for this specimen was 166.15°, which is the highest value achieved within the present study. Fig. 10 SEM images of a sample sprayed 4× with AKD-291 nm, dried and then kept at 40 °C. Scale bars correspond to 50 μm and 5 μm (inset), respectively. Bottom inset: Image of the maximum WCA measurement. Structure of crystalline platelets Upright platelets appear to be very efficient structures in terms of water repellence. Over the course of evolution, nature developed geometries very similar to the ones observed for AKD in the present study. The carnivorous pitcher plant ( Nepentes alata ) arranges crystalline wax platelets rectangular to the apparent surface to create superhydrophobic surfaces. 43,44 While different in terms of surface coverage and their degree of development from originally spherical particles, the AKD platelets observed in the present study were of fairly uniform geometry (lengths of about 1–5 μm and widths of approximately 100–300 nm). AKD structures on superhydrophobic paper surfaces 35 exhibited highly similar geometry, although a completely different process, i.e. the rapid expansion of supercritical CO 2 , was used. Since it is well known that crystalline regions may develop in AKD under certain conditions, 45 X-ray diffraction measurements were carried out in order to obtain additional information on the molecular structure of AKD platelets. An X-ray diffractogram of pure AKD grain melted at 103 °C and then re-solidified shows a characteristic diffraction pattern in the range of 2 θ 15–27° ( Fig. 11 ). This pattern is overlapping with the typical cellulose peak detected on measured wood samples and therefore was difficult to evaluate. 46 Furthermore, a typical diffraction peak at around 2 θ 7° arises in a region where no cellulose signal is observed. According to literature, this peak is the result of structures formed due to secondary interactions between alkenyl chains of AKD and thus indicative of crystallinity. 45 When AKD is applied to wood in molten state and then re-solidifies, no diffraction peak indicative of crystalline AKD structures is observed. By contrast, a specimen treated with aqueous AKD dispersion exhibiting surface structure as depicted in Fig. 6 , 7 , and 9 , shows a fairly distinct AKD diffraction peak at 2 θ 7°. It may thus be assumed that the specific geometry of AKD platelets is pre-determined by the formation of crystalline structures in AKD. Fig. 11 X-ray diffractograms of untreated beech wood, melted AKD on wood, pure AKD, and hierarchical AKD structure (platelets) on wood."
} | 6,509 |
34261646 | PMC8279518 | pmc | 6,887 | {
"abstract": "We 3D-print unified soft robots comprising fully integrated fluidic circuitry capable of sophisticated operations.",
"introduction": "INTRODUCTION Over the past decade, the field of soft robotics has established itself as distinctively suited for applications that would be difficult or impossible to realize using traditional, rigid robots ( 1 – 3 ). The reliance on compliant materials that are actuated by fluidic means (e.g., hydraulics and/or pneumatics) facilitates a number of inherent benefits for soft robots, particularly in terms of safety for human-robot interactions, lower costs, and adaptability in geometry for manipulating complex and/or delicate objects ( 4 – 6 ). At present, however, a critical barrier to the utility of soft robots stems from the requirement that increasing the number of independently operated soft actuators (or degrees of freedom) typically demands an equal or greater number of distinct control inputs ( 7 – 9 ). To reduce or obviate the need for such external control schemes, researchers have investigated a wide range of approaches for enhancing soft robot autonomy via fluidic logic ( 10 – 12 ). In contrast to efforts in which independent fluidic circuitry is manually connected to soft robots ( 13 , 14 ), there is growing interest in the ability to embed such functions directly inside of soft robotic systems ( 15 ). Of particular note is a hybrid strategy reported by Wehner et al. , which entailed using clean room–based multilayer soft lithography protocols to fabricate a microfluidic oscillator ( 16 ) and then harnessing a variety of manufacturing techniques—e.g., computer numerical control machining, multimaterial casting, embedded sacrificial direct ink writing, thermal curing/evacuation processes, and laser cutting—to ultimately achieve an untethered soft “octobot” capable of periodic actuations ( 17 ). Since its introduction, however, this manufacturing methodology has yet to find broad adoption in the soft robotics community ( 18 ), possibly due to the reliance on soft lithography–based fluidic circuits. Specifically, the use of multilayer soft lithography approaches for integrated fluidic circuit fabrication presents a number of challenges, including those associated with: (i) cost, time, and/or labor requirements for executing microfabrication protocols; (ii) access and training restrictions (e.g., to use clean room facilities and equipment); (iii) variability in device efficacy and/or reproducibility due to user skill–based manual alignment steps; and (iv) geometric (i.e., planar or “2.5D”) limitations inherent to photolithography and micromolding processes ( 19 ). Furthermore, while researchers have demonstrated a wide range of fluidic valving capabilities ( 16 , 20 – 22 ), enabling more sophisticated functionalities—particularly those based on pressure-gain operations—is not straightforward based on such fabrication methodologies ( 23 , 24 ). To bypass the aforementioned challenges at larger scales, Rothemund et al. introduced soft, bistable valves that serve as fluidic analogs to electronic Schmitt triggers ( 24 ). Researchers have integrated these valves with soft actuators to yield gripping, undulating, and rolling operations for soft robotic systems ( 24 – 26 ). Recently, Drotman et al. further extended the use of these valves to control the gait of an autonomous walking robot with embedded sensing operations ( 27 ). Despite these capabilities, one caveat is thesubstantial reliance on manual protocols by which the multicomponent valves and systems are assembled by hand (e.g., using glues and/or fasteners) ( 24 – 27 ), which can present similar challenges in terms of user skill–associated efficacy and reproducibility. Consequently, alternative approaches for manufacturing fluidic circuitry–based soft robots remain in critical demand ( 28 ). Here, we present a novel strategy for additively manufacturing unified soft robotic systems with fully integrated fluidic circuitry in a single print run via multimaterial “PolyJet three-dimensional (3D) printing” ( Fig. 1 ). Initially, modular components, such as fluidic circuit elements, interconnects, and access ports as well as soft robotic actuators and structural members ( Fig. 1A ), can be designed and assembled within computer-aided design (CAD) software to generate a 3D model of a soft robot with fully integrated fluidic circuitry ( Fig. 1B ). Although researchers have used a wide range of additive manufacturing technologies in the fields of soft robotics ( 29 – 31 ) and fluidic circuitry ( 32 – 36 ), we propose that PolyJet 3D printing is uniquely suitable for fabricating both classes of systems simultaneously as unified entities. PolyJet printing is an inkjet-based (“material jetting”) process in which multiple photoreactive and sacrificial support materials are dispensed in parallel (with continual ultraviolet dosing) to produce 3D objects in a line-by-line, layer-by-layer manner. Previously, researchers have reported the use of PolyJet printing for constructing soft actuators and robots (i.e., without fluidic circuitry) ( 37 – 40 ) as well as independent fluidic valves ( 41 ). In this work, however, we additively manufacture fully integrated soft robotic systems—i.e., including all of the soft actuators; body features (of arbitrary design); and fluidic circuit elements, interconnects, and ports—in a single print run ( Fig. 1, C and D , and movie S1). This process entails the concurrent printing of three distinct materials: (i) a compliant photopolymer ( Fig. 1C , black), (ii) a rigid photoplastic material ( Fig. 1C , white), and (iii) a sacrificial water-soluble support material ( Fig. 1C , yellow). One caveat to the use of the sacrificial support material is that it must be removed or dissolved from outer regions as well as internal voids and channels following the printing process ( Fig. 1E ). Researchers have demonstrated a number of techniques to minimize ( 42 , 43 ) and even bypass the support removal process completely ( 40 , 44 ); however, to promote broad accessibility, here, we use a hybrid approach that incorporates manual removal steps (e.g., on the order of tens of minutes) with autonomous dissolution protocols. In combination, the totality of the PolyJet-based additive manufacturing and postprocessing methodology—the vast majority of which being autonomous—can be executed in less than a day to realize 3D multimaterial soft robots with fully integrated fluidic circuitry (e.g., Fig. 1F ). Fig. 1 Design and additive manufacturing strategy for PolyJet 3D printing unified soft robotic systems comprising fully integrated fluidic circuitry in a single print run. ( A ) Modular 3D CAD models and analogous electronic circuit symbols of fluidic circuit elements, fluidic interconnects, soft actuators, and structural casing. ( B ) CAD model and corresponding analogous circuit diagram of a unified soft robot with a fully integrated fluidic oscillator circuit. ( C ) Conceptual illustration of multimaterial PolyJet 3D printing the soft robot using compliant (black), rigid (white), and water-soluble support (yellow) materials. ( D ) Sequential time-lapse images of the PolyJet 3D printing process. Scale bar, 5 cm; see also movie S1. ( E and F ) Fabrication results for the unified multimaterial soft robot with integrated fluidic circuitry (E) before and (F) after support material removal. Scale bars, 2 cm. Photo credits: Ruben Acevedo, University of Maryland College Park. To explore the utility of the presented strategy, first, we introduce and characterize a fundamental class of PolyJet-enabled fluidic circuit elements—namely, fluidic diodes, “normally closed” fluidic transistors, and “normally open” fluidic transistors that support facile geometric customization of their pressure-gain properties—from which sophisticated fluidic circuits can be built. We then investigate the operational performance of three soft robots with distinct integrated fluidic circuits designed with respect to fluidic analogs of conventional electrical signals, including (i) a soft robotic turtle that yields periodic, out-of-phase actuation routines for its soft limbs under constant-flow [“direct current (DC)”] input conditions; (ii) a soft robotic turtle that leverages embedded fluidic transistors with distinct pressure-gain properties to generate periodic swimming-inspired motions under sinusoidal [“alternating current (AC)”] fluidic input conditions; and (iii) a soft robotic hand capable of completing the first level of the original “Super Mario Bros.” video game in real time based on a single preprogrammed aperiodic (“variable current”) pressure input.",
"discussion": "DISCUSSION Fluidic circuitry provides powerful means to enhance soft robot autonomy and, in turn, reduce and/or eliminate the tethering requirements associated with conventional fluidic control schemes ( 47 ). In this work, we introduced the concept of additively manufacturing unified soft robotic systems with fully integrated fluidic circuitry in a single print run. To support the sophistication of the underlying fluidic circuits that govern the soft robot functionalities, we also presented and characterized a fundamental class of fluidic circuit elements that are compatible with the PolyJet 3D printing strategy, including fluidic diodes as well as normally closed and normally open fluidic transistors. Historically, it has been difficult to achieve and/or customize γ properties for conventionally manufactured fluidic operators ( 24 ). Thus, a particularly important feature of both sets of fluidic transistors reported here stems from the distinct source-to-drain and gate region diaphragms connected via an intervening piston, which allows for γ-associated behaviors to be tuned through straightforward geometric means (i.e., the ratio of the diaphragm diameters) as desired. Experimental results revealed that the fluidic circuit elements exhibited performance characteristics consistent with their electrical counterparts both independently and when integrated as part of larger integrated fluidic circuits. Although the fluidic circuits in this work were designed with respect to soft robots, it should be noted that fluidic valving and routing capabilities are widely used in chemical, biological, and biomedical fields ( 48 ), and thus, the fluidic processing approaches demonstrated here could be extended to support such applications. As the area of soft robotics is still relatively nascent, potential input modalities for driving soft robot operations remain diverse. Thus, in this work, we focused on investigating soft robots with integrated fluidic circuits based on inputs inspired by electrical signals associated with conventional robotics fields. First, we printed a soft robotic turtle capable of converting constant-flow inputs—analogous to DC electrical signals—to periodic, out-of-phase oscillations of its distinct limbs. We also designed and printed a distinct soft robotic turtle that generated periodic motions of its flippers under sinusoidal fluidic input conditions—analogous to AC electrical signals—based on an integrated fluidic circuit consisting of fluidic transistors with varying γ properties. In addition, we took advantage of such γ-enabled functionalities to create a soft robotic hand capable of hard-coded operations based primarily on the magnitude of a single, preprogrammed control input—akin to variable current electrical signals. We used a set pressure input program to autonomously control the soft robotic hand, resulting in target actuators pressing and depressing the buttons of an NES controller at specific times to complete the first level of the original Super Mario Bros. video game in real time. Although this use case served as an exemplar, the γ-based approach for controlling the fingers of the soft robotic hand could be extended to minimize the tethering requirements for emerging applications such as soft robotic gloves and rehabilitation devices ( 5 , 49 , 50 ). Furthermore, in contrast to conventional standards for evaluating soft robot capabilities that lack stringent performance metrics or associated penalties, the Super Mario Bros. demonstration is constrained with respect to externally established and invariable conditions (i.e., timing and level makeup) for which a single missed or inaccurately executed operation can result in complete failure. Given the recent concerns in the field of soft robotics regarding quantitative utility ( 51 ), we propose that future works should consider similar means of assessment that are founded on externally dictated, unyielding operational demands. Among the soft robots presented in this work, the constant flow–based soft robotic turtle is best suited to serve as a point of reference for comparing distinct methodologies for manufacturing fluidic circuit-based soft robots, as its oscillating behaviors are consistent with those often reported in the literature by other groups ( 15 , 17 , 24 – 27 ). For example, in contrast to soft lithography–based protocols previously used for fluidic circuit fabrication ( 13 , 15 – 17 ), which typically necessitate technical training and access to microfabrication equipment and clean room facilities ( 19 , 52 ), access to a PolyJet 3D printer (either directly or through a commercial 3D printing service) represents the only critical barrier to reproducing all of the soft robots and integrated fluidic circuits reported here. Although the support material removal protocols of the presented PolyJet-based strategy require a degree of manual labor (on the order of tens of minutes, e.g., movie S7), the elimination of essentially all other manual fabrication, integration, and assembly procedures associated with soft robotic actuators, structural/body features, and fluidic circuitry is a central benefit. In addition, as such manual protocols can lead to differences in performance based on user skill and training, it is expected that the automated PolyJet 3D printing process could provide advantages in terms of robot-to-robot repeatability while reducing failures caused by user error during manufacturing. It is important to note that the reported strategy could be improved further by applying recent techniques that instead use noncuring liquid-based support materials ( 40 , 44 )—with the caveat that such actions typically void the warranty of the 3D printer—to circumvent the vast majority of the support removal protocols used here. Such adaptations could, however, negatively affect the integrity of the multimaterial interfaces and/or operator-to-operator and, in turn, robot-to-robot efficacy and reproducibility. Thus, such endeavors should be undertaken with a high degree of caution. Although all of the fluidic circuit elements and soft robots were designed and fabricated as multimaterial entities, for applications that require entirely soft robots, the rigid materials could be replaced with compliant materials as desired. Similarly, the presented approach is not exclusively limited to PolyJet 3D printing and could be adapted for alternative additive technologies, such as using direct laser writing–based fluidic circuit elements ( 53 – 55 ) for soft microrobots or stereolithography-compatible soft materials ( 56 ) for meso-/macroscale systems. Nonetheless, the PolyJet-based strategy presented here offers distinctive promise to enhance accessibility within the field of soft robotics while supporting a level of reproducibility and design versatility (e.g., for the fluidic operators, integrated fluidic circuits, and the overall architectures of unified soft robotic systems) that has not been reported for alternative methodologies. Specifically, as the models for all of the fundamental fluidic circuit elements and soft robots in this work are available online (see the Supplementary Materials), researchers can readily download, modify on demand, and/or reproduce (e.g., 3D-print on site or via a commercial 3D printing service) all of the capabilities demonstrated here, thereby providing a new pathway for researchers spanning broad academic backgrounds to design, additively manufacture, and advance soft robotic systems that comprise fully integrated fluidic circuitry."
} | 4,065 |
22874123 | null | s2 | 6,889 | {
"abstract": "The computational evolution of gene networks functions like a forward genetic screen to generate, without preconceptions, all networks that can be assembled from a defined list of parts to implement a given function. Frequently networks are subject to multiple design criteria that cannot all be optimized simultaneously. To explore how these tradeoffs interact with evolution, we implement Pareto optimization in the context of gene network evolution. In response to a temporal pulse of a signal, we evolve networks whose output turns on slowly after the pulse begins, and shuts down rapidly when the pulse terminates. The best performing networks under our conditions do not fall into categories such as feed forward and negative feedback that also encode the input-output relation we used for selection. Pareto evolution can more efficiently search the space of networks than optimization based on a single ad hoc combination of the design criteria."
} | 238 |
31149818 | PMC6611076 | pmc | 6,891 | {
"abstract": "Oxygen\ndepletion in coastal waters may lead to release of toxic\nsulfide from sediments. Cable bacteria can limit sulfide release by\npromoting iron oxide formation in sediments. Currently, it is unknown\nhow widespread this phenomenon is. Here, we assess the abundance,\nactivity, and biogeochemical impact of cable bacteria at 12 Baltic\nSea sites. Cable bacteria were mostly absent in sediments overlain\nby anoxic and sulfidic bottom waters, emphasizing their dependence\non oxygen or nitrate as electron acceptors. At sites that were temporarily\nreoxygenated, cable bacterial densities were low. At seasonally hypoxic\nsites, cable bacterial densities correlated linearly with the supply\nof sulfide. The highest densities were observed at Gulf of Finland\nsites with high rates of sulfate reduction. Microelectrode profiles\nof sulfide, oxygen, and pH indicated low or no in situ cable bacteria\nactivity at all sites. Reactivation occurred within 5 days upon incubation\nof an intact sediment core from the Gulf of Finland with aerated overlying\nwater. We found no relationship between cable bacterial densities\nand macrofaunal abundances, salinity, or sediment organic carbon.\nOur geochemical data suggest that cable bacteria promote conversion\nof iron monosulfides to iron oxides in the Gulf of Finland in spring,\npossibly explaining why bottom waters in this highly eutrophic region\nrarely contain sulfide in summer.",
"introduction": "Introduction Coastal areas with\nlow oxygen (O 2 ) in bottom waters\nare expanding worldwide because of human activities. 1 , 2 Hypoxia (O 2 < 63 μM), anoxia (O 2 =\n0 μM), and the presence of toxic hydrogen sulfide (H 2 S), termed euxinia, can lead to loss of marine life and the development\nof “dead zones”. 1 Iron (Fe)\noxides and manganese (Mn) oxides in surface sediments can delay the\ndevelopment of bottom water euxinia, preventing the release of H 2 S from the sediment. 3 , 4 Recently, a group\nof multicellular filamentous bacteria was discovered, 5 which can promote the formation of Fe oxides\nand Mn oxides and can efficiently remove H 2 S from surface\nsediments. 6 − 8 These so-called cable bacteria belong to the Desulfobulbaceae family and link the oxidation of free H 2 S in deeper sediment horizons to the reduction of O 2 or nitrate (NO 3 ) by conducting electrons over centimeter\nscale distances. 5 , 9 The metabolic activity of cable\nbacteria results in the development of a suboxic zone (i.e., where\nO 2 and H 2 S are absent) and a unique pH profile, 10 characterized by a rise in pH near the sediment–water\ninterface and a strong acidification in the suboxic zone (of up to\n∼2 pH units). 5 , 11 This so-called “fingerprint\nfor cable bacteria” can be used as an indicator for their activity. 5 , 11 , 12 The strong acidification of the\npore water can lead to the dissolution of iron monosulfide (FeS) 6 , 8 and calcium (Ca), Fe and Mn carbonates. 7 When the Fe 2+ and Mn 2+ released from these\nminerals diffuse upward to the oxic zone, Fe oxides and Mn oxides\nmay form. 7 , 8 Cable bacteria can fundamentally alter the\nbiogeochemistry of coastal systems, as shown in a recent study for\na seasonally hypoxic coastal marine basin where the Fe oxides formed\nthrough their activity in spring prevented the release of H 2 S from the sediment during peak hypoxia in summer. 6 , 13 At\npresent, it is unknown whether this role of cable bacteria as “ecosystem\nengineers” can be generalized to other coastal zones experiencing\nseasonal hypoxia. Cable bacteria are tolerant to a wide range\nof salinities and temperatures,\nhave been observed in a variety of sediments, can occur in bioturbated\nsystems, and may be widespread in the seafloor. 11 , 14 Their ecological niche has been suggested to be primarily determined\nby the availability of H 2 S as an electron donor, either\nas FeS or dissolved H 2 S, and O 2 or NO 3 as an electron acceptor. 8 , 14 However, other factors\nmust also be at play because cable bacteria do not always establish\nwhen these conditions are met. Cable bacteria can co-occur with other\nsulfur oxidizing bacteria, such as Beggiatoaceae ,\nsuggesting competition for the same ecological niche. 6 , 13 Unlike cable bacteria, Beggiatoaceae are not capable\nof dissolving FeS, implying that their only source of H 2 S is from sulfate reduction. However, Beggiatoaceae are likely better adapted to low bottom water O 2 than\ncable bacteria. 15 , 16 Since field investigations\nof cable bacteria are relatively scarce\nit remains unclear how widespread their occurrence and how prevalent\ntheir role in delaying bottom water euxinia is. Here, we present field\ndata on the abundance of cable bacteria and the geochemical characteristics\nof 12 sites from contrasting depositional environments in the Baltic\nSea, as sampled in May and June 2016. We found that cable bacteria\nare most abundant in sediments of seasonally hypoxic sites with high\nrates of sulfate reduction. We infer that cable bacteria activity\nmay prevent the development of bottom water euxinia in the highly\neutrophic Gulf of Finland in summer.",
"discussion": "Results\nand Discussion Classification of Sites Based on Bottom Water\nOxygen The range in bottom water O 2 for 2014–2016\nvaried\ngreatly among the 12 sites ( Figure 2 A; Table S1 ). Bottom waters\nat GOF3 were characterized by high concentrations of O 2 (>150 μM) and were never hypoxic (<63 μM). Five\nof\nthe seasonally hypoxic sites—Arkona, LF1, 311, GOF5, and LL3A—in\ncontrast, showed a distinct seasonal cycle with high bottom water\nO 2 most of the year and relatively low O 2 in\nsummer (<63 μM for ∼60–90 days). Site JML was\nhypoxic most of the time and only briefly became oxic (in June 2015).\nBottom waters at sites LL19 and F80 were always anoxic. At the reoxygenated\nsites, the bottom water O 2 varied between 0 and 150 μM.\nAt the time of sampling, bottom water O 2 was low or absent\nat all of the seasonally hypoxic sites and the reoxygenated sites,\nexcept Arkona and LF1 ( Figure 2 B). Figure 2 (A) Range in bottom water O 2 (μM) for 2014–2016\nbased on the HELCOM database where available. The red dashed line\n(located at 63 μM) indicates the hypoxic boundary. The solid\nline between the boxes is the median, whereas the boxes represent\nthe lower and upper quartiles. The error bars indicate the minimum\nand maximum O 2 levels. (B) Bottom water O 2 concentrations\n(μM) in May/June 2016 derived from microelectrode profiles near\nthe sediment–water interface. (C) Areal density of cable bacteria\n(m cm –2 ). (D) Upward flux of sulfide toward the\nsediment−water interface/suboxic zone (mmol m –2 day –1 ). (E) Sulfate reduction rates (mmol m –2 day –1 ; black triangles represent\nthe downward flux of sulfate in mmol m –2 day –1 ). (F) Sedimentary FeS (AVS) and labile Fe(III) oxides\n(mmol m –2 ; integrated over top 2 cm; the dark and\nlight colors represent FeS and labile Fe(III) oxides respectively).\n(G) Macrofaunal abundance (ind. m –2 ). The error\nbars represent the standard deviation. (H) Bottom water salinity.\nThe black dots represent the bottom water salinity prior to the major\nBaltic inflows. (I) Sediment organic carbon (%; averaged over top\n2 cm). (J) Upward flux of ammonium (mmol m –2 day –1 ) toward the sediment−water interface. The\nstudy sites are classified based on bottom water redox conditions\nas described in Figure 1 . Abundance of Cable Bacteria\nin Baltic Sea Sediments Visual observations of the surface\nsediment onboard ship by light\nmicroscopy revealed the presence of filaments that were likely cable\nbacteria. These filaments were observed at the seasonally hypoxic\nsites GOF5 and LL3A and at the reoxygenated sites Bornholm and BY15A,\nwhile Beggiatoaceae were found at all sites ( Table 1 ). Thick microbial\nmats of Beggiatoaceae were observed only at the reoxygenated\nsites BY15 and BY15A in the Eastern Gotland basin ( Figure S1 ). A more detailed examination of the surface sediments\nusing FISH revealed that cable bacteria were present at all sites\nexcept for LL19 ( Figure S2 ). Cable bacterial\nabundances, however, varied greatly between sites and also with sediment\ndepth ( Figures S2 and S3 ). At the\noxic site GOF3, the areal density of cable bacteria was low (12 m\ncm –2 ; Figure 2 C). At four of the seasonally hypoxic sites—Arkona,\nLF1, 311, and JML—the areal density was 5 times higher (∼60\nm cm –2 ). The seasonally hypoxic sites GOF5 and LL3A,\nboth located in the Gulf of Finland, had the highest abundance of\ncable bacteria (121 and 150 m cm –2 , respectively).\nSuch densities are comparable with sediments with active cable bacteria\ncommunities as found, for example, in seasonally hypoxic Lake Grevelingen. 28 , 41 The two anoxic sites LL19 and F80 had the lowest abundance of cable\nbacteria (0 and 4 m cm –2 ), while at the reoxygenated\nsites—Bornholm, BY15, and BY15A—abundances varied (12\nand 55 m cm –2 ). The low amounts of cable bacteria\nat discontinuous depths at the reoxygenated sites are likely inactive\nremnant cells that have not yet decayed ( Figure\nS2 ). At sites BY15 and BY15A, the abundances of cable bacteria\nwere similar to those observed in reoxygenated Baltic Sea sediments\nin a recent study. 42 Evidence for\ncable bacteria activity was visible in high-resolution\ndepth profiles of pH, O 2 , and ∑H 2 S, at\nsites GOF5 and LL3A ( Figure S4 ). Sediments\nat these sites were characterized by the presence of a suboxic zone\n( Table 1 ) and a distinct\npH peak near the subsurface. However, with a minimum pH of ∼7,\nonly moderate acidification of the pore water in the suboxic zone\nwas observed when compared to the pH of ∼6.5, which is typical\nfor cable bacteria activity. 5 , 11 At the time of sampling,\nbottom water O 2 levels were extremely low at these sites\n( Table 1 ; 5–10\nμM saturation). Limited availability of O 2 typically\nresults in the collapse of the metabolic activity of cable bacteria. 16 The observed pH profiles at site GOF5 and LL3A\nthus may reflect a residue signal of cable bacteria activity shortly\nafter such a collapse. The depth profiles of pH, O 2 , and\n∑H 2 S at sites BY15 and BY15A, in contrast, resembled\nthe typical pH profiles for activity of Beggiatoaceae . 6 This pH signature is characterized\nby a low pH near the sediment surface as a result of proton formation\nand a high pH in the deeper sediment horizon caused by proton consumption\nby ∑H 2 S oxidation. 6 In summary, the highest abundances of cable bacteria were found\nat the seasonally hypoxic sites, in particular those located in the\nGulf of Finland. Furthermore, we found that the presence of cable\nbacteria does not imply that the characteristic fingerprint for cable\nbacteria activity is observed in the pore water. However, the pH,\nO 2 , and ∑H 2 S fingerprints, for sites\nwith the highest cable bacteria density (GOF5 and LL3A), did show\na strong similarity with the typical fingerprint for cable bacteria\nactivity. Controls on the Abundance and Activity of Cable Bacteria Metabolic activity of cable bacteria requires bottom water O 2 and availability of ∑H 2 S. 8 While bottom water O 2 was typically available\nat the oxic, seasonally hypoxic, and reoxygenated sites ( Figure 2 A), only the seasonally\nhypoxic sites had a high abundance of cable bacteria ( Figure 2 C). Here we assess the relationship\nbetween the abundance of cable bacteria at our 12 Baltic Sea sites\nwith ∑H 2 S supply from sulfate reduction and dissolution\nof FeS, macrofaunal abundance, salinity, sediment organic matter,\nthe rate of organic matter degradation, and potential competition\nof cable bacteria with Beggiatoaceae . Any free\n∑H 2 S that was not removed chemically or microbially\nmay have diffused into the zone were cable bacteria were active ( Figure S4 ). Diffusive ∑H 2 S fluxes\ninto the suboxic or oxic zone ( Figures 2 D, S5, and S6 ) varied greatly\nbetween sites with highest fluxes at the oxic site (GOF3; 1 mmol m –2 day –1 ), the three seasonally hypoxic\nsites in the Gulf of Finland (JML, GOF5, LL3A; 0.8–1.1 mmol\nm –2 day –1 ), and the two anoxic\nsites (LL19, F80; 0.7–1.6 mmol m –2 day –1 ). Sulfide fluxes were much lower at the other seasonally\nhypoxic sites (Arkona, LF1, 311; 0.01–0.09 mmol m –2 day –1 ) and the reoxygenated sites (Bornholm, BY15,\nBY15 A; 0.07–0.4 mmol m –2 day –1 ). Strikingly, the abundance of cable bacteria depends linearly on\nthe diffusive supply of ∑H 2 S, at all seasonally\nhypoxic sites, except JML ( Figure 3 A). Data for sediments populated by cable bacteria,\nin other systems, such as Lake Grevelingen (March 2012), 6 Wadden Sea (Mussel Reef; June 2013), 41 and a mangrove, 28 follow the same general trend ( Figure 3 A). In contrast, our anoxic and reoxygenated\nsites and the seasonally hypoxic site JML do not show such a correlation,\nlikely because O 2 was limiting during most of the year.\nWe conclude that, when sufficient bottom water O 2 is available\nduring a large part of the year, the supply of ∑H 2 S acts as the main control on the abundance of cable bacteria. Figure 3 (A) Linear\ncorrelation between diffusive supply of ∑H 2 S (mmol\nm –2 day –1 ) and\nareal density of cable bacteria (m cm –2 ). The toned\ndown sites in the background are omitted from the linear correlation,\nbecause other factors, such as insufficient bottom water O 2 , controlled the abundance of cable bacteria. (B) Relationship between\nmeasured sulfate reduction rates (mmol m –2 day –1 ) and areal density of cable bacteria (m cm –2 ). GOF3 is omitted from the plot, because of the exceptionally high\nsulfate reduction rate. The sulfate reduction rate at BY15A is not\navailable. The sample sites are classified based on the bottom water\nredox conditions as described in Figure 1 . Sulfate reduction can provide a key supply of sulfide to\ncable\nbacteria. 41 We found a large variation\nin rates at our 12 sites ( Figures 2 E, S5, and S7 ), with the\nhighest rate observed at the oxic site GOF3 (21.4 mmol m –2 day –1 ). Of the other sites, the seasonally hypoxic\nsites GOF5 and LL3A had the highest sulfate reduction rates (2.1 and\n2.5 mmol m –2 day –1 , respectively).\nComparison of the sulfate reduction rates and diffusive fluxes for\nGOF5 and LL3A shows that a significant proportion of the ∑H 2 S that was produced diffused to the suboxic or oxic zone.\nWhen we plot the areal density of cable bacteria against the sulfate\nreduction rates, no direct relation can be found. However, three groups\ncan be identified ( Figure 3 B) which consist of (1) sites with a high areal density of\ncable bacteria in concert with high sulfate reduction rates and seasonal\nbottom water O 2 availability; (2) sites with sufficient\nbottom water O 2 during a large part of the year, but relatively\nlow sulfate reduction rates and ∑H 2 S availability;\nand (3) sites with a low areal density of cable bacteria and low bottom\nwater O 2 availability during the year. At the two seasonally\nhypoxic sites GOF5 and LL3A in group 1, sulfate reduction rates were\nhigh (2.1 and 2.5 mmol m –2 day –1 , respectively) close to the sediment surface ( Figure S5 ). Such conditions likely allow cable bacteria to\naccess electrons released from ∑H 2 S by sulfate reduction\nclose to the oxic zone, stimulating a rapid and dense growth. 41 Iron monosulfide can also serves as a\nsource of ∑H 2 S for cable bacteria, with cable bacteria\ninducing the dissolution\nof FeS by pore water acidification. The Fe 2+ released by\ndissolution can subsequently be oxidized upon contact with O 2 to form Fe oxides. 8 , 12 When bottom water O 2 levels are low again, the Fe oxides are converted back to FeS upon\ncontact with ∑H 2 S. 13 Because\nof this “pool switching” mechanism and the fact that\nthe sediment FeS concentrations observed in June 2016 represented\none time point at the beginning of summer, we consider the sum of\nthe FeS and Fe oxides as a more representative term for the total\npool of potentially available FeS for cable bacteria (hereafter FeS\n+ FeOx) at the end of summer . The lowest FeS + FeOx pools\nwere found at both anoxic sites and the two reoxygenated sites BY15\nand BY15A (0.05 and 0.08 mol m –2 ; Figure 2 F). The highest FeS + FeOx\nwas found at site LL3A (0.7 mol m –2 ). Most of the\nFeS + FeOx (∼71%) at site GOF5 consisted of Fe oxides, likely\nas a result of cable bacteria activity prior to our sampling campaign\n( Figure 2 F). In contrast,\nat site LL3A, ∼99% of the FeS + FeOx consisted of FeS, likely\nbecause Fe oxides formed by cable bacteria already underwent conversion\nto FeS because of the low bottom water O 2 concentrations\nat the time of sampling. This pool of FeS, if dissolved in spring,\nfor example, in 100 days, is equivalent to a ∑H 2 S supply of 7 mmol –2 day –1 . Given\nthe sulfate reduction rate of 2.5 mmol m –2 day –1 at LL3A, this indicates that sediment FeS could be\na major source of ∑H 2 S for cable bacteria in this\nregion. In summary, the sites with the highest abundance of cable\nbacteria were characterized by a large FeS + FeOx pool, relatively\nhigh rates of sulfate reduction and upward fluxes of ∑H 2 S, and large seasonal variations in bottom water O 2 with high concentrations in winter and spring ( Figures 2 and 3 ). We conclude that both sulfate reduction and the dissolution of\nFeS act as a source of ∑H 2 S for cable bacteria in\nthe Gulf of Finland. Bioturbation can inhibit the development\nof cable bacteria by damaging\nthe bacterial filaments, causing a disruption of the electrochemical\nsignal. 11 Previously, high numbers of polychaetes\nbelonging to the genus Marenzelleria were observed\nat the oxic site GOF3. 43 Visual observations\nduring slicing of the cores for pore water collection also suggest\ndisturbance by macrofauna at this site. Polychaetes also dominated\nthe macrofaunal community at Arkona, GOF5, LL3A, and Bornholm, whereas\nbivalves were most abundant at LF1 and 311 ( Table 2 and Figures S9–S11 ). The genera of polychaetes (ind. m –2 ) differed\namong these four sites ( Figure S10A ). The\nthree major polychaetes observed at Arkona were Scoloplos , Bylgides , and Terebellides . At\nsites GOF5 and LL3A, Marenzelleria dominated, whereas Scoloplos were the most prevalent polychaetes at Bornholm.\nIn terms of macrofaunal biomass (AFDM), site GOF5 and LL3A had the\nhighest biomass of polychaetes (1.8 and 2.2 g m –2 , respectively; Figure S10B ). The biomass\nof polychaetes at Arkona was comparatively low (0.2 g m –2 ), even though the absolute number of polychaetes at Arkona was similar\nto that of GOF5 and LL3A ( Table 2 ). Macrofauna were absent at JML, LL19, and F80 ( Figure 2 G). In general, macrofaunal\nabundances at our 12 sites were relatively low (<1700 ind. m –2 ; Figure 2 G) when compared to other temperate coastal systems, such\nas, for example the Western Black Sea shelf (6 000–10 000\nind. m –2 ) 44 , 45 and the North Sea (2 400–21 000\nind. m –2 ). 46 The relatively\nimpoverished macrofaunal community in the Baltic Sea is a consequence\nof its natural constraints by brackish conditions that limit macrofaunal\nabundance and diversity and the recent human-induced increased prevalence\nof hypoxia. 47 We did not find a relationship\nbetween macrofaunal abundances (expressed as ind. m –2 and AFDM g m –2 ) and cable bacterial abundances\n( Figure S12A,B ), suggesting that at most\nsites the impoverished macrofaunal communities did not significantly\nhamper growth of cable bacteria. Interestingly, the areal density\nof cable bacteria and biomass (AFDM) of polychaetes was highest at\nsites GOF5 and LL3A ( Figure S12C ). This\nfurther indicates that the polychaetes did not disturb the growth\nof cable bacteria. However, at our oxic site GOF3, which was characterized\nby high bottom water O 2 and a high diffusive supply of\n∑H 2 S, the low areal density of cable bacteria was\nlikely due to intense bioturbation by Marenzelleria . A similar low areal density of cable bacteria was observed in a\nrecent study at a permanently oxic site in the Eastern Gotland Basin\nand was also attributed to the presence of Marenzelleria . 42 Table 2 Macrofaunal Individuals\nper Square\nMeter for Six Taxonomic Classes in May and June 2016 site Polychaeta Sipunculida Oligochaeta Bivalvia Malacostraca Insecta total GOF3 N/A N/A N/A N/A N/A N/A N/A Arkona 1111 ± 393 0 69 ± 62 306 ± 243 139 ± 124 69 ± 76 1694 ± 593 LF1 23 ± 33 139 ± 57 0 463 ± 87 0 0 625 ± 113 311 23 ± 33 81 ± 48 0 185 ± 65 0 12 ± 26 301 ± 104 GOF5 1100 ± 225 12 ± 26 0 58 ± 129 12 ± 26 46 ± 33 1227 ± 249 LL3A 995 ± 741 0 0 0 0 12 ± 26 1007 ± 734 JML 0 0 0 0 0 0 0 LL19 0 0 0 0 0 0 0 F80 0 0 0 0 0 0 0 Bornholm 1278 ± 717 0 0 0 0 83 ± 28 1361 ± 705 BY15 0 0 0 0 0 0 0 BY15A 0 0 0 0 0 0 0 Cable bacteria can tolerate a wide range of salinities, because\nthey occur in marine, brackish, and fresh water environments. 48 Bottom water salinity varied widely among our\n12 sites, from 7.5 to 18.5, but showed no relationship with the abundance\nof cable bacteria ( Figures 2 H and S12D ). Bottom water salinities\nat our study sites were slightly higher than under normal circumstances\nbecause of recent inflows of saline water from the North Sea ( Figure 2 H). Organic carbon\nin the upper 2 cm of the surface sediment varied from 1 to 13 wt %\n( Table 1 ), and again,\nthere was no relationship with the abundance of cable bacteria ( Figures 2 I and S12E ). The rate of anaerobic degradation of organic\nmatter, here approximated by the ammonium flux toward the sediment–water\ninterface ( Figures 2 J and S8 ) showed the same linear trend\nwith cable bacterial abundances as ∑H 2 S at most\nseasonally hypoxic sites ( Figure S12F ). The absence of a correlation between the abundance of cable bacteria,\nmacrofaunal abundances, salinity, and sediment organic carbon ( Figures 2 and S12 ) highlights that in sediments with no to moderate\ndisturbance by bioturbation, the availability of O 2 and\n∑H 2 S are the key controls that determine the abundance\nof cable bacteria. While at the oxic site GOF3 both O 2 and\n∑H 2 S were abundantly present, cable bacteria were\nlikely inhibited by high macrofaunal activity. At the anoxic sites\nLL19 and F80, there was insufficient O 2 . At the reoxygenated\nsites Bornholm, BY15, and BY15A, O 2 concentrations likely\nremained too low for cable bacteria. Beggiatoaceae were abundantly present as thick mats, and because they are better\nadapted to such low bottom O 2 conditions, they likely outcompeted\nthe cable bacteria. This is supported by our observation that reoxygenation\nof the overlying water for sediment from site BY15A in the laboratory\ndid not result in cable bacteria activity ( Figure\nS13A ). The seasonally hypoxic sites provided the best\nconditions for cable\nbacteria, with the variation in abundance between sites explained\nby the diffusive supply of ∑H 2 S ( Figure 3 A). These sites had a high\nO 2 availability during a major part of the year in concert\nwith a high diffusive supply of ∑H 2 S and a large\nFeS + FeOx pool ( Figure 2 F). Because the bottom water O 2 levels at both sites were\nextremely low at the time of sampling (∼5–10 μM; Table 1 ), cable bacteria\nwere not very active. Strikingly, the activity of cable bacteria in\na sediment core retrieved from GOF5 could be stimulated rapidly (within\n∼5 days) upon reoxygenation of the overlying bottom water ( Figure S13B ). In summary, this indicates\nthree requirements for a high abundance\nof cable bacteria: (1) high bottom water O 2 availability\nduring a major part of the year; (2) availability of ΣH 2 S; (3) no to moderate disturbance by macrofaunal bioturbation.\nSuch conditions are found at our study sites in the Gulf of Finland.\nOther factors, such as the availability and degradation of organic\nmatter and bottom water salinity (within the salinity range at our\nsites, i.e. 7.5–18.5) are of less importance. The strong dependence\nof cable bacteria on the availability of ∑H 2 S is\ntypical for chemoautotrophic sulfur-oxidizing bacteria (e.g., Nelson\nand Jannasch 49 ). However, recent work on\ncable bacteria indicates that they are likely heterotrophs. 50 Apparently, the requirements of cable bacteria\nfor organic carbon are relatively easily met, allowing the availability\nof ΣH 2 S to become a key control. Biogeochemical\nImpact of Cable Bacteria Cable bacteria\nactivity can strongly impact sedimentary Fe and S cycling, 6 , 7 , 13 because the acidification of\nthe pore water (pH ∼6.5; Figure S13B ) 5 , 11 can facilitate the dissolution of FeS and promote\nthe formation of Fe oxides. 8 , 12 The FeS + FeOx pool\nin the upper part of the sediment ( Figure 2 F) provides insight into the maximum amount\nof Fe oxides that can be formed on a seasonal basis. This amount of\nFe oxides ultimately controls how much ∑H 2 S can\nbe sequestered before free ∑H 2 S is released into\nthe water column during bottom water anoxia. In the Gulf of\nFinland at site GOF5, the metabolic activity of cable bacteria and\nassociated pore water acidification in spring likely contributed to\nthe formation of the Fe oxides (0.17 mol m –2 ) observed\nin the surface sediment in June 2016 ( Figures 2 F and 4 A). The estimated\ndepletion of FeS at GOF5 over the first 2.5 cm is 0.17 mol m –2 ( Figure 4 B). Assuming\nthat all Fe 2+ released upon dissolution of FeS would precipitate\nas Fe oxides upon contact with O 2 , this would give an increase\nin Fe oxides of 0.17 mol m –2 , which is in line with\nour observations. The abundant presence of Fe oxides in surface sediments\nhas previously been shown to delay euxinia in seasonally hypoxic Lake\nGrevelingen. 6 At our site GOF5, the sulfate\nreduction rate was 2 mmol m –2 day –1 . This would imply that if the activity of the cable bacteria ceases\nbecause of the onset of hypoxia, the Fe oxide layer can temporarily\ndelay the escape of ΣH 2 S from the sediment for a\nperiod of ∼85 days. Figure 4 Solid-phase profiles of (A) labile Fe(III) oxides\n(FeOx) and (B)\nFeS (AVS) for GOF5 in June 2016. An alternative mechanism for the development of an Fe oxide\nlayer\ncould be bioirrigation by Marenzelleria . These Marenzellaria are capable of pumping O 2 into\npore waters, thereby enhancing the oxidation of reduced Fe. 51 However, the number of Marenzellaria that was observed at GOF5 was relatively moderate (1100 ±\n225 ind. m –2 ; Table 2 ), and results of a reactive transport model for a\nsimilar site suggest that much higher population densities >3000\nind.\nm –2 are required to have a significant effect on\nthe formation of Fe oxides. 51 Bioirrigation\nby Marenzelleria also typically leads to oxidation\nof the sediment down to a depth of several centimeters. 51 Because we observed Fe oxides only within the\ntop 1 cm of the surface sediment at site GOF5, this confirms that\nthe role of Marenzelleria with respect to oxidation\nof the sediment by bioirrigation was likely negligible. The insignificant\nimpact of Marenzelleria on biorrigation is further\nsupported by bromide incubations performed on intact sediment cores\nretrieved from site LF1 and Arkona ( Supporting\nInformation 1.10; Figure S14 ). These incubations indicate very\nlow rates of bioirrigation at site LF1 and Arkona, despite the presence\nof polychaetes ( Table 2 ). Water column monitoring data indicate that the bottom waters\nin\nthe Gulf of Finland are rarely euxinic ( Figure 2 A). We suggest that cable bacteria are responsible\nfor the absence of ∑H 2 S in the bottom water of the\nGulf of Finland in summer, by inducing the formation of strong surface\nenrichments of Fe oxides in winter and/or spring. The Gulf of Finland\nis only the second system for which this Fe oxide buffer mechanism\nhas been suggested, after Lake Grevelingen, and the first with a relatively\nlow bottom water salinity (∼9–11 versus ∼32). 52 The Fe oxide buffer mechanism induced by cable\nbacteria is likely of importance in many other eutrophic, brackish\ncoastal areas characterized by moderate of disturbance by bioturbation,\nhigh bottom water O 2 and a high sediment supply of ∑H 2 S."
} | 6,994 |
28070216 | PMC5217482 | pmc | 6,892 | {
"abstract": "Permeable sediments are common across continental shelves and are critical contributors to marine biogeochemical cycling. Organic matter in permeable sediments is dominated by microalgae, which as eukaryotes have different anaerobic metabolic pathways to prokaryotes such as bacteria and archaea. Here we present analyses of flow-through reactor experiments showing that dissolved inorganic carbon is produced predominantly as a result of anaerobic eukaryotic metabolic activity. In our experiments, anaerobic production of dissolved inorganic carbon was consistently accompanied by large dissolved H 2 production rates, suggesting the presence of fermentation. The production of both dissolved inorganic carbon and H 2 persisted following administration of broad spectrum bactericidal antibiotics, but ceased following treatment with metronidazole. Metronidazole inhibits the ferredoxin/hydrogenase pathway of fermentative eukaryotic H 2 production, suggesting that pathway as the source of H 2 and dissolved inorganic carbon production. Metabolomic analysis showed large increases in lipid production at the onset of anoxia, consistent with documented pathways of anoxic dark fermentation in microalgae. Cell counts revealed a predominance of microalgae in the sediments. H 2 production was observed in dark anoxic cultures of diatoms ( Fragilariopsis sp.) and a chlorophyte ( Pyramimonas ) isolated from the study site, substantiating the hypothesis that microalgae undertake fermentation. We conclude that microalgal dark fermentation could be an important energy-conserving pathway in permeable sediments."
} | 404 |
33125242 | PMC7667636 | pmc | 6,894 | {
"abstract": "We\nreport a dissociative electron attachment study to 2-furoic\nacid (C 5 H 4 O 3 ) isolated in a gas phase,\nwhich is a model molecule consisting of a carboxylic group and a furan\nring. Dissociation of furan by low energy electrons is accessible\nonly via electronic excited Feshbach resonances at energies of incident\nelectrons above 5 eV. On the other hand, carboxylic acids are well-known\nto dissociate via attachment of electrons at subexcitation energies.\nHere we elucidate how the electron and proton transfer reactions induced\nby carboxylation influence stability of the furan ring. Overlap of\nthe furan and carboxyl π orbitals results in transformation\nof the nondissociative π 2 resonance of the furan\nring to a dissociative resonance. The interpretation of hydrogen transfer\nreactions is supported by experimental studies of 3-methyl-2-furoic\nand 5-methyl-2-furoic acids (C 6 H 6 O 3 ) and density functional theory (DFT) calculations.",
"conclusion": "5 Conclusions In the\npresent work, we have measured dissociative electron attachment\nin 2-furoic acid, 3-methyl-2-furoic acid, and 5-methyl-2-furoic acid.\nA rich fragmentation pattern has been revealed, with the strongest\ndissociation channel for all three compounds coming from the cleavage\nof the hydroxyl bond and production of the (M–H) − anion at electron energies around 1.2 eV. At higher energies (M–H) − anions are formed via core excited Feshbach resonances\nthat are not detected for 3M-2FA. The behavior can be explained by\nexclusive formation of the (M–H) − only from\nthe 3 position of the molecule or low stability of the 3M-2FA dehydrogenated\nanion at these energies, which undergoes further fragmentation. Assignment of individual core excited Feshbach resonances to individual\nRydberg states of the molecule based on PES indicate that the nondissociative\nπ –1 s 2 resonance of furan becomes\ndissociative in 2FA upon overlapping the orbital with that of the\ncarbonyl group. This is important with respect to electron transfer\nreactions after irradiation of complex molecular systems. DFT\ncalculations of energetic thresholds for individual DEA reaction\nchannels of 2FA are in good agreement with a direct bond cleavage\nmechanism for most of the observed fragments. However, complex rearrangement\nand fragmentation of the neutral cofragments may also occur during\nthe DEA. Experiments with methyl-substituted molecules help us to\nbetter identify the most probable reaction pathways, including hydrogen\nmigration.",
"introduction": "1 Introduction Electron scattering by\nfuran-containing molecules has attracted\nsignificant attention in recent years (e.g., refs ( 1 − 9 )). This is caused by a more general interest in electron interactions\nwith biomolecular systems due to their involvement in the radiation\nchemistry of living tissues. 10 − 13 The molecular structure of furan can be considered\nas a building block of biomolecules, particularly sugar and more complex\nstructures such as nucleic acids. Carboxylic acids are studied as\na model components of complex biomolecules as well, 14 − 16 particularly\nas important prototypical systems for hydrogen bonds. 17 It has been proposed theoretically 18 , 19 and demonstrated experimentally for the formic acid, 20 oxalic acid, 21 or\npyruvic acid 22 dimers that the formation\nof a double hydrogen bond via carboxylic groups can induce ultrafast\nproton transfer between two bonded monomers. The double hydrogen bond\nstructure of dimer has also been observed for 2-furoic acid 23 − 25 (2FA)—a molecule studied here, which possesses both a heterocyclic\nring and a carboxylic group. Substitution on a furan ring can\nsignificantly affect its ability\nto attach low energy electrons as well as its susceptibility to fragment.\nAlready the first studies of the electron attachment to furan, tetrahydrofuran,\nand fructose 1 demonstrated that fructose\nis a much better electron scavenger than bare or hydrogenated furan.\nA large (>10 × 10 –20 m 2 ) cross\nsection\nfor electron attachment to deoxyribose was then reported by Ptasińska\net al. 26 Another example is the severe\ndamage caused to the furan ring after electron transfer from cytidine\nin a complex nucleoside, deoxycytidine monophosphate. 27 Substitution of furan hydrogens by electronegative fluorine\natoms not only enhances its ability to attach low energy electrons\nat subexcitation energies, 28 but also enables\nbreakage of the ring by these species. 29 Moreover, several examples of chemical intramolecular mechanisms\nof hydrogen migration in the charged state of the furan and its substituents\nwere observed. 30 , 31 Finally, substitution can affect\nthe chemical and biochemical interaction of the ring structures. 32 , 33 It is therefore interesting to study how substituion of the ring\ninfluences the interaction with low energy electrons, particularly\nfor 2FA with several important applications. 2FA, C 5 H 4 O 3 , is also known as\npyromucic acid or α-furancarboxylic acid or furan-2-carboxylic\nacid. 2FA can be synthesized by the oxidation of either furfuryl alcohol\nor furfural. 34 An industrial way of achieving\n2FA involves the Cannizzaro disproportionation reaction of furfural\nin an aqueous sodium hydroxide solution. 35 , 36 The main industrial applications of 2FA include pharmaceuticals,\nagrochemicals, fragrances, and flavors. 37 − 39 Due to its highly nonlinear\noptical properties, 2FA finds its application in optic technologies. 40 , 41 Of particular importance for the present study is a proposed antitumor\nactivity of 2FA. 42 Electron affinic antitumor\ndrugs exhibit strong synergistic effects with radiation, which have\nbeen subscribed to the interaction with secondary low energy electrons. 43 Presently, in the literature, there is\na paucity of experimental\nelectron scattering studies with 2FA, likely because of its low volatility\nand sensitivity to air, complicating its handling and sublimation\nfor gas phase studies. The use of a direct insertion probe technique\nfor sample sublimation in the present experiment significantly improves\nthe handling of 2FA and may be advantageous also for other studies\nof low volatility samples. In addition to 2FA measurements we\ninvestigated 3-methyl-2-furoic\nacid (3M-2FA) and 5-methyl-2-furoic acid (5M-2FA) molecules. These\nmeasurements helped us to better understand the mechanism of 2FA dissociation\nupon electron attachment.",
"discussion": "4 Results and Discussion Theoretically, we explored four minimum energy conformers for 2FA.\nFirst, the carboxylic moiety can be oriented in two ways with respect\nto the furan ring, and second, the hydrogen atom in the carboxylic\nacid can be oriented either toward or away from the ring. Graphical\nrepresentations of these conformers and calculated relative energy\nlevels are shown in Table 1 . Choosing the conformer does not affect much the position\nof the theoretical threshold energy. For the purpose of our analysis\nit is not critical to choose which energy conformer we use in the\ncalculations. The calculated energy thresholds between two, most energetically\nseparated conformers introduce the difference only by ∼230–280\nmeV, depending on the basis set. For DEA threshold calculations, we\nuse the lowest energy conformer of 2FA. Table 1 Relative\nElectronic Energies (in kJ\nmol –1 ) of 2FA Acid Conformers Compared with Available\nTheoretical Results from Halasa et al. 55 The DEA fragmentation processes\nto 2FA lead to a large number of\nanionic products. Generally, DEA is a two-step process that can be\ndescribed by the following equation: 2 where\nwe can distinguish creation of a transient\nparent anion M #– of a target molecule M and then\nfragmentation into a detected negative fragment ion M a – and\nan undetected neutral fragment M b . In the following sections,\nwe will describe the mechanisms of partial dissociation channels of\nthe 2FA after electron attachment. 4.1 Hydrogen Loss Channel The dominant\nfragmentation channel results in the formation of the stable, closed\nshell anion, (M–H) − , by the loss of a neutral\nhydrogen atom: 3 The DEA reaction resulting in hydrogen\nloss from a molecule itself 56 , 57 or from the hydroxyl\nmoiety of a carboxylic group is a well-known process. 3 , 22 , 58 In the present case, it occurs\nprimarily at an electron energy of ∼1.2 eV as demonstrated\nin the energy dependent ion yields in Figure 2 . In 2FA, there are four hydrogen atoms which\nmay be cleaved from the parent ion. The threshold energy calculations\nin Table 3 confirm\nthat only the hydrogen in the carboxylic acid group fulfills the energy\nonset of the low energy resonance. Figure 2 2FA molecule model with depicted carbon\nnumbers (top); ion yields\nfor the most prominent DEA fragments of 2FA, 3M-2FA, and 5M-2FA (bottom). Cleavage of hydrogen atoms from the furan ring\ncan be achieved\nonly by significant energy input of above 2.5 eV. Such fragmentation\ncan occur via core excited Feshbach resonances observed as a broader\nand overlapping structures in the (M–H) − anion\nyield from 5 to 11 eV ( Figure 2 ). Based on the calculations, cleavage of any of the hydrogen\natoms of the furan ring is possible at these energies. However, our\nsupporting experiments with methyl-substituted compounds indicate\nthat only the hydrogen from position 3 is cleaved. We can see that\nFeshbach resonances are well visible in the spectra of 2FA and 5M-2FA\n( Figure 2 a,c), but\nthey disappear in the spectrum of 3M-2FA ( Figure 2 b), where the hydrogen in position 3 is replaced\nwith a methyl group. The differences in computed energies for\n2FA indicate that hydrogen\nloss from position 3 may be favored by ≈0.5 eV in comparison\nto the other ones. While the energy may explain the preference for\nthe channel in the case of ergodic dissociation of the transient anion,\nit does not explain why other channels are closed. An alternative\nexplanation may be that the (M–H) − anion\nof 3M-2FA, at these higher energies, is metastable and undergoes dissociation\nto smaller anionic fragments as will be discussed in sections 4.2 and 4.3 . 4.2 Electronically Excited Feshbach Resonances In Figure 3 , we\npresent the ion yield curves for DEA channels of the three targets\ntogether, in decreasing intensity order. In Figure 3 , the values of the calculated threshold\nenergies for the individual dissociation channels are included to\nenable comparisons with the theoretical evaluations of the dissociation\nthresholds. Table 3 presents the threshold energies calculated for several possible\nreaction pathways for each detected DEA channel. The uncertainty in\nthe identification of the reaction pathways is caused by the fact\nthat the energy threshold depends on the neutral fragmentation products.\nHence, finding an agreement with the experimental signal onset requires\ncalculation of several reaction pathways and often an inclusion of\ncomplex rearrangement reactions. Figure 3 Ion yields for individual DEA fragments\nof 2FA (red), 3M-2FA (semitransparent\ngreen), and 5M-2FA (semitransparent black). The calculated threshold\nenergies (from Tables 3 and 4 ) for individual dissociation channels\nrefer to 2FA. The blue curve in panel a is a photoelectron spectrum\nfrom ref ( 59 ) shifted\nby 4 eV. The blue transparent vertical stripes are to enhance the\nrelation between the PES and the DEA spectra. The observed anionic fragments, with the exception of (M–H) − (1.2 eV resonance) and C 2 H – (discussion in section 4.3 ), are formed above 5 eV. In Figure 3 , we can distinguish three dominant energy\nbands, centered around 5.3, 6.7, and 9.4 eV, which correspond to core-excited\nresonances. These can fall into two main classes: the Feshbach resonances\nwith a cationic core and two electrons in the s 2 Rydberg\norbital or core-excited shape resonances (possibly supported by a\ndipole moment of the parent excited state). The first class has been\nshown to be operative in a broad range of molecules; 60 − 62 the second class has been shown to be operative in selected organic\nmolecules. 63 In the case of simple amides,\nthe question of which mechanism is operative sparked a recent discussion. 64 − 66 A useful tool to identify the contribution of the Feshbach resonances\nis to compare the DEA spectra with an available photoelectron spectrum\n(PES) and semiempirical molecular orbital (MO) calculations for 2FA.\nThe relationships between energies available from photoelectron and\nelectron attachment spectroscopies is well-known. The spectra are\ntypically shifted by a constant energy of a few electronvolts. 2 , 22 , 67 − 70 The PES data reveals three distinct\nbands in the region of 9–13 eV. We use the same MO notation\nas Klapstein et al. 59 to denote these bands.\nThe first band in the PES, located at 9.32 eV, is related to ionization\nfrom the ring π 3 orbital. The second, composed of\ntwo overlapping bands, the ring π 2 orbital and the n O ′ band of the carbonyl group, appears at 10.74 eV. The third band,\ndue to out-of-plane lone-pair orbital on the hydroxyl oxygen, n O ″ , is located at 11.90 eV. The next higher, unassigned, band is visible\nin the experimental photoelectron spectrum at 13.5 eV. These bands\nwere predicted by MO calculation using the modified neglect of diatomic\noverlap (MNDO) method, and by the HAM/3 method. 59 Here we focus only on comparisons to the experimental PES.\nWhen we collate the DEA bands above 5 eV with the PES, we can assign\nbands as in Table 2 . Table 2 Assignment of the Bands from the PES\nSpectrum to DEA 2FA ; DEA FN Band Locations from\nFuran a DEA 2FA → PES (assignment) DEA FN 5.3 → 9.32 (π 3 ) 6 6.7 → 10.74 ( n O ′ , π 2 ) 9.4 → 13.50 10.5 a All data in electronvolts. The energy differences between corresponding DEA and\nPES bands\ncan be estimated to ∼4 eV. Typically, previous reports found\nthe energy shift in the range 3.3–4.5 eV depending on the molecular\ntarget. 2 , 22 , 60 The PES band\nassigned to the n O ″ band is not visible in the DEA spectrum;\nhowever, it may contribute to the adjacent ion yield signals, making\nthem broader. Using the same energy shift, as for other bands, we\ncan estimate its position to be at ∼7.9 eV in the DEA spectrum.\nThe correlation with the photoelectron spectrum indeed suggests that\nthe observed DEA bands are mediated by the formation of Feshbach resonances\nwith two electrons in the same Rydberg orbitals. It should be noted\nthat for a molecule of this size with a complicated excited structure\nother core-excited processes can be operative at overlapping energies. We can compare the present results with those of furan (FN) 1 to gain more insight. FN fragments into three\npronounced DEA channels of C 2 HO – ( m / z = 41), (FN–H) − ( m / z = 67), and C 3 H 3 – ( m / z = 39) fragments, in decreasing intensity\norder. The present work allows us to observe the quantitative agreement\nin theses anionic fragments between FN and 2FA; the same ion yield\nintensity order can be observed in Figure 3 , panels c, d, and g, respectively. Concerning\nthe energetics, the FN ion yields band peaks at 6 and 10.5 eV. The\nfirst band is sharp and intense, and the latter is less intense and\nbroader. Such bands are observed also in the spectrum of 2FA, slightly\nshifted toward lower energies and observed as 5.3 and 9.4 eV bands,\nrespectively (see Table 2 ). However, we can see in Figure 3 that, except for the m / z = 67 fragment corresponding to the loss of carboxyl group from 2FA,\nthe intensity of the first 5.3 eV band is negligible. For all ions,\na new and intense band appears in the spectrum of 2FA (in Figure 3 ) at an energy of\n6.7 eV, which corresponds to the overlapping π 2 orbital\nof the ring and carbonyl oxygen as discussed above. We can see that\ninitially bound excited state corresponding to electron removal from\nthe π 2 orbital in FN is transformed to dissociative\nstate after carboxylation in 2FA. In addition to the fragments,\nwhich were previously observed for\nfuran, we also observe new fragments in 2FA. The most intense is OH – ( m / z 17) resulting\nfrom dissociation of the carboxylic group. Another fragment that can\nbe directly assigned to the carboxylic group is COOH – . However, this fragment was observed with low intensity at the detection\nlimit of our setup, so the spectrum is not shown here. Even though\none can expect formation of these ions at energies typical for DEA\nto formic acid, 46 this is not the case.\nThe main Feshbach resonance in formic acid peaks at 7.6 eV 71 − 73 corresponding to the electron removal from the n O ″ orbital\nof the hydroxyl oxygen. The OH – and COOH – fragments in the present case, however, peak at 6 and 9 eV, where\nbare HCOOH dissociates with only low yields. 73 Either such electronic states localized on HCOOH group become dissociative\nin 2FA or the entrance states are that of the furan ring. Correct\nassignment of these bands will require high level ab initio calculations. Substitution of the carboxylic group then also\nopens several new\nfragmentation channels of the furan ring. The most intense of these\nchannels results in the formation of C 2 H – ( m / z 25) followed by C 3 H 3 O – ( m / z 55) and C 4 H – ( m / z 49). Most of these fragmentation channels can be described\nby a simple reaction mechanism as shown in Table 3 . Table 3 B3LYP/aug-cc-pVTZ Threshold Energies\nfor Individual DEA Fragmentation Channels DEA to 3M-2FA and 5M-2FA results in similar anionic\nproducts. When\ncomparing the yields for 3- and 5-methylated furoic acid with that\nof 2FA, we can see similar resonance trends, both quantitatively and\nqualitatively. Increase of the signal intensity in the case of m / z 39 may be assigned to competition of\nthis fragmentation channel with that of m / z 25, and a similar increase in the case of m / z 55 may be assigned to competition with the channel\nresulting in the formation of m / z 41 anion. The COOH loss, m / z 67\nfor 2FA, corresponds to m / z 81 anions\nobserved for methylated molecules. However, substitutions cause significant\nchanges in the signal of m / z 67\nand 25 anions, which will be discussed in section 4.3 . 4.3 Hydrogen Migration The anionic channel\nleading to the formation of C 2 H – from\n2FA shows two distinctive resonance bands in the ion yield spectrum\npeaking at ∼5 eV and around 9 eV. The higher energy band was\npreviously assigned as a Feshbach resonance. Fragmentation resulting\nin a stable anion and two neutral fragments predict energy thresholds\nof 8.15, 7.77, and 7.49 eV, for reactions 1, 2, and 3 in Table 4 . These are good estimates for this band. However, the experimental\nonset for the lower band appears at ≈4 eV. Table 4 Possible Dissociation Channels for\nC 2 H – DEA Channel of the 2FA Molecule a a Dissociation\nschemes 1, 2, and\n3 depict elementary fragmentation of the molecule, whereas more complex\nreaction channels where hydrogen migrates from/to carboxyl group are\npresented in reactions 2a, 3a, and 3b or within the ring, reactions\n4 and 5. The threshold energies are calculated using B3LYP/aug-cc-pVTZ. First, what is the nature of\nthis band? The band seems to be slightly\nshifted with respect to the main 5.3 eV band contribution observed\nin the spectrum of other anions. This apparent shift may be caused\nby the overlap with higher energy bands for all the anions except\nC 2 H – , which is well separated. However,\nsuch overlap does not explain the low energy onset of the resonance. Therefore it is possible that the band results from a shape resonance.\nWe explored the Rydberg states of the system using the semiempirical\nformula of Gallup; 74 however, we did not\nfind any high lying shape resonances (LUMO → 0.07 eV, LUMO\n+ 1 → 2.24 eV, LUMO + 2 → 3.29 eV). Still, the dipole\nsupported inner shell resonances may be possible as described by Khvostenko\net al. 63 In furan the resonance is positioned\nat 5.55 eV, 63 which is ∼0.5 eV below\nthe corresponding S 1 state. A similar shift was proposed\nfor amides in the recent study of Li et al., 75 and the same shift due to a dipole interaction may also well describe\nthe low lying C 2 H – resonance in the present\ncase. Irrespective of the nature of this resonance, we can see from Table 4 that it cannot be\nexplained by any direct dissociation channel, so a more complex reaction\nmust be involved. Several complex dissociation channels resulting\nin formation of\nC 2 H – anions are listed in Table 4 . These include reaction channels\nwhere hydrogen migrates from/to carboxyl group, reactions 2a, 3a,\nand 3b, or within the ring, reactions 4 and 5. The lowest threshold\nof 3.19 eV was estimated for reaction 5 with only one neutral cofragment.\nAny of these reactions can explain the lowest energy band in the C 2 H – anion signal, but identifying the most\nlikely will require further experiments or an estimation of the different\nreaction barriers. The second interesting observation is the\nformation of m / z 67 anion from the\nmethyl-substituted\nspecies. The anion formed by simple cleavage of the carboxyl group\ncorresponds to m / z 66. Formation\nof m / z 67 anion therefore requires\na more complex fragmentation pathway, with hydrogen migration from\nthe carboxyl group to the furan ring. It is now interesting\nto explore how methyl substitution at different\npositions in the molecule influences hydrogen migration. The ion yield\nfor C 4 H 3 O – ( m / z 67) in Figure 3 d significantly changes with methyl substitutions.\nThe three main bands positioned at 5.3, 6.7, and 9.4 eV are visible\nin 2FA. For 3M-2FA the band located around 6.7 eV is significantly\nsuppressed, whereas for 5M-2FA a decrease in ion yield is observable\nto the band around 5.3 eV, leaving the middle band (6.7 eV) the strongest.\nMethylation also results in a slight decrease of the intensity of\nthe 9.4 eV band. When exploring the yields of C 2 H – ( m / z 25) in Figure 3 e, we can see only\ntwo resonances in the\ncase of 2FA around 5 and 9 eV, which do not change upon substitution\nat the 5 position of the ring. However, substitution at the 3 position\nof the ring results in complete disappearance of the low 5.3 eV resonance\nand in the formation of a new peak in the anion yield around 6.7 eV. The present experimental results do not allow us to unambiguously\nidentify what causes these strong changes in the hydrogen migration\nafter methyl substitution at the 3 position of the ring. We can only\nspeculate about the mechanisms involved. The disappearance of\nthe 5.3 eV resonance in the yield of the m / z 25 anion after 3-methylation may be\npartially described by the new signal at this energy in the yield\nof m / z 39 anion, corresponding to\nthe C 3 H 3 – anion due to the attached methyl group. This gives\nus only two options for how the C 2 H – can\nbe extracted from the furan ring at low energies. The carbons of the\nresulting C 2 H – anion are, in the first\noption, the 2 and 3 carbons of the furan ring and, in the second option,\nthe 3 and 4 carbons of the ring. Returning to Table 4 , only reaction mechanisms 2a, 3a, and 3b\nare consistent with this explanation. The appearance of a new resonance\nin the yield of m / z 25 anion at\n6.7 eV after 3-methylation can be then linked to the disappearance\nof this resonance in the yield of m / z 67 anion. It seems that, after cleavage of the carboxylic group\nfrom 3M-2FA, the anion becomes highly unstable and dissociation via\nrelease of C 2 H – will be much easier than\nstabilization of the ring. As the mechanism requires hydrogen migration,\nthe low stability of the 3M-2FA anion may be an explanation also for\nthe disappearance of the signal of Feshbach resonances in the (M–H) − anion signal described in section 4.1 . As a further note, our experimental\nmeasurements are performed\nunder conditions which are similar to our earlier work on pyruvic\nacid, 22 where we observed the formation\nof dimer anions and consequently parent anions. In the present study,\nwe did not observe the creation of stable molecular parent anions,\nnor dimers. This again indicates a strong interaction of the ring\nwith the COOH group. Consequently, the ability of the carboxyl group\nto form intermolecular hydrogen bonds is reduced. The binding energy\nof the carboxyl group bonded dimer of 2FA is ∼0.26 eV as reported\nby Ghalla et al., 24 which is less than\nhalf of the energy for formic acid (∼0.62 eV 76 )."
} | 6,074 |
38658521 | PMC11043267 | pmc | 6,895 | {
"abstract": "The unsustainable and widespread utilization of fossil fuels continues to drive the rapid depletion of global supplies. Biodiesel has emerged as one of the most promising alternatives to conventional diesel, leading to growing research interest in its production. Microbes can facilitate the de novo synthesis of a type of biodiesel in the form of fatty acid methyl esters (FAMEs). In this study, Saccharomyces cerevisiae metabolic activity was engineered to facilitate enhanced FAME production. Initially, free fatty acid concentrations were increased by deleting two acetyl-CoA synthetase genes ( FAA1, FAA4 ) and an acyl-CoA oxidase gene ( POX1 ). Intracellular S-adenosylmethionine (SAM) levels were then enhanced via the deletion of an adenosine kinase gene ( ADO1 ) and the overexpression of a SAM synthetase gene ( SAM2 ). Lastly, the S. cerevisiae strain overproducing free fatty acids and SAM were manipulated to express a plasmid encoding the Drosophila melanogaster Juvenile Hormone Acid O -Methyltransferase ( Dm JHAMT). Using this combination of engineering approaches, a FAME concentration of 5.79 ± 0.56 mg/L was achieved using these cells in the context of shaking flask fermentation. To the best of our knowledge, this is the first detailed study of FAME production in S. cerevisiae . These results will provide a valuable basis for future efforts to engineer S. cerevisiae strains for highly efficient production of biodiesel. Supplementary Information The online version contains supplementary material available at 10.1186/s13568-024-01702-7.",
"introduction": "Introduction \nOngoing concerns about the impact of global climate change and energy security have widespread implications for environmental integrity, sustainability, and food security (Kaljuvee and Kuusik 1999 ; Kotcher et al. 2019 ). In an effort to address fossil fuel depletion, pollution, and the negative effects of these fuels on the climate, there is an urgent need to move away from fossil fuel dependency in favour of renewable energy sources (Khan and Fu 2020 ). Biodiesel has emerged as a promising fossil fuel alternative with the potential to be produced in an ecologically friendly manner (Alishah Aratboni et al. 2019 ; Snowdon R Fau - Friedt 2012 ). \nBiodiesel consists of the ethyl- and/or methyl-esters of linear alkyl chains, and serves as a clean-burning renewable fuel that can be directly used in current diesel engines without any need for additional modifications (Nady et al. 2020 ). As compared to fossil fuels, biodiesel exhibits excellent ignition properties and can reduce the emission of CO 2 and other fumes by 78% (Elkelawy et al. 2022 ). \nBiodiesel production is primarily achieved via the transesterification of animal fat- or vegetable oil-derived triacylglyceride oils with methanol or ethanol using an alkaline catalyst (Leung et al. 2010 ). However, as these oils are also consumed by humans, using them to produce biodiesel would result in higher prices for both vegetable oil and the resultant biodiesel (Verma and Kuila 2020 ). The crops required to produce these oils also require extensive agricultural space and an extended growth period (Zhang et al. 2021 ). To overcome these issues, research groups have explored the production of biodiesel using yeast, fungi, and algae capable of lipid biogenesis (Bhatia et al. 2017 ; Rasmey et al. 2020 ; Zhang et al. 2021 ). The microbe-based production of biodiesel necessitates access to large-scale cultivation and harvesting systems (Rasmey et al. 2020 ). The growth of these microbes needs to be regulated with great care to establish optimal conditions for large-scale biodiesel generation. The establishment of metabolically engineered microbial species capable of directly synthesizing biodiesel represents a promising alternative to the use of biodiesel derived from plant or animal fats. \nThe Steinbüchel group made the first effort to achieve fatty acid ethyl esters (FAEEs) synthesis in Escherichia coli (Elbahloul and Steinbüchel 2010 ). To achieve this goal, E. coli was engineered to use two orthogonal pathways to synthesize fatty acyl-CoA and ethanol, with these substrates then processed by a wax ester synthase to facilitate the synthesis of FAEEs (Elbahloul and Steinbüchel 2010 ). The Keasling group further optimized E. coli- mediated FAEEs synthesis, achieving biodiesel yields of 1.5 g/L (Zhang et al. 2012 ). To simplify biodiesel synthesis pathways, a Mycobacterium marinum (MmMT)-derived S - adenosyl- L- methionine (SAM)-dependent methyltransferase was introduced into E. coli by the Lykidis group to directly catalyze endogenous fatty acid and SAM reactions to facilitate fatty acid methyl esters (FAMEs) production (Nawabi et al. 2011 ). However, the MmMT methyltransferase exhibits a high degree of specificity with a preference for less common fatty acids harboring a 3-hydroxy group, resulting in low FAMEs yields (Nawabi et al. 2011 ). To address this issue, the Saken group identified Drosophila melanogaster Juvenile Hormone Acid O-Methyltransferase ( Dm JHAMT) as an enzyme with broader fatty acid specificity (Sherkhanov et al. 2016 ). The introduction of Dm JHAMT into E. coli and the further enhancement of intracellular SAM supply levels was sufficient to increase FAME yields to 0.56 g/L (Sherkhanov et al. 2016 ). \n Saccharomyces cerevisiae has been extensively studied as a model microorganism and has many advantageous characteristics that make it well suited to commercial use, including resistance to phage contamination, the ability to undergo high-density fermentation, and a high level of robustness suitable for growth under industrial conditions (Jin and Cate 2017 ; Leber et al. 2015 ; Lian and Zhao 2015 ). These characteristics have led to growing interest in its use as a host to facilitate fatty acid-derived chemical and biofuel production. Under normal conditions, C16 and C18 fatty acids comprise the majority of the lipid content present within S. cerevisiae cells, and these fatty acids are suitable precursors that can be used to generate biodiesel (Keasling 2013 ; Leber et al. 2015 ). Accordingly, several research teams have engineered S. cerevisiae to produce FAEEs via the heterologous expression of wax ester synthase or an acyl-CoA: alcohol acyltransferase (De Jong et al. 2015 ; Valle-Rodríguez et al. 2014 ; Zhou et al. 2016 ). To date, however, there have been no reports of the metabolic engineering of yeast to facilitate FAMEs production. \nIn the present study, S. cerevisiae cells were engineered to achieve FAMEs production. Initially, genetic engineering was used to modulate the free fatty acid and SAM metabolic pathways to increase the levels of these FAME precursors. The Dm JHAMT methyltransferase was then introduced into the resultant free fatty acid- and SAM-overproducing S. cerevisiae strain. Together, this combined engineering strategy was used to successfully achieve the first reported instance of de novo FAMEs synthesis in S. cerevisiae.",
"discussion": "Discussion Biodiesel has emerged as a promising alternative to fossil fuels and attracted increasing attention from researchers (Ramos et al. 2022 ; Sharma et al. 2020 ). Several research teams have achieved de novo microbe-mediated biodiesel production (Table 1 ). In E. coli , an FAEE yield of up to 1.5 g/L has been reported, and is the highest FAEE yield to date (Zhang et al. 2012 ). The highest measured FAEE yield in S. cerevisiae is 0.52 g/L (Yu et al. 2012 ). Unlike FAEEs, however, there have been far fewer studies focused on FAME synthesis. Yunus et al. engineered cyanobacteria to facilitate the conversion of CO 2 into FAMEs at a yield of up to 120 mg/L in 10 days (Yunus et al. 2020 ). In E. coli , Nawabi et al. were able to induce FAME biosynthesis via the introduction of exogenous methyltransferases, although the utilized methyltransferase exhibits a high degree of specificity for rare 3-hydroxy group-containing fatty acids, contributing to poor FAME yields (Nawabi et al. 2011 ). To address this issue, Sherkhanov et al. identified Dm JHAMT as a broad-spectrum methyltransferase, and they introduced this methyltransferase into E. coli that had been engineered to overproduce SAM and fatty acids, resulting in the production of 0.56 g/L of FAMEs (Sherkhanov et al. 2016 ). No studies to date, however, have reported on the use of S. cerevisiae for FAME biosynthesis, which was achieved for the first time in the present study. \n Table 1 Comparisons of biodiesel production from different organisms Microorganisms Products Titre Cultivation condition Reference \n E. coli \n FAEEs 1.5 g/l Shake flask (Zhang et al. 2012 ) \n S. cerevisiae \n FAEEs 230 mg/L Shake flask (Teo et al. 2015 ) \n S. cerevisiae \n FAEEs 25 mg/L Shake flask (Thompson and Trinh 2014 ) \n S. cerevisiae \n FAEEs 6.3 mg/L Shake flask (Shi et al. 2012 ) \n S. cerevisiae \n FAEEs 5 mg/L Shake flask (Runguphan and Keasling 2014 ) \n S. cerevisiae \n FAEEs 34 mg/L Shake flask (Shi et al. 2014 ) \n S. cerevisiae \n FAEEs 17.2 mg/L Shake flask (Valle-Rodríguez et al. 2023 ) \n S. cerevisiae \n FAEEs 0.52 g/L Shake flask (Yu et al. 2012 ) \n E. coli \n FAMEs 0.56 g/L Shake flask (Sherkhanov et al. 2016 ) \n E. coli \n FAMEs 70.5 μm Shake flask (Nawabi et al. 2011 ) Synechocystis sp. FAMEs 120 mg/L Shake flask (Yunus et al. 2020 ) \n S. cerevisiae \n FAMEs 4.69 mg/L Shake flask This study \n Given that the FAEE and FAME yields achieved to date are well below commercial levels, additional efforts are required, particularly with respect to precursor supplies and enzyme activity. SAM and free fatty acids serve as FAME precursors, and efforts to increase their concentrations are thus needed to ensure efficient FAME biosynthesis. Many strategies for increasing free fatty acid concentrations in yeast cells have been explored to date. Zhou et al., for example, blocked fatty acid activation and degradation to generate yeast with high free fatty acid yields via the introduction of an optimized acetyl-CoA pathway, the expression of a fatty acid synthase with greater efficiency, and the overexpression of endogenous acetyl-CoA carboxylase (Zhou et al. 2016 ). In a fed-batch fermentation system, their engineered strain was able to achieve a free fatty acid yield as high as 10.4 g/L (Zhou et al. 2016 ). Methionine adenosyltransferase synthesizes SAM from ATP and methionine, and SAM subsequently serves as the primary biological methyl donor as it contains an active methylthioether group. To date, studies of yeast have revealed many genes involved in SAM accumulation, and metabolic engineering approaches can improve the efficiency of SAM synthesis in S. cerevisiae (Kanai et al. 2017 ). Zhao et al. established a yeast strain capable of more efficient SAM synthesis by combining SAM2 gene overexpression and blocking the SAM decarboxylation pathway, achieving a SAM yield of 12.47 g/L with a fed-batch cultivation approach (Zhao et al. 2016 ). In this study, the deletions of the FAA1, FAA2 , and POX1 genes were conducted to enhance free fatty acid concentrations within S. cerevisiae cells, while ADO1 deletion and SAM2 overexpression were performed to enhance SAM accumulation (Fig. 4 ). The introduction of the DmJHAMT gene ultimately enabled a FAME yield of 5.79 ± 0.56 mg/L from these engineered yeast. This study represents the first report of the de novo synthesis of FAMEs in S. cerevisiae. While free fatty acid and SAM concentrations in these yeast cells were enhanced via metabolic engineering, this strategy only resulted in a FAME yield of 5.79 ± 0.56 mg/L. The largest barrier to the optimization of FAME production may be related to the suboptimal performance of the Dm JHAMT enzyme. Further protein engineering strategies focused on Dm JHAMT may represent a feasible means of enhancing FAME levels. Therefore, a series of rational design efforts based on known or simulated protein structures and directed evolution will be carried out in the future. \n Fig. 4 Fatty acid synthesis pathway in S. cerevisiae and engineering strategies for biodiesel synthesis. Overexpression targets are shown in green, and knockout targets are in red. Metabolic pathways that need to be blocked are marked with a cross. Double arrows represent multiple steps. EMP, Embden-meyerhof pathway; TCA, tricarboxylic acid; pyruvate decarboxylase; SAM, S-adenosylmethionine; SAH, S-adenosine homocysteine; FFA, free fatty acid; FAMEs, fatty acid methyl esters. SAM2 encodes S-andenosyl-l-methionine synthetase; FAA1 and FAA4 encode fatty acyl-CoA synthetase; ADO1 encodes adenosine kinase; POX1 encodes acyl-CoA oxidase; DmJHAMT encodes O-Methyltransferase"
} | 3,185 |
26185074 | null | s2 | 6,897 | {
"abstract": "Microbes produce a biofilm matrix consisting of proteins, extracellular DNA, and polysaccharides that is integral in the formation of bacterial communities. Historical studies of polysaccharides revealed that their overproduction often alters the colony morphology and can be diagnostic in identifying certain species. The polysaccharide component of the matrix can provide many diverse benefits to the cells in the biofilm, including adhesion, protection, and structure. Aggregative polysaccharides act as molecular glue, allowing the bacterial cells to adhere to each other as well as surfaces. Adhesion facilitates the colonization of both biotic and abiotic surfaces by allowing the bacteria to resist physical stresses imposed by fluid movement that could separate the cells from a nutrient source. Polysaccharides can also provide protection from a wide range of stresses, such as desiccation, immune effectors, and predators such as phagocytic cells and amoebae. Finally, polysaccharides can provide structure to biofilms, allowing stratification of the bacterial community and establishing gradients of nutrients and waste products. This can be advantageous for the bacteria by establishing a heterogeneous population that is prepared to endure stresses created by the rapidly changing environments that many bacteria encounter. The diverse range of polysaccharide structures, properties, and roles highlight the importance of this matrix constituent to the successful adaptation of bacteria to nearly every niche. Here, we present an overview of the current knowledge regarding the diversity and benefits that polysaccharide production provides to bacterial communities within biofilms."
} | 423 |
39885323 | PMC11782530 | pmc | 6,900 | {
"abstract": "Methanogenic archaea (methanogens) possess fascinating metabolic characteristics, such as the ability to fix molecular nitrogen (N 2 ). Methanogens are of biotechnological importance due to the ability to produce methane (CH 4 ) from molecular hydrogen (H 2 ) and carbon dioxide (CO 2 ) and to excrete proteinogenic amino acids. This study focuses on analyzing the link between biological methanogenesis and amino acid excretion under N 2 -fixing conditions. Among five hydrogenotrophic, autotrophic methanogens, Methanothermobacter marburgensis was prioritized and further cultivated in closed batch cultivation mode under N 2 -fixing conditions. M. marburgensis was grown on chemically defined minimal medium with different concentrations of ammonium in a H 2 /CO 2 /N 2 atmosphere. This enabled the quantification of ammonia uptake, N 2 -fixation, amino acid excretion and the conversion of H 2 /CO 2 to CH 4 . To quantify N 2 -fixation rates in a mass balance setting a novel method has been established. The method utilizes the pressure drop below a certain threshold pressure in closed batch cultivation mode – the th reshold p ressure for N 2 -fix ation (THp N2fix ). Using the THp N2fix method, volumetric N 2 -fixation rates of M. marburgensis as high as 0.91 mmol L −1 h −1 were determined. Excretion of amino acids was found with highest detected values of glutamic acid, alanine, glycine and asparagine. The highest total amino acid excretion of 7.5 µmol L −1 h −1 was detected with H 2 /CO 2 /N 2 at an ammonium concentration of 40 mmol L −1 . This study sheds light on the link between methanogenesis, biological N 2 -fixation, and proteinogenic amino acid excretion. The concomitant production of amino acids and CH 4 could become of biotechnological relevance in an integrated approach coupling biomethanation and N 2 -fixation in a biorefinery concept. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-025-87686-1.",
"conclusion": "Conclusions This study sheds new light on the link between methanogenesis, biological N 2 -fixation, and proteinogenic amino acid excretion. We show the inherent industrial potential of M. marburgensis for amino acid excretion under N 2 -fixing conditions. The THp N2fix method can be used to determine N 2 -fixation rates of hydrogenotrophic, autotrophic, diazotrophic methanogens – without the need for GC measurements. Moreover, the THp N2fix methodology may serve as the basis for establishing high-throughput screening of methanogens and other gas-fermenting organisms where a pressure drop or increase occurs. This study has implications for research in microbial physiology, ecology, and biotechnology of amino acid excretion, methanogenesis and N 2 -fixation and may serve as basis for developing applications in gas fermentation and Archaea Biotechnology.",
"introduction": "Introduction The greenhouse gasses (GHGs) carbon dioxide (CO 2 ), methane (CH 4 ) and nitrous oxide (N 2 O) accumulate in the atmosphere of Earth where they contribute heavily to global warming and climate change. The accumulation of GHGs mainly results from anthropogenic combustion of fossil fuels, industrial processes and agriculture 1 – 3 . One example for high CO 2 emissions in industrial processes is the production of ammonia via the Haber-Bosch process. The Haber-Bosch process catalyzes the reaction of molecular hydrogen (H 2 ) and molecular nitrogen (N 2 ) to ammonia and demands a metal catalyst, high temperature (400–500 °C) and high pressure (150–350 bar) and therefore results in 1.5 tons CO 2 per ton NH 3 produced 4 – 6 . From a nitrogen perspective, 174 million tons of NH 3 are produced each year globally applied as artificial fertilizers to ensure an increased harvest of crops 4 , 7 . Thus, biological nitrogen N 2 -fixation (diazotrophy), to fertilize soil and produce substances relevant for human nutrition, such as amino acids, is one solution to reduce global CO 2 emission tremendously. Due to the highly stable triple bond between the nitrogen atoms, it is difficult for microorganisms to fix N 2 directly from the environment. Although this makes N 2 unavailable as a biological nitrogen source for most organisms, some microorganisms from the domain of Archaea and Bacteria developed the trait to fix N 2 . This process is coupled to highly expensive metabolic costs of 16 ATP per N 2 fixed 8 – 10 . In Bacteria, diazotrophy is found in symbiotic and free-living species. The symbiosis ranges from the known example between legumes and intracellular diazotrophs, such as Azobacter spp. or Pseudomonas spp., to symbiotic relationships between diazotrophic bacteria and sponges, gymnosperms, or even insects 11 . Among the Archaea diazotrophy occurs among the methanogenic archaea (methanogens) 12 – 15 . Methanogens are anaerobic microorganisms and known for the ability to generate methane (CH 4 ) as the end product of their energy metabolism. According to their substrate utilization spectrum, methanogens can be divided into different metabolic groups: hydrogenotrophic (H 2 , formate or simple alcohols), aceticlastic (acetate), methylotrophic (compounds containing a methyl group), H 2 -dependent methylotrophic (methylated compounds with H 2 as electron donor), and methoxydotrophic (methoxylated aromatic compounds). Some methanogens are able to grow autotrophicly and hydrogenotrophicly by reducing CO 2 with H 2 to CH 4 , and play a crucial role in the global carbon cycle 16 . The first proof of diazotrophy in methanogens was shown in Methanosarcina barkeri strain 227 (DSM 1538) 17 and Methanothermococcus thermolithotrophicus DSM 2095 18 . Since then, diatrozophy has been further studied in M. thermolithotrophicus 10 , 18 , Methanococcus maripaludis 19 – 21 , M. barkeri 22 , Methanosarcina mazei 23 , 24 , Methanocaldococcus sp. 14 and in Methanothermobacter marburgensis 18 . One way to make use of diazotrophy in biotechnology could be to store the fixed N 2 in nitrogen containing metabolic end products, such as proteinogenic amino acids. Proteinogenic amino acids are applied in a variety of industrial sectors, such as food and feed, agriculture, pharmaceuticals, or packaging and housing. The production of amino acids through bacterial fermentation has marked an important branch of biotechnology for several decades 25 . Through genetic engineering of the main microorganisms for amino acid production, Escherichia coli and Corynebacterium glutamicum , were turned into highly optimized microbial cell factories for commercial production of amino acids 26 – 32 . However, the metabolic potential of archaea with regard to amino acid excretion has up to now been vastly overlooked 33 – 36 . In the emerging research and development field of Archaea Biotechnology 33 , 34 , the production of proteinogenic amino acids, which are important nutritional compounds, could be linked to the production of CH 4 , which is an important biofuel 35 . Several studies already showed the biotechnological potential of methanogens in biomethanation 37 – 47 . Moreover, methanogens were recently reported to excrete proteinogenic amino acids 35 , 48 – 51 . However, to our knowledge, the interplay between diazotrophy, amino acid production and CH 4 production has not been described yet in the scientific literature. The aim of our research was to examine the physiological and biotechnological characteristics of biological CO 2 - and N 2 -fixation in connection to proteinogenic amino acid excretion and CH 4 production. Among five methanogens, M. marburgenis was prioritized to investigate N 2 -fixation, CH 4 production, and amino acid excretion characteristics in closed batch cultivation mode at different NH 4 + concentrations. Moreover, to unambiguously prove biological N 2 -fixation in a closed batch system and in a mass balance setting, a method for proving biological N 2 -fixation by methanogens without N-labelling techniques was also developed. The method focusses on headspace gas conversion in an isobaric setting in closed batch cultivation mode by undercutting a certain pressure – the theoretical threshold (THp N2fix ). Since M. marburgensis was reported to be able to grow solely on N 2 19 and is able to excrete proteinogenic amino acids 35 we hypothesized that M. marburgensis utilizes N 2 as the sole source of nitrogen for proteinogenic amino acid production and excretion.",
"discussion": "Discussion Agriculture and the production of artificial N-fertilizers are an indirect source of GHG emissions through releasing N 2 O via nitrification of ammonia (NH 3 ) 2 , 4 . To generate this ammonia fertilizer, the Haber-Bosch process is the main industrial procedure for synthetic N 2 -fixation and responsible for a release of 1.5 tons of CO 2 per ton of NH 3 produced 6 . Therefore, the identification of microbial strains optimized for molecular N 2 -fixation could reduce the amount of chemically produced ammonia via the Haber-Bosch process. Physiologically, an ideal strain for N 2 -fixation in a biorefinery concept should include certain properties, such as high specific growth rate, high specific CO 2 conversion and high N 2 -fixing abilities. Furthermore, the strain should be cultivatable in chemically defined minimal medium exclusively with inorganic compounds and substrates. Among the methanogens, several candidates would fulfill these prerequisites. For this reason, suitable methanogens should be identified using a fast and simple screening methodology regarding their N 2 -fixation characteristics. Therefore, this study proposes a fast and easy detection method of diazotrophic characteristics of autotrophic, hydrogenotrophic methanogens using THp N2fix as proof of N 2 -fixation and concomitant amino acid production. Moreover, in the context of “power to gas” technology, biomethanation of CO 2 43 – 45 , 48 in combination with N 2 -fixation for the excretion of amino acids could become of high economic interest 48 – 51 . To our knowledge there is no study yet that examined combined CO 2 - and N 2 -fixation regarding amino acid excretion. However, this study should only be seen as proof of principle. For a high throughput screening of THp N2fix and quantification of NUR, online measurements of the headspace pressure would be clearly desirable 40 , 41 . In addition, in-line or at-line measurements of biomass and of NH 4 + concentrations would be needed to be able to examine the growth and production kinetics of methanogens under N 2 -fixing conditions. Furthermore, methanogens may use carbonates as an additional carbon source 59 . A carbon source in the media would render the THp N2fix method as an indicator for N 2 -fixation impossible, as additional H 2 utilization beyond the 4:1 ratio would potentially occur. For that reason, it was necessary to exclude carbonate from the media for the purpose of converting CO 2 to CH 4 only from the gas that had been supplied in the serum bottle headspace. Gassing with H 2 /CO 2 /N 2 generally resulted in lower growth ( Supplementary Fig. 4 ) when compared to gassing with only H 2 /CO 2 ( Supplementary Fig. 2 , Supplementary Fig. 3 ). This could be explained via constrained growth by the expensive N 2 -fixing process 9 . Within this set-up, undercutting the THp N2fix while fixing N 2 was successful in experiments where NH 4 + was limiting, ranging between 2 and 10% and in a time frame from 40 to 77 h of cultivation (Fig. 3 ). The qN 2 of M. marburgensis (Data Sheet 2: NUR_qN2) detected in this study are in the range 5–20% the maximum specific CH 4 production rates 54 . This high physiological capacity of M. marburgensis for diazotrophic growth seems to agree with that of M. thermolithotrophicus 10 . Regarding the capability of methanogens for N 2 -fixation, M. marburgensis harbours a nif gene cluster 13 and the organism was already examined in continuous culture regarding its growth yield under N 2 -fixing conditions 18 . Thus, M. marburgensis has been shown to be capable to growing on NH 4 + -free medium solely using N 2 as nitrogen source. However, here we show that NH 4 + addition is required to enable N 2 -fixation of M. marburgensis and that there seems to be an optimal concentration of NH 4 + between 2.5 and 10% for optimal N 2 -fixation. At higher concentrations of NH 4 + (50% and 100%) (Fig. 3 ), the pressure rarely dropped below the THp N2fix , and it seemed that N 2 -fixation was slightly inhibited or at least reduced. This could be related to the fact that M. marburgensis is usually grown at NH 4 + concentrations of 20 to 40 mmol L − 1 38 , 40 , 53 , 54 which acts as the preferred nitrogen source, or that the tested NH 4 + concentrations above 20 mmol L −1 physiologically down-regulates the nitrogenase activity 20 , 60 . NH 4 + switch off can occur in methanogens. This means that there is an inactivation of N 2 -fixation once a thermodynamically superior nitrogen source is available 20 . A switch-off of the nitrogenase activity with addition of NH 4 + as a superior and more easily accessible nitrogen source, as seen in M. maripaludis 60 could not be fully confirmed here. However, during growth of M. maripaludis on Ala only a partial switch-off of the nitrogenase activity has been observed 60 . Thus, it might be possible that in M. marburgensis the uptake of Ala allows for N 2 -fixation in the presence of NH 4 + ( Supplementary Fig. 5 ). N 2 fixation in the presence of NH 4 + was studied e.g., in M. barkeri 17 and in M. maripaludis 60 , but in both studies N 2 -fixation in the presence of NH 4 + had not been observed. Except for our study, N 2 -fixation in the presence of NH 4 + has only been reported in a single experiment 61 . In experiments with washed M. marburgensis biomass compared to non-washed biomass, metabolic expensive N 2 -fixation was proven even with higher NH 4 + concentrations (Figs. 3 and 4 ). This might have occurred due to the slow growth and introduced stress because of the washing step. Thus, amino acid excretion could also be the reason for the partially inconsistent and higher HUR:CUR ratio and C-balances ( Supplementary Table 3 ). The results presented in this study confirm that M. marburgensis is a diazotrophic organism. However, we could not yet confirm earlier findings that the sole N-source of M. marburgenis can be N 2 19 . In independent unpublished experiments with M. marburgensis using N 2 as sole source of nitrogen, the organism did also not grow or produce CH 4 (Nevena Maslać, personal communication). We show that the N 2 is converted by M. marburgensis during growth on H 2 /CO 2 /N 2 in chemically defined minimal medium into proteinogenic amino acids (Fig. 5 ) which are excreted into the growth medium. One could argue that the amino acids were not actively or passively excreted. However, in a previous study it has been shown that cell lysis was not a substantial source of amino acid excretion by M. marburgensis 35 . In this study a variety of amino acids were excreted by M. marburgensis with the highest total amount of up to 7.5 µmol L −1 h −1 in early time points (Data sheet 1: HPLC Productivity). Interestingly, Ala seems to be consumed in later time points of 5%, 7.5% and 10% experiments (Fig. 5 ). This finding might provide insight for the varying NUR (Fig. 4 ), as NH 4 + could hinder nitrogenase enzyme activity, whereas the utilization of Ala could switch to an intermediate regulatory response 60 . Amino acid excretion rates of M. marburgensis under N 2 -fixing conditions can currently not compete with the genetically engineered and high amino acid producing organisms C. glutamicum and E. coli 26 . Highest concentrations of Glu reached 0.045 g L −1 compared to genetically modified C. glutamicum of 40 g L −1 (Table 1 ). Nevertheless, they succeeded to create a modified C. glutamicum from a wild type with no L-Lysine production to 0.6 to 4.0 g L −1 h −1 28 . Furthermore, conventional amino acid excreting cell factories are engineered to only overproduce one specific amino acid and are, unlike M. marburgensis , not able to fix atmospheric N 2 for their production 26 , 29 . Moreover, the carbon and energy substrate for E. coli and C. glutamicum are carbohydrates and M. marburgensis utilizes CO 2 and H 2 . With the synthetic biology tools that have become available to genetically enhance methanogens 55 , 62 – 66 , some of these organisms may become cell factories for proteinogenic amino acid production. \n Table 1 Comparison of amino acid concentrations from this study with M. marburgensis to E. coli and C. glutamicum . Organism Amino acid Closed batch value this study / g L −1 Value from other studies / g L −1 Reference \n C. glutamicum \n 1 \n Asp 0.0031 0.46 \n 67 \n \n C. glutamicum \n 1 \n Leu n.d. 0 \n 68 \n \n C. glutamicum \n 1 \n Lys 0.0014 0 \n 27 \n \n C. glutamicum \n 2 \n Ala 0.0109 110 \n 26 \n \n C. glutamicum \n 2 \n Arg 0.0011 92 \n 69 \n \n C. glutamicum \n 2 \n Gln 0.0015 73.5 \n 70 \n \n C. glutamicum \n 2 \n Glu 0.0448 40 \n 26 \n \n C. glutamicum \n 2 \n Ile n.d. 32 \n 71 \n \n C. glutamicum \n 2 \n Leu n.d. 20 \n 68 \n \n C. glutamicum \n 2 \n Lys 0.0014 120 \n 27 \n \n C. glutamicum \n 2 \n Met n.d. 7 \n 72 \n \n C. glutamicum \n 2 \n Ser 0.0005 44 \n 73 \n \n C. glutamicum \n 2 \n Try 0.0030 65 \n 26 \n \n E. coli \n 2 \n Ala 0.0109 114 \n 74 \n \n E. coli \n 2 \n Phe 0.0001 60 \n 26 \n \n E. coli \n 2 \n Thr 0.0008 90 \n 26 \n \n E. coli \n 2 \n Val n.d. 60–70 \n 26 \n n.d.: not detected. 1 wild-type microorganism. 2 genetically engineered microorganisms."
} | 4,407 |
22647231 | null | s2 | 6,902 | {
"abstract": "A recent resurgence in basic and applied research on photosynthesis has been driven in part by recognition that fulfilling future food and energy requirements will necessitate improvements in crop carbon-fixation efficiencies. Photosynthesis in traditional terrestrial crops is being reexamined in light of molecular strategies employed by photosynthetic microbes to enhance the activity of the Calvin cycle. Synthetic biology is well-situated to provide original approaches for compartmentalizing and enhancing photosynthetic reactions in a species independent manner. Furthermore, the elucidation of alternative carbon-fixation routes distinct from the Calvin cycle raises possibilities that novel pathways and organisms can be utilized to fix atmospheric carbon dioxide into useful materials."
} | 198 |
40307847 | PMC12063673 | pmc | 6,904 | {
"abstract": "Abstract Sulfuric acidic hot springs (<pH 4.0, >37°C) are found in volcanic regions worldwide, where various bacteria, archaea, and the unicellular red algae Cyanidiophyceae dominate. Regarding heterotrophic eukaryotes, the only known species was the thermophilic amoeboflagellate Tetramitus thermacidophilus (class Eutetramitea, phylum Heterolobosea), which feeds on surrounding bacteria and archaea. In this study, we investigated three sulfuric hot springs (34.7°C–50°C, ∼pH 2.0) in Japan to determine whether other heterotrophic eukaryotes inhabit these environments. As a result, we isolated and identified cultures of four species capable of surviving at pH 2.0 and 40°C: Allovahlkampfia sp. (Eutetramitea, Heterolobosea); Nuclearia sp. and Parvularia sp. (Nucleariidea, Cristidiscoidea); and Vannella sp. (Discosea, Amoebozoa). Phylogenetic analyses suggest that these four species independently evolved from mesophilic and neutrophilic ancestors, separate from each other. Additionally, Platyophrya sp. (Colpodea, Ciliophora) and two species of Neobodo (Euglenozoa, Kinetoplastea) were also found in the same environment, while their maximum survival temperatures were 35°C and 30°C, respectively. Among these, all species except Neobodo were confirmed to grow exclusively by feeding on Cyanidiococcus sp., a dominant species of Cyanidiophyceae in the environment. Thus, various lineages of heterotrophic unicellular eukaryotes have independently developed acidophilic and thermotolerant traits, allowing them to colonize sulfuric hot springs.",
"introduction": "Introduction Life on Earth is remarkably diverse, thriving in environments ranging from temperate ecosystems to the most extreme habitats. Extremophiles, unique groups of organisms capable of surviving and thriving under extreme physical and chemical conditions, challenge traditional notions of the limits of life. These environments include extreme temperatures, high salinity, intense acidity or alkalinity, elevated pressure, and high levels of radiation (Shu and Huang 2022 , Rappaport and Oliverio 2023 ). The study of extremophiles has revealed remarkable biochemical and physiological adaptations that enable survival under such stresses. This knowledge not only enhances our understanding of evolutionary biology (Shu and Huang 2022 , Rappaport and Oliverio 2023 ) but also offers innovative applications (Littlechild 2015 ). For instance, extremophiles contribute to environmental remediation by breaking down pollutants or immobilizing heavy metals, and their heat- and salt-stable enzymes, such as DNA polymerases, have become invaluable tools in various industrial and biotechnological processes (Littlechild 2015 ). Research on extremophiles has largely focused on prokaryotes (bacteria and archaea), which dominate extreme habitats and are easier to study (Shu and Huang 2022 ). In contrast, protists (unicellular eukaryotes) have been historically overlooked due to assumptions that their complex structures and high energy demands make them poorly suited for extreme environments (Rappaport and Oliverio 2023 ). Technical challenges in isolating protists and limited knowledge of their diversity have further contributed to their neglect. However, advances in environmental metagenome sequencing are revealing the presence of diverse protists in extreme ecosystems, challenging earlier assumptions and broadening our understanding of life in such conditions (Rappaport and Oliverio 2023 ). Still, the detected nucleic acids may originate from organisms that temporarily entered the environment from nearby habitats or from dead cells, necessitating confirmation through other methods, such as cultivation. Sulfuric hot springs (generally <pH 4.0 and >37°C), found worldwide around volcanic areas, are examples of habitats for extremophiles. The extremely low pH of these waters is due to the dissolution and oxidation of sulfur, sulfur dioxide, and hydrogen sulfide exposed to water and oxygen, which produces sulfuric acid. In addition, oxidation of sulfide minerals (e.g. pyrite) in underground rocks and sulfur-oxidizing bacteria also leads to sulfuric acid production (Dopson and Johnson 2012 ). In addition, the low pH facilitates the solubility of metals in water; therefore, these acidic waters tend to have high concentrations of metals. As a result, organisms thriving in such environments must cope with toxic metals in addition to high (≥40°C) or moderately high (37°C–40°C) temperatures and low pH, all of which are lethal to most eukaryotes. Thus, these organisms are referred to as polyextremophiles (Rappaport and Oliverio 2023 ). Presumably because of these requirements, the variety of unicellular eukaryotes found in sulfuric hot springs is, as described below, extremely limited. On the other hand, several types of unicellular eukaryotes have been found in highly acidic mesophilic environments or in neutral environments with high temperatures. So far, analyses of eukaryotic diversity in highly acidic environments have been largely biased toward artificial environments, such as mine-derived streams and lakes, which are highly acidic and generally at moderate or lower temperatures (acid mine drainage; AMD) (Amaral-Zettler et al. 2002 , 2011 , Hao et al. 2010 , Rappaport and Oliverio 2023 ). Previous studies have shown that species diversity declines sharply below pH 3 (Packroff and Woelfl 2000 , Wollmann et al. 2000 ). In such environments, in addition to various lineages of microalgae, heterotrophic organisms such as some ciliates belonging to families Urotricha and Vorticella, heliozoans Actinophrys spp. (class Raphidomonadea, phylum Stramenopila), and amoebae Vahlkampfia spp (class Eutetramitea, phylum Heterolobosea) have been identified (Deneke 2000 , Packroff 2000 , Wollmann et al. 2000 ). Additionally, multicellular organisms, including fungi (Rappaport and Oliverio 2023 ), chironomids, and rotifers (Wollmann et al. 2000 ), have also been found in highly acidic environments. In high-temperature conditions (≥40°C) with weak acidity or a higher pH (>pH 4.0), amoebae such as Acanthamoeba spp. (class Discosea, phylum Amoebozoa) and those belonging to the family Naegleriidae (class Eutetramitea, phylum Heterolobosea), including Naegleria spp., have been identified. However, these organisms are thermotolerant rather than thermophilic, capable of growing at temperatures above 40°C but exhibiting optimal growth at lower temperatures (Reeder et al. 2015 ). As thermophiles with an optimal temperature above 40°C, several amoeba strains belonging to the phyla Amoebozoa and Heterolobosea have been found. These include the amoebozoan Echinamoeba thermarum (class Tubulinea; Baumgartner et al. 2003 ) and Heterolobosean amoebae such as Marinamoeba thermophila (family Tulamoebidae, class Eutetramitea), found in marine hydrothermal vents (De Jonckheere et al. 2009 ); Oramoeba fumarolia (family unassigned, class Eutetramitea), isolated from marine sediment near a fumarole (De Jonckheere et al. 2011 ); and Fumarolamoeba ceborucoi (family unassigned, class Eutetramitea), found in volcanic fumaroles (De Jonckheere et al. 2011 ) (classification of Heterolobosea follows Pánek et al. 2025 ). On the other hand, most thermoacidophiles thriving in sulfuric hot springs are nonphotosynthetic prokaryotes (archaea or bacteria) and eukaryotic microalgae belonging to the red algae class Cyanidiophyceae, which includes the genera Galdieria, Cyanidium, Cyanidiococcus , and Cyanidioschyzon (Walker et al. 2005 , Cho et al. 2023 , Stephens et al. 2024 ). Regarding eukaryotes, Cyanidiophyceae are usually the only group found in hotter regions (37°C–60°C, with an optimal temperature of 40°C–50°C), forming blue-green mats despite being red algae, as their common ancestor lost the red photosynthetic pigment phycoerythrin (Seckbach 1994 ). In cooler regions, however, the habitat of Cyanidiophyceae begins to overlap with that of other eukaryotic algae, such as green algae, euglenids, and diatoms (Gross 2000 , Ferris et al. 2005 ). It is well known that cyanobacteria and other phototrophic prokaryotes are absent in environments with a pH below about 4 (Ward et al. 2012 ). Thus, Cyanidiophyceae had long been recognized as the only phototrophic organisms and eukaryotes found in acidic hot waters. However, later, the photosynthetic euglenid Euglena sp. CRRdV (Sittenfeld et al. 2002 ) and the unicellular green alga Chlamydomonas pitschmannii (Pollio et al. 2005 ) were found in sulfuric hot springs. \n Euglena sp. CRRdV, which is closely related to E. mutabilis , another acid-tolerant euglenoid, was found in an acidic hot mud pool (34°C–45°C; pH 2–4) located near the Rincón de la Vieja volcano (northwestern Costa Rica) (Sittenfeld et al. 2002 ). In the culture, its upper temperature limit for growth was 40°C (Sittenfeld et al. 2002 ). Chlamydomonas pitschmannii was found in the Pisciarelli hot spring located in the hydrothermal system of the Campi Flegrei Caldera (Italy) (Pollio et al. 2005 ). In the culture, it showed a pH tolerance limit of 1.5, with the best growth occurring at a pH between 2.0 and 2.5. The maximum temperature for growth was 42°C, while the optimal temperature was 37°C. Thus, these two algae are thermotolerant (able to survive at high temperatures but not requiring them for optimal growth) rather than truly thermophilic (thriving and growing best at high temperatures), unlike Cyanidiophyceae. The only nonphotosynthetic eukaryotic organism ever found in sulfuric hot springs is the amoeboflagellate Tetramitus thermacidophilus (family Vahlkampfiidae, class Eutetramitea, phylum Heterolobosea) (Baumgartner et al. 2009 , Reeder et al. 2015 ). This organism was found and isolated from three locations around the world: a sample mainly consisting of filamentous bacteria from the Caldera Uzon in Kamchatka (Russia; strain Cu8) (Baumgartner et al. 2009 ), a sediment covered with Cyanidiophyceae in Pisciarelli Solfatara near Naples (Italy; strain Pzc6; both thrive at pH 1.2–5 and 28°C–54°C, with an optimum at pH 3 and 45°C) (Baumgartner et al. 2009 ), and a sample from Boiling Springs Lake (BSL), an acid geothermal feature in Lassen Volcanic National Park in California (strain BSL; optimal growth at 38°C–50°C and pH 2–5), dominated by Cyanidiophyceae (Reeder et al. 2015 ). As far as it has been investigated, T. thermacidophilus feeds on bacteria and archaea present in the environment and does not appear to prey on algae, including Cyanidiophyceae (Baumgartner et al. 2009 , Reeder et al. 2015 ). Regarding the metagenomic analysis of environmental samples from sulfuric hot springs, the presence of some eukaryotic lineages was suggested at moderately high temperatures around 37°C–40°C. However, no evidence was found for the presence of eukaryotic organisms other than those mentioned above at temperatures of 40°C or higher (Brown and Wolfe 2006 , Oliverio et al. 2018 , Rappaport and Oliverio 2023 , Stephens et al. 2024 ). Here, we have investigated whether the scarcity of heterotrophic organisms in sulfuric hot springs, particularly under conditions of ≥40°C and ∼pH 2 (with only T. thermacidophilus having been found so far), is due to their phylogenetic rarity, or whether other lineages might also exist at low frequencies. By investigating samples from three sulfuric acidic hot springs in Japan, seven species of heterotrophic unicellular eukaryotes, belonging to a wide variety of eukaryotic lineages, were found in mats of Cyanidiophyceae, with five preying on them. Among these, four species were capable of growth at pH 2 and 40°C. Although the optimal temperature for all of these species was below 40°C, indicating that they are thermotolerant rather than thermophilic, unlike T. thermacidophilus , these results suggest that several heterotrophic eukaryotic lineages independently acquired resistance to low pH, moderately high temperatures, and probably high metal concentrations in order to colonize niches in sulfuric hot springs.",
"discussion": "Discussion In volcanic sulfuric acidic hot springs, which are characterized by high or moderately high temperatures and extreme acidity, the only known heterotrophic protist had been T. thermacidophilus (class Eutetramitea, phylum Heterolobosea), identified from three locations worldwide and exhibiting optimal growth at pH 3 and 45°C (Baumgartner et al. 2009 , Reeder et al. 2015 ). In this study of sulfuric acidic hot springs (34.7°C–50°C, 34.7°C–50°C, ∼pH 2.0), we identified four additional heterotrophic protists capable of growing or surviving at 40°C and pH 2.0 (Fig. 11 ; Table 2 ): Nuclearia sp. NuKS-1 and Parvularia sp. PaGS-1 (both in class Nucleariidea, phylum Cristidiscoidea), Allovahlkampfia sp. AlKS-1 (class Eutetramitea, phylum Heterolobosea), and Vannella sp. VaKS-1 (class Discosea, phylum Amoebozoa) (Fig. 5 ). Additionally, while their maximum survival temperatures were 35°C and 30°C, respectively (Fig. 9 ; Table 2 ), the ciliate Platyophrya sp. PlGS-1 (class Colpodea, phylum Ciliophora) (Fig. 6 ) and the kinetoplastids Neobodo sp. NbGS-1 and NbTG-1 (class Kinetoplastea, phylum Euglenozoa) (Fig. 7 ) were also found in the same environment. In addition, as discussed below, the results of the phylogenetic analyses suggest that the former four species independently evolved from their respective mesophilic, neutrophilic ancestors through the development of acidophilic and thermotolerant traits. Although the four species could grow at 40°C or at slightly higher temperatures (with a maximum of 42°C for Allovahlkampfia sp. AlKS-1), these organisms are not thermophilic but thermotolerant, with optimal temperatures below 40°C, unlike the thermophilic T. thermacidophilus and the unicellular red algae Cyanidiophyceae (Fig. 9 ). The rarity of thermophiles, but not thermotolerant eukaryotes, inhabiting moderately high-temperature acidic environments is apparently consistent with the results of previous metagenomic analyses (Rappaport and Oliverio 2023 ). Figure 11. Phylogenetic distribution of thermo-acidophilic and thermotolerant acidophilic organisms in eukaryotes. Photosynthetic and heterotrophic species (capable of surviving at ≥40°C and ≤pH 4.0) are shown in green and magenta text, respectively. Magenta and orange circles indicate thermophilic (optimal temperature ≥40°C) and thermotolerant (optimal temperature <40°C) organisms, respectively. Organisms newly identified in this study are underlined. The tree topology is based on a previous paper (Strassert et al. 2021 ). Table 2. Growth temperature and pH, prey organisms used in culture, and phylogeny of unicellular eukaryotes identified in sulfuric acidic hot springs in this study. Species Location condition Culture temp. (°C) Opt. temp. (°C) Culture pH Opt. pH Prey Taxonomy \n Cyanidiococcus spp. All 30–50 40–50 0.5–5.0 2.0 – Cyanidiophyceae Rhodophyta Archaeplastida \n Nuclearia sp. NuKS-1 Kusatsu P2 50.0 °C pH 2.3 25–41 35 1.8–7.0 3.0 \n Cyanidiococcus , E. coli Nucleariidea Cristidiscoidea Obazoa \n Parvularia sp. PaGS-1 Goshogake P1 38.1 °C pH 2.45 25–40 25 1.2–7.0 2.0 \n Cyanidiococcus , E. coli Nucleariidea Cristidiscoidea Obazoa \n Allovahlkampfia sp. AIKS-1 Kusatsu P1 45.7 °C pH 2.07 25–42 30 1.6–7.0 5.0 \n Cyanidiococcus , E. coli Eutetramitea Heterolobosea Discoba \n Vannella sp. VaKS-1 Kusatsu P1 45.7 °C pH 2.07 25–40 35 1.4–5.0 3.0 \n Cyanidiococcus , E. coli Discosea Amoebozoa \n Platyophyra sp. PIGS-1 Goshogake P2 34.7 °C pH 2.28 25–35 30 2.0–7.0 3.0 \n Cyanidiococcus \n Colpodea Ciliophora Alveolata \n Neobodo sp. NbGS-1 Goshogake P2 34.7 °C pH 2.28 25–30 30 2.0–7.0 2.0–5.0 \n Cyanidiococcus , E. coli Kinetoplastea Euglenozoa Discoba \n Neobodo sp. NbTG-1 Tamagawa P1 42.5 °C ND ND ND ND ND ND Kinetoplastea Euglenozoa Discoba “–” indicates that no prey was used for cultivation, and “ND” indicates that it was not determined. In the taxonomy column, the class (top), phylum (middle), and, if applicable, supergroup (bottom) are indicated. Thus far, Nuclearia species have been isolated from environments with moderate or lower temperatures (<37°C) and neutral to slightly acidic pH levels (≥pH 4) (Fig. 3d and i ; e.g. Dirren and Posch 2016 ). However, environmental metagenomic analyses of the highly acidic, red-colored AMD in Río Tinto (15°C–25°C, pH 2.0; Spain) (RT10 5E 39; Amaral-Zettler et al. 2002 , 2011 ), and AMD in Xiang Mountain (30°C, pH 3.0; China) (AMD-Euk-14; Hao et al. 2010 ) identified Nuclearia spp. from each location. Of these, AMD Euk-14 (Hao et al. 2010 ) is most closely related to Nuclearia sp. NuKS-1 (Fig. 3d ). Thus, the result suggests that the common ancestor of Nuclearia sp. NuKS-1 and AMD-Euk-14 evolved from a mesophilic, neutrophilic ancestor by acquiring thermotolerant and acidophilic traits. Regarding the genus Parvularia , only one strain, Parvularia atlantis ATCC 50694, isolated from a lake in Atlanta, has been described (López-Escardó et al. 2018 ). However, environmental metagenomic analyses of the AMD in Río Tinto (Amaral-Zettler et al. 2002 , 2011 ) identified three clones of Parvularia (RT5iin16; O138306E03; RT5iin14; Fig. 3i ). Although it remains unclear when Parvularia sp. PaGS-1 acquired acidophilic and thermotolerant capabilities during evolution, the phylogenetic tree pattern and the distribution of organisms across various environments (Fig. 3d and i) suggest that Nuclearia sp. NuKS-1 and Parvularia sp. PaGS-1 independently evolved acidophilic and thermotolerant traits. Thus far, Heterolobosean Allovahlkampfia species have been isolated from soil, temperate freshwater, or tree bark, and they feed on bacteria. However, all of these species, except for Allovahlkampfia sp. AlKS-1, as far as their isolation environments have been documented, have been found in environments with moderate or lower temperatures (<37°C) and neutral to slightly acidic pH levels (≥pH 4) (Fig. 4e ; e.g. Geisen et al. 2015 ). In culture, none of these strains, except for Allovahlkampfia sp. AlKS-1, has shown growth at 37°C; some have grown at 30°C, while others have grown only at lower temperatures (Geisen et al. 2015 ). The thermoacidophilic amoeboflagellate T. thermacidophilus (Baumgartner et al. 2009 , Reeder et al. 2015 ) also belongs to class the Eutetramitea but is classified in the family Vahlkampfiidae, unlike Allovahlkampfia , which belongs to the family Acrasidae (Pánek et al. 2025 ) (Fig. 4e ). Thus, Allovahlkampfia sp. AlKS-1 has evolved acidophilic and thermotolerant traits independently from T. thermacidophilus . Several species of amoeba Vannella have been isolated from soil, freshwater, and seawater. To date, except for Vannella sp. VaKS-1, there have been no reports of their occurrence in high or moderately high temperatures (>37°C) or highly acidic environments (<pH 4.0) (Smirnov et al. 2007 , Kudryavtsev et al. 2021 ). However, the phylogenetic analysis showed that Vannella sp. VaKS-1 is included in a clade to which many environmental metagenomic sequences are assigned (Fig. 5d ). Among them, four sequences were obtained from AMD in Río Tinto [15°C–25°C, pH 2.0; O218406H12; O127706H07; B324106B01; O138006A05: (Amaral-Zettler et al. 2011 )], while two sequences were from surface waters off the coast of the northeastern Red Sea (25°C at the time of sampling; RS.01f.10 m.23; RS.01f.10 m.11; Acosta et al. 2013 ). Thus, acid tolerance may have been acquired by the common ancestor of this clade or independently established in each subgroup at a later stage. As for the thermotolerance, Vannella sp. VaKS-1 may have acquired it relatively recently after diverging from other members. The ciliates belonging to the class Colpodea are widespread in various environments including, terrestrial habitats, such as soil, leaf litter, tree holes, freshwater, and seawater (Foissner 1993 , Foissner et al. 1999 , Dunthorn et al. 2009 , Vd'ačný and Foissner 2019 ). The phylogenetic analysis showed that Platyophrya sp. PlGS-1 is closely related to P. vorax and P. spumacola which have been found in environments ranging from acidic (pH 2.5–4) to neutral soils (Foissner 2000 ) (Fig. 6e ). In addition, P. bromelicola , which formed a monophyletic clad with the three species (Fig. 6e ), was isolated from tank water of a bromeliad tree (Foissner and Wolf 2009 ), which is generally acidic, with a pH between 4.0 and 6.5 (North et al. 2023 ). Thus, it is likely that among Colpodea, Platyophrya has adaptively evolved to acidic environments. The genus Neobodo , along with other free-living bodonids and obligate parasitic trypanosomatids, belongs to the class Kinetoplastea of phylum Euglenozoa (Adl et al. 2019 ). Neobodo species are bacterivores that inhabit a wide variety of environments, including freshwater, seawater, and soils (von der Heyden and Cavalier-Smith 2005 , Flegontova et al. 2018 ). The phylogenetic analysis showed that both Neobodo sp. NbGS-1 and Neobodo sp. NbTG-1 are closely related to an environmental genomic sequence from AMD in Río Tinto (15°C–25°C, pH 2.0; Spain; Uncultured eukaryote O138106G09) (Amaral-Zettler et al. 2011 ) and to another sequence from thermal and acidic green biofilms in a fumarole (Mexico; Uncultured Bodonidae METASED30; temperature and pH information not provided). These four strains are related to Neobodo sp. G97, whose sequence was detected by PCR from the gut of a tsetse fly and is presumed to have originated from environmental water (Votýpka et al. 2021 ). Other Neobodo strains have been isolated from nonacidic environments (Fig. 7 ; e.g. von der Heyden and Cavalier-Smith 2005 ). These results suggest that the common ancestor of Neobodo spp. inhabiting sulfuric hot springs in Japan and AMD in Río Tinto evolved into an acidophile from a neutrophilic ancestor. In this study, except for Neobodo spp., each organism or its closely related species was found at only one of the three sulfuric acidic hot springs. Furthermore, previous environmental DNA-based metagenomic analyses have not identified any closely related species of these organisms. However, their environmental population density was extremely low compared to Cyanidiophyceae; thus, the survey and sampling were probably far from saturation. Further investigations are needed to determine whether these organisms are widely distributed in sulfuric acidic hot spring environments, similar to Cyanidiophyceae. Nevertheless, all the organisms that morphologically matched those that proliferated in the Cyanidiophyceae blue-green mat when incubated at 40°C or at room temperature (Fig. 2 ) were successfully cultured and analyzed in this study. In our sampling, after incubation of the algal mats, a greater variety (Fig. 2 ) and a higher number of presumed heterotrophic eukaryotic cells were observed in the samples that were transported and incubated at room temperature compared to those transported and incubated at 40°C. This is probably because the optimal temperature for these organisms was lower than 40°C, as shown by their isolated cultures (Fig. 9 ). For some organisms, especially those isolated from Kusatsu Hot Spring, the reason why the water temperature at the sampling site exceeded the survival limits observed in their isolated cultures (Table 2 ) is still unclear. During sampling, we submerged the collected Cyanidiococcus mats in nearby spring water. We also collected and observed only the respective spring waters, but we did not find any organisms in the water samples, suggesting that the isolated organisms were in the algal mats. It is likely that some organisms temporarily entered from surrounding cooler areas (e.g. the edges of the flow). However, we cannot rule out the possibility that factors such as the structure of the mat (e.g. biofilms) or coexistence with other organisms may raise the upper temperature limit for survival. In this regard, the isolated Vannella sp. VaKS-1 did not proliferate at 40°C when fed with E. coli , but it was able to grow when fed with Cyanidiococcus under our culturing conditions. The thermoacidophilic T. thermacidophilus feeds on archaea and bacteria but does not appear to prey on Cyanidiophyceae (Baumgartner et al. 2009 , Reeder et al. 2015 ). To date, no predators of Cyanidiophyceae have been reported. However, the heterotrophic protists found in this study were isolated from Cyanidiophyceae mats (Fig. 1 ), primarily composed of Cyanidiococcus , and were found to feed on Cyanidiococcus (except for Neobodo , whose feeding behavior remains unclear) (Figs 3 – 7 and 10 ). These newly identified organisms, other than the ciliate Platyophrya sp. PlGS-1, were also able to grow by feeding exclusively on E. coli (Figs 8 and 9 ), suggesting that they likely consume bacteria and archaea in their natural environment as well. Additionally, although not analyzed in detail in this study, a rotifer isolated from the environment (Fig. 2f ) was also able to grow by feeding exclusively on Cyanidiococcus . The seven heterotrophic organisms identified in this study are all acidophiles, with optimal pH ranges between 2 and 5 (Fig. 8 ). However, unlike the unicellular red alga Cyanidiococcus , which was found in the same environment and can grow at pH 1.0 and even pH 0.5 (Fig. 8 ), these heterotrophs are not as highly specialized for extreme acidity and fail to survive at such low pH levels (Fig. 8 ). In terms of temperature, while Cyanidiococcus has a maximum growth temperature of 50°C, the heterotrophic organisms identified in this study have maximum growth temperatures below 40°C (Fig. 9 ). Among them, four species are thermotolerant, capable of marginal growth or survival at 40°C or at slightly higher temperatures, but they are not thermophiles like Cyanidiococcus (Fig. 9 ; Table 2 ). Even though their cells are not specifically adapted to conditions such as pH 2.0 and 40°C, Cyanidiophyceae, their primary prey, remains abundant throughout the year. Thus, it is presumed that the population size of these heterotrophic unicellular eukaryotes is limited solely by abiotic environmental factors such as temperature and pH. Additionally, since there are few competing organisms for prey and few predators targeting them, the overall environment may still be favorable for their survival. This study revealed that various lineages of heterotrophic unicellular eukaryotes have independently developed acidophilic and thermotolerant traits, enabling them to colonize moderately high-temperature, extremely acidic sulfuric hot springs (Fig. 11 ). To elucidate the mechanisms and evolutionary processes underlying these adaptations, future research involving genomic, physiological, and structural analyses will be necessary. To this end, the information and cultures provided by this study serve as valuable resources for future investigations."
} | 6,765 |
38632231 | PMC11024123 | pmc | 6,907 | {
"abstract": "Endowing textiles with perceptual function, similar to human skin, is crucial for the development of next-generation smart wearables. To date, the creation of perceptual textiles capable of sensing potential dangers and accurately pinpointing finger touch remains elusive. In this study, we present the design and fabrication of intelligent perceptual textiles capable of electrically responding to external dangers and precisely detecting human touch, based on conductive silk fibroin-based ionic hydrogel (SIH) fibers. These fibers possess excellent fracture strength (55 MPa), extensibility (530%), stable and good conductivity (0.45 S·m –1 ) due to oriented structures and ionic incorporation. We fabricated SIH fiber-based protective textiles that can respond to fire, water, and sharp objects, protecting robots from potential injuries. Additionally, we designed perceptual textiles that can specifically pinpoint finger touch, serving as convenient human-machine interfaces. Our work sheds new light on the design of next-generation smart wearables and the reshaping of human-machine interfaces.",
"introduction": "Introduction Perception is a crucial function of human skin. While flexible film-based electronics have been developed as electronic skins 1 – 6 , textile-based electronics can provide superior flexibility, air permeability, and comfort for wearable electronics 7 – 10 . Integrating perceptual function into textiles will revolutionize the way humans interact with electronic devices and contribute to the further development of smart wearables. Recent advances in electronic textiles have enabled communication, sensing, display, power supply, and other functions 11 – 14 . However, perceptual textiles capable of exclusively and accurately pinpointing human touch and helping humans/robots recognize and respond to dangers, such as fire, water, and fracture, have yet to be reported. To address this challenge, flexible, strong, and conductive fibers are required as the basic units. While metallic wires and nanocarbon-based fibers offer good conductivity, they are prone to electrical and mechanical failures under tension or cyclic bending deformation 15 – 17 . Geometric designs, such as serpentine, kirigami, or wrinkled structures, can impart certain stretchability to fibers 18 – 21 , but pose challenges in terms of compatibility with textiles and increase the risk of failure and cost. Alternatively, ion gels feature inherent flexibility, conductivity, transparency, and robustness 22 , 23 . They can transfer electricity through mobile ions, similar to biological tissues 24 , 25 . Therefore, robust gel fibers that are biocompatible, mechanically strong, and highly ionic-conductive have the potential to be woven into the desired intelligent textiles. In this work, we prepared a highly strong, conductive, and stable silk fibroin-based ionic hydrogel (SIH) fiber and realized the fabrication of intelligent perceptual textiles capable of precisely detecting external dangers and human touch. The SIH fiber, composed of natural silk fibroin, ionic liquid ([Emim]BF 4 ), and glycerol, was prepared through a continuous wet spinning process. It has a semi-crystalline and oriented structure similar to natural silkworm silk, leading to a high tensile strength of 4 MPa, which can be further increased to 55 MPa via post-stretching. The SIH fiber also possesses a notable extensibility of up to 530%, more than 20 times that of natural silkworm silk (~25%), which can be ascribed to the plasticization of ionic liquid [Emim]BF 4 , glycerol, and water. Importantly, the incorporation of [Emim]BF 4 endows the SIH fibers with stable and high ionic conductivity up to 0.45 S·m –1 . Furthermore, we demonstrated the application of SIH fibers in perceptual textiles. The circuits with integrated SIH fibers can show instantaneous and characteristic responses to stimuli such as fire burning, water immersing, sharp object cutting, and finger touching, endowing the SIH fiber-based intelligent textiles with sensing capabilities for protection. Besides, we demonstrated that the designed perceptual textiles are capable of precisely and specifically detecting the occurrence and location of touch. Therefore, the successful fabrication of silk fibers with intrinsic ion conductivity and excellent mechanical properties will promote a significant advancement in the functionalization and utilization of silkworm silk fibers. Furthermore, the development of perceptual textiles capable of exclusively and accurately pinpointing human touch will revolutionize human-machine interfaces, offering excellent flexibility and comfort, bringing great convenience in intelligent living to humans.",
"discussion": "Discussion In summary, we have developed a revolutionary perceptual textile capable of sensing dangers and human touch, utilizing a SIH fiber with stable ionic conductivity, high strength, excellent extensibility, and good flexibility. The remarkable fiber was prepared through a continuous wet spinning process, and comprised of natural derived fibroin, ionic liquid, and glycerol. Due to its semi-crystalline and oriented structure, the fiber shows an unparalleled strength of up to 55 MPa, surpassing that of other reported ionic hydrogel fibers (less than 10 MPa). The hydrogelation enables the fiber to elongate up to 530%, which is 20 times higher than that of nature silk fibers. The incorporated [Emim]BF 4 endows the fiber with excellent and stable conductivity of 0.45 S·m –1 . Based on the superior mechanical properties and conductivity of the SIH fiber, we have designed and fabricated perceptual textiles. The textiles can help humans or robots electrically respond to potential hazards, such as fire, water, and sharp objects. Besides, the perceptual textiles integrated a surface-capacitive touch circuit system can detect and locate finger touch, making them ideal for use in wearable human-machine interfaces. Our perceptual textiles have been successfully used to remotely control a robot hand, showcasing their potential in applications such as remote control and communication. The perceptual textiles developed in this work represent a new and comfortable human-machine interface. Given the impressive mechanical properties, stable and good conductivity of the SIH fiber, along with the sustainability, biosafety, and biodegradability of its precursor materials, we anticipate that it will serve as an excellent candidate material for the fabrication of other functional units and thus contribute to development of human-friendly, comfortable and all-fiber-based integrated intelligent wearable systems."
} | 1,654 |
39934172 | PMC11814417 | pmc | 6,908 | {
"abstract": "DNA nanotechnology and especially the DNA origami method are primal tools to create precise nanoscale objects. For DNA origami, a long ssDNA scaffold strand is folded by a multitude of smaller staple strands into base-pair accurate shapes, allowing for precise modification and incorporation of guest molecules. However, DNA origami are limited in size, and thus is the area that can be controlled with nanoscale precision. Prior methods of creating larger assemblies were either costly or lacked structural control. Here, we incorporate two methods of modularity into one exemplary modular DNA origami (moDON). The modularity allows for the creation of over 50,000 diverse monomers and subsequently the assembly of a plethora of fully addressable designer superstructures while keeping the construction cost very low. The here-introduced methods for modularity in DNA origami design offer an efficient, cost-effective solution for constructing precisely organized, and fully addressable structures on a variety of scales.",
"introduction": "Introduction The concept of modularity is widely used in a plethora of scientific and engineering disciplines 1 . Modularity divides structures or systems into subunits, that can be independently addressed and changed, without affecting the rest of the structure or system. Subunits may even be reusable for different structures or systems. This allows for low-effort, cost-effective assembly of, e.g. buildings 2 , hardware (through standardization of subparts), machine code (also: object-oriented programming) or genetic code 3 . In nature, well-defined subunits of only a few nm in size 4 assemble into higher-order structures 5 , 6 with sizes ranging from a few hundred nanometers (nm) to several micrometers (µm). As synthetic biology progresses, researchers aim to mimic such structures, both on a molecular and on a cellular level. Therefore, the need for modular building blocks capable of controllably assembling and disassembling into and from higher-order structures on biologically relevant time scales has become more urgent. Structural DNA nanotechnology has become a widely used construction approach in synthetic biology. Especially the DNA origami 7 technique, where a long, circular scaffold strand, extracted from the M13mp18 bacteriophage, is folded into any desired 2D or 3D shape using short synthetic staple strands, presents an indispensable tool for nanoscale engineering. DNA origami nanostructures (DONs) ranging from just a few nm to over a hundred nm can be easily designed and synthesized, with freely available CAD software such as caDNAno 8 . To date a large variety of DONs have been published, ranging from the simplest 2D structures 7 to functional DNA origami motors 9 . A very attractive feature of DONs is their complete addressability for guest molecule placement with base pair (bp) precision, enabling e.g. the formation of fluorescent nanorulers 10 , plasmonic nanosensors 11 , 12 , nm-sized force sensors 13 , 14 , or to study complex biological questions 15 – 17 . Despite all of these advantages, a bottleneck in DON design is the overall final size limit, determined by the scaffold strand (the M13mp18 genome is made up of 7249 bases). Although different insertions have led to scaffold sizes of close to 9000 bases, the overall size increase in 3D DONs is marginal. Therefore, a variety of different strategies have been reported to increase DON sizes. These include the use of a lambda/M13 hybrid phage as a scaffold source 18 , the development of orthogonal scaffold strands 19 , the use of unscaffolded structures 20 – 22 , origami slats 23 , 24 , or hierarchical assembly into finite 25 , 26 or periodic 26 – 28 structures. Each of these methods has at least one major drawback: With the lambda/M13 hybrid phage, synthesis of scaffolds larger than 50 000 bases 18 , equivalent to ~six regular origami was achieved. A similar size was achieved with superstructures from orthogonal scaffolds, reaching sizes equivalent to five regular origami. However, both approaches have similar bottlenecks: Firstly, the effort for scaffold production is increased tremendously and secondly, the unique scaffold sequence requires the same amount of unique staples to be designed and synthesized individually, increasing effort and monetary cost proportionally to superstructure size. The latter problem also affects unscaffolded nanostructures employing single-stranded tile (SST) assembly, where every ssDNA tile needs to be designed and synthesized individually. Nonetheless, using 10,000 unique ssDNA tiles, superstructures in the Gigadalton (GDa) scale have been constructed 22 . The largest superstructures reported to date were constructed using so-called DNA origami slats 24 , through connection of several long 6 or 12 helix bundles (HBs) via well-positioned, complementary ssDNA handles. These structures reached several µm and GDa in size and their synthesis is comparably cost effective. However, assembly is so far limited to planar 2D structures with low degrees of complexity in assembly. Yet it is the most efficient strategy reported to date for these kinds of large, flat and fully addressable superstructures. Another approach is the construction of superstructures with ssDNA connectors 29 – 32 . With this approach, different 3D shapes of superstructures can be constructed, but each connection requires many unique ssDNA connectors, making it susceptible to unwanted interactions and incomplete connections. GDa sizes can also be reached through hierarchical assembly, where successively larger structures are assembled in multiple subsequent steps. This approach has led to structures measuring 0.5 µm (2D), depicting a μm-sized image of the Mona Lisa 33 , while 3D structures could reach up to the GDa scale 25 . The high structural control and potential size make it an excellent approach for the formation of superstructures. Nevertheless, design and construction are faced with different challenges. For the hierarchical assembly of superstructures, connectivity always faces one of two problems: On the one hand, assemblies of self-complementary DONs lead to repetitive, homomultimeric finite 25 or periodic 26 , 27 superstructures. This approach is simple and cost-effective, as in principle only one type of monomer is required. However, it has the great disadvantage of an overall loss of structural control of the superstructure beyond the repetitive subunit, as only form and size of the subunit is controlled, yet superstructure formation occurs without control over the final structure. Consequently, it is impossible to controllably retain the site-specific addressability of the structure beyond the single monomer or subunit. On the other hand, assemblies made up of many different DONs, each complementary to one another, but not themselves, yield finite, heteromultimeric superstructures 9 , 19 . With this approach the structural control over the superstructure is high and the site-specific addressability is retained. However, the downsides are the requirement for a multitude of different DONs, leading to high design effort and increasing costs, proportional to superstructure size, and/or the need to be assembled in several steps, resulting in a tremendous decrease in overall yield. Therefore, currently, concepts combining both low cost and high structural versatility are missing. Such concepts could alleviate the issues associated with traditional hierarchical assembly and allow for the controlled assembly of large superstructures, or a multitude of orthogonal smaller structures, with retained site-specific addressability across the entire superstructure. In response to this challenge, we developed two methods for modularity and combined them in one structure, the moDON, a single modular DNA origami formed from one set of staples, but capable of folding into tens of thousands of different connection configurations. The large number of orthogonal connection sites allows us to increase the number of monomers in one-pot assemblies, circumventing the extraction and purification of intermediates as generally required in traditional hierarchical assembly. We were able to controllably assemble and disassemble manifolds of moDONs into large superstructures as well as smaller, more complex structures, in the xy- as well as in the z-plane. We engineered modularity into two different connection methods. The first method of modularity, in xy-direction, was created for shape-matching connection sites 25 , 26 . We achieved the multitude of connection sites by engineering a re-routable scaffold. Until now DNA origami was designed under the paradigm of creating one single, fixed scaffold routing for each structure. With this method alone, we were able to construct hundreds of individual moDON monomers with different shape-matching connection sites. The second method of modularity is a three-strand connection system in the z-direction. Here we engineered orthogonal binding sites with well-positioned, directional, and optimally sized handles, circumventing the problems occurring in other three-strand-systems, lacking rigidity, losing control over connection stoichiometry, and/or the impractical use of many different unique strands for each connection. The z-connections were engineered to be directional, non-branching, not self-passivating, removable, rigid, orthogonal, with a low number of duplexes per connection, and only made up of one single DNA sequence per connection site. Together with the xy-connections, this allowed for the construction of > 50 000 unique moDON monomers. We demonstrate the construction of arbitrary finite superstructures with connections in three dimensions. Additionally, we created periodic structures with monomeric and multimeric repetitive subunits reaching more than 1 GDa in size. Superstructures could be assembled in one-step reactions from monomers, without the need for purification and isolation of intermediate products, which is required in traditional hierarchical assembly. This greatly simplifies construction and simultaneously increases the overall yield. Further, we show that xy- and z-connections are fully orthogonal allowing for parallel and selective assembly and disassembly of each individual connection using different orthogonal triggers (see Figure S1 for overview). Finally, we show that the site-specific addressability of each monomer in the superstructure is fully retained by placing gold nanoparticles (Au NPs) at specific positions in the superstructures, demonstrating a highly controllable, efficient, and cost-effective strategy for the formation of large and orthogonal superstructures.",
"discussion": "Discussion In summary, we developed re-routable scaffolds as design paradigm for DNA origami to introduce modularity to shape-matching connections and developed a modular three-strand-system that circumvents the loss over structural control, dynamic convertibility, and the need for many unique connectors, which were previously problematic for these connections. We showcased the power of modularity in DNA origami design by engineering one moDON, with tens of thousands of different connection site configurations. The large number of orthogonal connections allows one-step assemblies into a diverse variety of superstructures. These can be finite (super-)structures with fully retained site-specific addressability, or periodic structures reaching large scales of µm and GDa. We introduced modularity to two connection strategies 26 , 27 , 35 . Modularity in xy-direction was achieved via scaffold re-routing, subverting the traditional way of DNA origami design of rigid scaffold routings. Modularity in z-direction was achieved with well-positioned, directional, and optimally sized handles for an orthogonal three-strand-system, which circumvents all issues of this kind of connection: lack of rigid, unintended passivation, branching, non-removable connections, lack of directionality, and/or the need for many unique connections strands for each connection. With six positions in the xy-direction, each being able to form one of two connection sites (α-ζ) or be passive, and two opposite positions in the z-direction being able to form one of eight connection sites or be passive, the total number of unique monomers that can be formed through the modularity of just one single moDON is: \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$\t {(2\\,{{\\rm{xy}}}\\; {{\\rm{configs}}}.+{{\\rm{passive}}})}^{6{{\\rm{sites}}}} \\, * \\,{(8\\,{{\\rm{z}}}\\; {{\\rm{configs}}}.+{{\\rm{passive}}})}^{2 \\, {{\\rm{sites}}}} \\\\ \t=59049 \\, {{\\rm{unique\\; monomers}}}$$\\end{document} ( 2 xy configs . + passive ) 6 sites * ( 8 z configs . + passive ) 2 sites = 59049 unique monomers With 59 049 unique monomers possible from a single moDON set of staples, our method significantly reduces cost (total number of staples needed are only ~ 2.1 × the number needed for a single structure (Table S18 )), and design time compared to individual design of each DNA origami. To change the configuration of one connection site only ~ 2.5 % of the staples need to be exchanged. The modular construction of DNA origami significantly decreases the cost and effort for superstructure design whilst having a large versatility in possible assemblies. However, the modularity alone does not introduce fundamentally different kinds of DNA-DNA connections and is not able to fully solve existing issues with respective connection strategies. Connections based on protrusions and indentations have a low stability, limited by the number of base pairs in each connection. An increase of bp in the connection, on the other hand, could undermine structural stability or facilitate the formation of topological traps. Extending the number of DNA layers in the connections from two to four 28 could also increase the stability but would increase the complexity of the modular design. Protrusions and indentations are also limited by the amount of helix configurations, while still avoiding symmetries and similarities of connection sites, as well as the formation of overhanging scaffold loops, prone to off-target interactions. The rigidity of z-connections is prone to missing connector strands. In general, unintended and unpredictable interactions between ssDNA strands become more likely with an increasing number of nt involved. Unfortunately, this presents a universal problem for all DNA-DNA connections, and an efficient way to overcome these limitations is yet to be found. However, compared to other methods of superstructure assembly, the modular design excels in versatility, cost-effectiveness, and/or ease of handling. By increasing the number of connection sites and configurations thereof, we were able to increase the number of monomers in one-step reactions, increasing the size of superstructures constructed in one-pot assemblies. Further, the cost for the superstructures is decreased tremendously, since each of the diverse origami configurations is a variation of the same moDON. Both factors combined alleviate the problems of high cost, lack of versatility, and yield loss through repeated extraction and purification of intermediates in traditional hierarchical assembly. We showed further, that the moDON superstructures can also be hierarchically assembled, enabling the construction of even larger, more complex structures. However, 3D SST assemblies 20 – 22 exceed all other published structures to date in versatility and addressability, including ours, as every tile is separately adjustable and addressable. The largest fully addressable SST structures reached ~ 0.5 GDa 22 , ~ 10 times larger than the fully addressable structures here, but the monetary cost, on the other hand, was far beyond $ 100,000. Here, modular-designed origami are much more cost-effective and easier to handle than the SST structures from 30,000 ssDNA oligomers. The largest fully addressable structures achieved to date are origami crisscross assemblies 24 , reaching several µm and several GDa for structures in 2D. The size and weight of our fully addressable assemblies is far below those. Here, however, the moDON assemblies excel in the variety of scales and forms they can be assembled into (e.g. twisted and curved structures), and also disassembled from. Other three-strand systems 27 , 29 – 32 (to the here presented z-connections) were able to connect origami to various forms. But those connection strategies need a large number of unique, long ssDNA strands, and/or do not have control over the connection properties, like stoichiometry or rigidity, whereas the here presented z-connections only require one single kind of 21nt-short ssDNA strand for each stable, directional, orthogonal, and dynamically controllable connection, decreasing the risk of unwanted interactions and increasing the number of orthogonal connections designable. Using only one unique z-connector per connection site, to control the dynamic assembly and disassembly process makes this approach also feasible for incorporation into strand displacement networks, unlike structures needing dozens of unique sequences. Further, the small number of ssDNA nucleotides involved in each modular xy-connection in combination with the structurally orthogonal indentations and protrusions makes this design strategy immune to off-target interactions. Additionally, the here constructed moDON folds with quantitative yield, which leads to a comparably higher effective yield of the superstructures, a trait that is often overlooked and a prominent issue in many structures reported thus far for (hierarchical) assembly. In its capacity to assemble and disassemble into and from a large variety of shapes and scales on biologically relevant size- and timescales, the moDON could have great potential in synthetic biology. For example, the long tubular superstructures (Figs. 3 g and 4c ) could lend themselves as cytoskeleton-mimicking structures, where assembly and disassembly could be controlled by DNA-strand displacement circuits 36 , 37 . Especially the z-connections with their minimized number of connection strands, and still retained structural control are suited for the dynamic assembly and disassembly in structures. Compared to DNA nanotubes, mostly used to mimic the natural cytoskeleton in synthetic biology, the here constructed periodic tubes have a very large subunit, predisposing them for the incorporation of more complex approaches to multifunctionality, e.g. molecular motors 9 that could alternatively also be used for the construction of artificial flagella in synthetic biology, or hinged elements 38 . The here presented methods of modularity should prove a powerful addition in the toolbox of DNA origami construction. We expect the xy-z orthogonal modularity and connections design to be adaptable to any other DNA origami structure, not only the moDON, potentially even allowing for further expansions of orthogonal connection sites, and presenting itself as an addition or even alternative to hierarchical assembly. All in all, these features present the modular approach to DON superstructure construction as a promising tool for the construction of large superstructures with high control, suitable for synthetic biology applications."
} | 4,883 |
39065414 | PMC11280047 | pmc | 6,909 | {
"abstract": "Biological nitrogen fixation in legume plants depends on the diversity of rhizobia present in the soil. Rhizobial strains exhibit specificity towards host plants and vary in their capacity to fix nitrogen. The increasing interest in rhizobia diversity has prompted studies of their phylogenetic relations. Molecular identification of Rhizobium is quite complex, requiring multiple gene markers to be analysed to distinguish strains at the species level or to predict their host plant. In this research, 50 rhizobia isolates were obtained from the root nodules of five different Pisum sativum L. genotypes (“Bagoo”, “Respect”, “Astronaute”, “Lina DS”, and “Egle DS”). All genotypes were growing in the same field, where ecological farming practices were applied, and no commercial rhizobia inoculants were used. The influence of rhizobial isolates on pea root nodulation and dry biomass accumulation was determined. 16S rRNA gene, two housekeeping genes recA and atpD, and symbiotic gene nodC were analysed to characterize rhizobia population. The phylogenetic analysis of 16S rRNA gene sequences showed that 46 isolates were linked to Rhizobium leguminosarum ; species complex 1 isolate was identified as Rhizobium nepotum , and the remaining 3 isolates belonged to Rahnella spp., Paenarthrobacter spp., and Peribacillus spp. genera. RecA and atpD gene analysis showed that the 46 isolates identified as R. leguminosarum clustered into three genospecies groups (B), (E) and (K). Isolates that had the highest influence on plant dry biomass accumulation clustered into the (B) group. NodC gene phylogenetic analysis clustered 46 R. leguminosarum isolates into 10 groups, and all isolates were assigned to the R. leguminosarum sv. viciae .",
"introduction": "1. Introduction Field pea ( P. sativum L.) is one of the most popular legume plants in crop rotations in Lithuania [ 1 , 2 ]. Legume plants’ inclusion in rotations offers a multitude of advantages which are significant to agricultural productivity and sustainability. Legume plants’ cultivation enhances soil quality through various mechanisms. These include the improvement of soil structure, which promotes better water infiltration and retention [ 3 , 4 ]. It also breaks pest cycles and makes weed control easier [ 5 , 6 , 7 ]. Additionally, legume plants such as field peas improve the growth of soil microbial biomass which plays a crucial role in nutrient cycling and overall soil fertility [ 8 ]. One of the most significant benefits of incorporating legume crops into rotations is their ability to form symbiosis with rhizobia, fix atmospheric nitrogen and increase the nitrogen accumulation in the soil. This natural process reduces the need for mineral nitrogen fertilizers in subsequent non-legume crops [ 6 , 7 , 9 ]. In general, the integration of field pea or other legume plants into crop rotations helps implement the European Union (EU)’s strategies of “Bringing nature back into our lives” [ 10 ] and “The Farm to Fork” [ 11 ], which aim to reduce pesticides and fertilizers in European agriculture. The amount of nitrogen fixed by pea plants can vary due to many factors such as soil’s chemical and physical properties, environmental conditions, weed and insect infestation, crop rotation and the rhizobia population [ 12 , 13 , 14 , 15 ]. Rhizobia population can vary due to the same factors listed above, however, biotic factors, such as the competitiveness of microorganisms and commercial inoculants, affect the Rhizobium population as well [ 16 , 17 ]. Rhizobium spp. includes more than 90 species, which can nodulate different legume plants [ 18 ]. The R. leguminosarum species was initially regarded as the sole species able to establish symbiosis with P. sativum . R. leguminosarum includes different symbiovars nodulating different legume plants. For example, R. leguminosarum sv. viciae forms a symbiosis with plants of the Viciaea tribe, like vetches ( Vicia L.) and pea ( P. sativum L.), R. leguminosarum sv. trifolii nodulates clovers (i.e., Trifolium L.), and R. leguminosarum sv. phaesolii infects common beans ( Phaseolus vulgaris L.) [ 18 ]. However, more species were introduced as symbionts of the legume tribe Vicieae , such as R. fabae [ 19 ], R. pisi [ 20 ], R. lentis [ 21 ], R. binae [ 21 ], R. bangladeshense [ 21 ], and R. anhuiense [ 22 ]. The population of rhizobiain pea fields remains poorly documented, especially in Lithuania and under organic farming conditions. This study aimed to analyze the rhizobia population and genetically characterize rhizobial strains isolated from different organically grown pea genotypes, and evaluate the strain impact on plant nodulation and biomass accumulation. Five different pea genotypes selected for a wide range rhizobia isolation: “Respect”, “Astronaute”, “Bagoo”, “Lina DS” and “Egle DS” were characterized by abundant grain yield, resistance to loading and disease infection. Lina DS” and “Egle DS” are new and less studied Lithuanian pea genotypes listed in national and EU variety lists in 2021. Different genes 16S rRNA, atpD, recA and nodC were analysed for rhizobial isolates’ genetic identification.",
"discussion": "3. Discussion The 50 bacterial isolates were isolated from five different pea genotypes, which were growing in the same field. Soil chemical analysis showed that the humus content in soil was low at 2.84%, however, other soil parameters were suitable for pea growth and nutrient assimilation [ 23 , 24 ]. Mobile phosphorus and potassium quantity in analysed soil was high, and soil pH was near neutral, so these minerals were not fixated into insoluble compounds and they were available for the peas [ 25 ]. Mineral nitrogen content in the soil was not high at 9.82 mg kg −1 or 47.1 kg ha −1 [ 26 ]. This amount of nitrogen should have less or no negative impact on the nodulation of different pea genotypes. A pea nodulation test showed that not all isolates formed nodules repeatedly. No nodules were formed on pea roots where isolates RSP03, RSP06, RSP09, ASTR04, EGLE01 and BAGOO04 were inoculated, however, plant biomass analysis showed that isolates ASTR04, RSP03 and RSP06 had a significant positive influence on dry plant biomass formation. Genetic analyses revealed that these three isolates ASTR04, RSP03 and RSP06, isolated from “Astronaute” and “Respect” genotypes, belonged to Rahnella spp., Peribacillus spp. and Paenarthrobacter spp. genera, respectively. However, some research has demonstrated that these three genera members could be beneficial for legume plants on the different plant growth-stimulating mechanisms [ 27 , 28 , 29 ]. For example, Rahnella sp. BIHB 783 isolate is described as a cold-adapted psychrotroph which produces indole-3-acetic acid, indole-3-acetaldehyde, indole-3-acetamide, indole-3-acetonitrile, indole-3-lactic acid and other plant beneficent compounds. This strain also has the ability to solubilize organic and inorganic phosphorus forms. In the same study, Rahnella sp. BIHB 783 isolate inoculation significantly increased pea shoot length, dry weight and yield by 25%, 32% and 37%, respectively [ 27 ]. We found a few studies where scientists demonstrated that pea lateral root number increased significantly after Peribacillus simplex 30N-5 inoculation [ 28 ]. Also, the fresh weight of soybean root was significantly improved and was two times higher after P. simplex R180 inoculation [ 29 ]. Paenarthrobacter nitroguajacolicus UP1 isolated from pea rhizosphere showed phosphorus and potassium solubilization activity and siderophore production in in vitro experiments [ 29 ]. Our results showed that the tree isolates ASTR04, RSP03 and RSP06 increased dry plant biomass significantly compared to the control. The average dry plant biomass increased 1.7-fold where isolates ASTR04 and RSP06 were inoculated, and 1.9-fold where isolate RSP03 was inoculated. Plants inoculated with isolate RSP03 accumulated significantly higher biomass compared to ASTR04- and RSP06-inoculated plants. It is possible that these isolates’ consortium with rhizobia could improve legume plant growth and development, however, additional plant analysis is necessary to confirm that presumption. Phylogenetic 16S rRNA gene sequence analysis showed that 47 isolates were assigned to Rhizobium spp., among them RSP09, EGLE01 and BAGOO04 isolates that did not form nodules. RSP09 isolate clustered with R. nepotum 39/7 type strain. R. nepotum belongs to a phylogenetic clade formerly known as agrobacteria and is considered as non-symbiotic and nonnodulating bacteria [ 30 ]. Few studies reported that R. nepotum can fix nitrogen, and produce auxins and siderophore [ 31 , 32 ]. Another work observed that R. nepotum inoculation had a positive effect on soybean leaf area, pod weight and number of pods [ 33 ], however, this bacterial species is less studied with plants. In our research, no significant differences in average dry plant biomass formation were observed between RSP09-inoculated plants and the control variant. Phylogenetic analysis of 16S rRNA gene sequences revealed that 46 isolates, including nodules not formed from isolates EGLE01 and BAGOO04, were attributed to. the R. leguminosarum group, which included R. leguminosarum bv. viciae USDA 2370, R. laguerreae FB206, R. sophorae CCBAU 03386, R. anhuiense CCBAU 23252, R. acidisoli FH13, R. ruizarguesonis UMP1133 and R. hidalgonense FH14 type strains. According to the scientific literature, 16S rRNA gene sequences of these type strains in the R. leguminosarum group are similar in 99.9% of cases [ 34 , 35 ], therefore, it is problematic to determine isolates’ phylogeny with a specific species in this cluster. Phylogenetic analysis of Rhizobium spp. 16S rRNA sequences is not indicative of genospecies [ 35 ]. According to scientific research, R. leguminosarum is too diverse to be considered a single species, and it can be described as a species complex [ 36 ]. Kumar et al. were the first to divide the R. leguminosarum isolates into five genospecies A–E [ 36 ]. The R. leguminosarum isolates were separated by average nucleotide identity (ANI), and values below 95%, represent separate species [ 37 ]. Recently extensive genomic analysis has subdivided R. leguminosarum into 18 different genospecies, 5 of which include the type strains of named species: R. laguerreae , R. sophorae , R. ruizarguesonis , R. indicum and R. leguminosarum itself [ 35 ]. The results of our performed phylogenetic atpD and recA gene analysis showed that 46 Rhizobium spp. isolates divided into three genospecies: 30 isolates clustered with R. leguminosarum bv. viciae 3841 (genospecies B), 15 isolates formed close clustering with R. leguminosarum USDA 2370T (genospecies E) and only one ASTR05 isolate was clustered to R. leguminosarum sv. phaseoli FA23 (genospecies K). All 46 isolates were identified as R. leguminosrum . Most of Rhizobium spp. isolates isolated from “Egle DS” and “Lina DS” genotypes clustered within the same genospieces (B). According to the Lithuanian Research Centre for Agriculture and Forestry (LAMMC), breeders “Egle DS” and “Lina DS” are related genotypes. “Egle DS” was developed by crossbreed “Respect” and “Audit” genotypes, and “Lina DS” was developed by crossbreed “Grafila” and “Respect”. However, R. leguminosarum isolates isolated from “Respect”, “Bagoo” and “Astronaute” genotypes dispersed into (B) and (E) clusters evenly. Several studies indicated that nodulation and BNF are influenced not only by the population of rhizobia in the soil but also by the genotype of the legume plant. Different legume genotypes can attract different rhizobia strains, while the same rhizobia strain can fix different amounts of nitrogen in different plant genotypes [ 38 , 39 , 40 ]. Those differences are associated with specific flavonoids produced by different legume genotypes responsible for rhizobia attraction to the root hairs. Rhizobia, in turn, produces the nod factors that induce morphological changes in their host plant root system [ 41 ]. This means that symbiosis formation is controlled by multiple genetic factors from pea and rhizobia [ 42 ]. It is possible that “Egle DS” and “Lina DS” genotypes produce similar flavonoids and attract genetically similar rhizobia strains. Another option is that genetically related rhizobia produce similar nod factors which are easily recognized by related genotypes. Currently, the nodC gene is commonly used as a phylogenetic marker of Rhizobium spp. symbiovars [ 43 ]. The reason is that nod genes encode species-specific modifications to the nod factor structure and, related to this, specific nod genes have been shown to be major determinants of legume host specificity [ 44 , 45 , 46 ]. NodC gene analysis showed that 43 R. leguminosarum isolates spread into 10 different groups and were assigned to R. leguminosarum sv. viciae symbiovar. This also revealed that some isolates from different genospecies have the same nodC genes and are clustered in the same groups. Some studies suggest that nod and nif genes are symbiotic, adaptive genes and, in many cases, have an evolutionary history independent of the rest of the genome. Nod loci could be transferred by lateral gene transfer, even across divergent chromosomal lineages [ 47 ]. Horizontal transfer of nodulation genes can potentially help rhizobia adapt to a new host plant or enable genetically different bacteria with similar nod genes to form symbiotic interactions with the same legume plants [ 47 , 48 ]. Phylogenetic analysis results confirmed nodulation results, i.e., that these isolates form a symbiosis with pea plants. The same 43 isolates formed nodules on Lithuanian pea genotype “Egle DS” roots. However, nodC gene amplification of EGLE01, BAGOO04 and ASTRO01 isolates was not successful. Amplification of these isolates was repeated five times, different PCA conditions were applied, but the nodC gene was not amplified. AtpD and recA gene analysis of EGLE01, BAGOO04 and ASTRO01 revealed that these isolates can be identified as R. leguminosarum . Nodulation test results indicated that the ASTRO01 isolate can form nodules on pea roots, while EGLE01, and BAGOO04 do not form nodules and symbiosis with pea plants. Whereas nodC gene amplification with these isolates failed, we can only presume that ASTRO01 could have been assigned to R. leguminosarum sv. viciae symbiovar, and that the remaining two EGLE01 and BAGOO04 isolates may potentially be categorized into other symbiovars, which are incapable of forming a symbiosis with pea plants. However, nodC gene phylogenetic analysis results of 43 R. leguminosarum isolates confirmed that this gene could be used for symbiotic relationship determination between R. leguminosarum isolates and different legume plants [ 49 , 50 ]. Nitrogen-fixing rhizobia strains increase biomass accumulation in legumes, and consequently dry plant biomass serves as a pertinent indicator for the potential evaluation of rhizobia nitrogen fixation [ 51 , 52 ]. Pea biomass analysis with the “Egle DS” genotype revealed that 17 out of 18 rhizobia isolates, which were the most effective on dry plant biomass formation, belonged to the same genospieces (B), and only the isolate EGLE10 grouped in the cluster (E). Most of the dry plant biomass-increasing isolates were isolated from Lithuanian pea genotypes, and seven rhizobia isolates were obtained from “Egle DS”, while six strains were obtained from “Lina DS”. This revealed that the highest plant biomass was achieved when isolates of the same genotype or related “Lina DS” genotype were used for plant mono-inoculation. These results could be important for highly effective commercial Rhizobium inoculate formulation and adaptation for specific pea genotypes. The nodulation results showed the most nodulating rhizobia isolates dispersed similarly into both (B) and (E) clusters. Few of the most nodulated RSP08, RSP05, LIN03, LIN10, ASTR03, EGLE04, LIN04, LIN07 and EGLE05 isolates also exhibited high efficiency on plant biomass accumulation. However, some of rhizobia isolates formed a high nodule number, but had less or no effect on plant biomass formation. Correlation analysis revealed that the relationship between nodulation and dry plant biomass accumulation was significantly positive, but weak (r = 0.2293). The weak correlation suggests that, while dry plant biomass tends to increase with the number of nodules, the nodulation is not the sole determinant of plant biomass accumulation. The growth of a plant and its biomass accumulation are complex processes that depend on many external and internal factors. These findings confirmed the results of the other authors, indicating that a high nodule number is an important parameter demonstrating rhizobia nodulation ability, however, this does not guarantee biomass accumulation or efficient nitrogen fixation for the plant. Plant biomass formation and nitrogen accumulation depend on plant genetic characteristics, soil composition, climate and rhizobia–strain–nitrogen–fixation efficiency [ 53 , 54 ]. This has scientifically proven that legumes’ symbiosis with partially efficient or inefficient rhizobia strains can interfere with root nodulation through efficient rhizobia strains, consequently reducing biological nitrogen fixation for the plant [ 40 , 55 , 56 ]. In summary, phylogenetic analysis of recA and atpD genes showed that rhizobia diversity in the field where ecological farming was applied and no commercial rhizobia inoculants were used, was not high, and most of the isolates were genetically similar and clustered in two main clusters (B) and (E), and only one isolate was attributed to (K) cluster. Nevertheless, most of the isolates significantly increased pea biomass accumulation and intensively formed nodules. Many studies have shown that rhizobia inoculants can improve nodulation, nitrogen fixation and grain yield [ 57 , 58 , 59 , 60 ], however, highly effective and competitive rhizobia strains need to be selected to ensure efficient BNF [ 61 , 62 ]. Our isolated and analysed LIN03, EGLE05 and ASTR03 rhizobia strains have high potential as biofertilizers, and these isolates formed a high number of nodules and increased pea biomass more than 2.5-fold compared to the control. However, nitrogen fixation efficiency, competitiveness and plant reinoculation with the same and different genotype strains should additionally be analysed in further investigations to create highly effective Rhizobium inoculates for specific pea genotypes or wide-range applications."
} | 4,683 |
37938734 | PMC9723589 | pmc | 6,913 | {
"abstract": "Most marine sponge species harbour distinct communities of microorganisms which contribute to various aspects of their host’s health and physiology. In addition to their key roles in nutrient transformations and chemical defence, these symbiotic microbes can shape sponge phenotype by mediating important developmental stages and influencing the environmental tolerance of the host. However, the characterisation of each microbial taxon throughout a sponge’s life cycle remains challenging, with several sponge species hosting up to 3000 distinct microbial species. Ianthella basta , an abundant broadcast spawning species in the Indo-Pacific, is an emerging model for sponge symbiosis research as it harbours only three dominant symbionts: a Thaumarchaeotum, a Gammaproteobacterium, and an Alphaproteobacterium. Here, we successfully spawned Ianthella basta , characterised its mode of reproduction, and used 16S rRNA gene amplicon sequencing, fluorescence in situ hybridisation, and transmission electron microscopy to characterise the microbial community throughout its life cycle. We confirmed I. basta as being gonochoric and showed that the three dominant symbionts, which together make up >90% of the microbiome according to 16S rRNA gene abundance, are vertically transmitted from mother to offspring by a unique method involving encapsulation in the peri-oocytic space, suggesting an obligate relationship between these microbes and their host.",
"introduction": "Introduction Sponges are found in marine and freshwater environments from the tropics to the poles, playing a particularly important role in the survival and productivity of coral reefs where they recycle and retain (in)organic nutrients [ 1 , 2 ]. Their evolutionary success is, in part, derived from the metabolic integration between the sponge host and its diverse community of symbiotic microbes [ 3 , 4 ]. For example, sponge-associated microbes can remove waste products, provide the host with nutrients, and produce secondary metabolites that aid in host defence [ 5 – 9 ]. The host must ensure high fidelity of transmission of important microorganisms [ 10 ], therefore these sponge symbionts are expected to be heritable, either through vertical transmission (from the parent), horizontal transmission (from the environment), or a combination of both (i.e. ‘mixed mode’; [ 11 – 16 ]). Symbiont acquisition mode can shape sponge phenotype, particularly when microbes mediate important developmental stages. For example, vertically inherited symbionts of the sponge Amphimedon queenslandica supply their host with the amino acid arginine, something the host cannot produce itself [ 17 – 19 ], and arginine is also used to produce nitric oxide, a signalling compound that triggers settlement and metamorphosis of A. queenslandica larvae [ 20 ]. Vertical transmission ensures the faithful transfer of microbes critical to host survival. However, sponges that rely solely on vertical transmission can lose the ability to adapt to changing environments [ 21 , 22 ]. Horizontal transmission, where microbes are taken up from the surrounding seawater, allows for the acquisition of genetically diverse microbial strains that may enable the host to better acclimatise to changing environmental conditions [ 12 , 23 , 24 ]. However, this mode of transmission can lead to the loss of beneficial symbionts as well as to the acquisition of pathogens [ 11 , 21 , 25 ]. The mode and mechanism of transmission can thus provide insight into the nature of the host-symbiont relationship, yet for most sponge species the mechanisms of symbiont transmission remain unknown. The development of model systems where the microbiome is fully characterised throughout the host life cycle would greatly facilitate studies on how individual symbionts mediate host development and adaptation [ 17 ]. Ianthella basta is an oviparous, verongiid demosponge, a group for which little is known about their sexual reproduction [ 26 , 27 ]. This sponge harbours only three dominant symbionts that make up >90% of the microbial community: a Thaumarchaeotum, a Gammaproteobacterium and an Alphaproteobacterium [ 7 , 28 ]. The remaining <10% of the microbial community is composed of low abundant or transitory taxa. This reduced microbial complexity enables the characterisation of the mode of transmission for each symbiont throughout the sponge’s life cycle, a task that is intractable in sponges with more complex microbiomes. Here, we successfully spawned I. basta in indoor aquaria, confirmed its mode of reproduction, and characterised its late-stage gametes. Using a combination of 16S rRNA gene amplicon sequencing, fluorescence in situ hybridisation (FISH) and transmission electron microscopy (TEM) we characterised the microbial community of I. basta adults and offspring. We reveal that all dominant symbionts are vertically transmitted through a unique mechanism, where microbes are packaged into the peri-oocytic space before the release of the oocyte into the seawater, ensuring fidelity of transmission, and incorporation into the oocyte after its release.",
"discussion": "Discussion Marine sponges harbour distinct communities of microbes, that can shape sponge phenotype by mediating important developmental stages and influencing the environmental tolerance of the host. However, the characterisation of each microbial taxa throughout a sponge’s life cycle remains challenging. Here, we characterised the microbiome of the model sponge Ianthella basta throughout its life cycle to facilitate future studies on how individual symbionts mediate host development and adaptation. To this end, we successfully spawned the oviparous sponge I. basta , confirming it is a gonochoric species (separate male/female sexes) that releases gametes after the new moon in August on the Great Barrier Reef. We subsequently characterised the microbiome throughout all life stages of the sponge and revealed that the three dominant sponge symbionts are vertically transmitted from female adults to oocytes through a unique mechanism. In this mechanism, microbes are packaged into the space between spherulous cells and the oocyte (termed peri-oocytic space) before the release of the oocyte into the seawater and are incorporated into the oocyte after its release, suggesting an obligate relationship between these symbionts and their host. To characterise I. basta ’s life cycle at the start of development, we first investigated gamete production to determine the mode of gametogenesis. Reproductively mature I. basta individuals simultaneously released gametes in aquaria 1-day after the new moon in August 2018, consistent with previous field observations in which I. basta contained gametes in August 2010, but not in September 2010 [ 27 ]. Spermatogenesis in I. basta occurred within 5-days of broadcast spawning and male and female gametes did not co-occur in individuals, confirming I. basta is gonochoric. This finding follows the general pattern of gonochorism being the dominant sexual phenotype for oviparous sponges, with 13 out of the 22 orders belonging to the Demospongia showing gonochorism [ 45 ]. The microbial community of I. basta was subsequently characterised throughout the sponge’s life cycle using 16S rRNA gene amplicon sequencing. The microbiome of both adults and offspring was dominated by three symbionts (i.e. a Thaumarchaeotum, Gammaproteobacterium, and Alphaproteobacterium) and was significantly different from the surrounding seawater. The three dominant sponge symbionts were present in low abundance in all tank water (<0.1% abundance) and absent from the inflowing seawater, with the exception of an extremely low abundance of the Thaumarchaeotum in one of the four seawater replicates. As these Thaumarchaeotum reads likely result from a contamination during sequencing, we suggest that the observed signals of all three dominant symbionts in the tank water originated from the sponge host and conclude that the symbionts are vertically transferred. Vertical transmission of the three dominant symbionts would also be expected as the larvae and embryos are lecithotrophic (reliant on internal yolk reserves for nourishment). Horizontal acquisition of microbes generally occurs post-larval settlement and metamorphosis when the aquiferous system develops [ 46 – 49 ], and recruits are actively filter feeding [ 23 , 50 ], which occurred in I. basta juveniles between 5 and 12 days post-settlement. To determine the mechanism of vertical transmission of the three dominant symbionts, we further processed oocytes in the mother adult tissue as well as released oocytes and embryos for FISH and TEM. In other commonly studied reef species, like broadcast spawning corals or the sea anemone Nematostella vectensis , microbes are thought to be transmitted through the mucus surrounding the gametes, while brooding corals could seed bacteria into developing larvae [ 51 – 54 ]. In sponges, symbionts are most commonly transferred prior to release of the eggs or larvae via both phagocytosis and other mechanisms. For example, in the viviparous homosclerophorid Corticium candelabrum , symbionts aggregate around late-stage oocytes, which are initially engulfed and digested as food [ 55 ] and then migrate into the intercellular spaces of the embryo during cleavage after the egg is internally fertilised [ 12 ]. In other species, symbionts are initially phagocytosed from the mesohyl by nurse cells which subsequently transfer the symbionts to the oocytes through cytoplasmic bridges e.g. [ 50 ]. In contrast, in I. basta , FISH indicated that the three dominant symbionts aggregate directly around the oocytes in the adult tissue (Fig. 5 ), without being incorporated, which was supported by TEM (Fig. 2 ). Although FISH images showed that the Thaumarchaeotum occasionally occurred within the oocytes, this was not supported by TEM observations despite quantitative FISH analysis indicating that archaea should be present at sufficient numbers to be detectable by TEM (see Supplementary note 3 and Table S5 ). The same discrepancy between FISH and TEM images (i.e. a signal versus no detection) has previously been reported in oocytes from the oviparous sponge Ectyoplasia ferox [ 50 ] and may be an artefact of sample processing for FISH (i.e. fixation, embedding, and sectioning for FISH might transfer some cells). Taken together, the results suggest that the three dominant symbionts surround oocytes, but do not directly inhabit the oocyte while inside the mother tissue. However, upon release into the seawater, both FISH and TEM showed the incorporation of microbes, which included the three dominant symbionts. Interestingly, TEM showed that symbionts were expelled along with the egg, hosted in the peri-oocytic space between the oocyte and follicle membrane of maternal cells, and incorporated into the egg through phagocytosis before the disintegration of the envelope of maternal cells (Fig. 3 ). Thus, we show that symbiont incorporation takes place only after the eggs are released into the water column. This timing differs from the method of symbiont incorporation of the only two other species in the order Verongiida studied to date, Aplysina cavernicola and Aplysina aerophoba , which phagocytose microbial symbionts from the mesohyl into late-stage oocytes prior to rather than post release [ 26 , 56 ]. The novel mechanism of vertical transmission identified in I. basta ensures fidelity of transmission, maintains a tight association between I. basta and its microbial community, and suggests an obligate relationship between the dominant symbionts and their host. Once transmitted to new offspring, the dominant Thaumarchaeotum occurred at a significantly higher relative abundance (20 ± 9% in adults, 46 ± 3% in oocytes and 48 ± 11% in embryos), potentially indicating its importance in I. basta ’s development. One way through which Candidatus Nitrosospongia ianthellae could mediate developmental stages is by producing nitric oxide through nitrite reduction catalysed by its highly expressed nitrite reductase NirK [ 7 ]. Nitric oxide is a signalling compound that induces metamorphosis in the sponge A. queenslandica [ 20 , 57 ]. It also regulates metamorphosis in various other marine invertebrates [ 58 – 61 ], where it either induces or inhibits settlement. Whether the Thaumarchaeotum and the compound nitric oxide are involved in the settlement and metamorphosis of I. basta remains to be confirmed, thus further experiments to clarify the role of the Thaumarchaeotum and nitric oxide in sponge larval settlement and metamorphosis are required. While all three dominant symbionts are vertically transferred, our data also suggest that several other microbial taxa may be acquired post-settlement, likely when the sponge begins pumping. Although present in low abundance, these microbes can be repositories of key functions [ 62 ]. Hence, we also characterised their potential mode of transmission. ASV_4786, belonging to the order Chlamydiales, of which the members are obligate intracellular symbionts, was found in female adults and oocytes and is likely acquired vertically by oocytes. In contrast, ASV_4387 (class Gammaproteobaceria, genus Endozoicomonas) is putatively horizontally transmitted (i.e. taken up from the seawater when the sponge starts pumping) as it was only present in adults and seawater. Here we identified the mode of transmission of the major microbiome members in the tropical reef sponge I. basta . We show that the three dominant symbionts, comprising >90% of the microbial community according to 16S rRNA gene abundance, are vertically transmitted through a unique mechanism. This mechanism involves the encapsulation of microbes in the peri-oocytic space and the incorporation of the microbes into the oocytes after their release into the surrounding seawater, suggesting an obligate relationship between these microbes and their host. Consequently, this study provides a valuable framework for future manipulative experiments to study the role of the three dominant symbionts in the development of I. basta . Furthermore, it established a fully characterised model organism to further study the molecular mechanisms for symbiont acquisition in demosponges."
} | 3,582 |
39255046 | PMC11789975 | pmc | 6,914 | {
"abstract": "Abstract As a globally abundant source of biomass, lignocellulosic biomass has been the centre of attention as a potential resource for green energy generation and value‐added chemical production. A key component of lignocellulosic biomass, lignin, which is comprised of aromatic monomers, is a potential feedstock for value added chemical production. The cleavage processes of the linkages between monomers to obtain high value products, however, requires significant investigation as it is a complex, non‐facile process. This study focuses on the photocatalytic valorization of a β‐5 lignin model compound, a key linkage in the lignin structure. It was found that greater yields of aromatic products were obtained from the photocatalytic conversion of β‐5 lignin model compound using carbon nitride (CN) when compared to Evonik P25 titanium dioxide (TiO 2 ). Products of the β‐5 model compound photocatalytic conversion were determined and C−C bond cleavage was observed. It was also determined that the solvent participated in the reactions with the introduction of a cyano group to one of the products. Radical quenching experiments revealed that superoxide radicals participated in the CN photocatalytic conversion. These results reveal for the first time the products and possible mechanism of the photocatalytic transformation of β‐5 model compounds using CN photocatalysis.",
"conclusion": "Conclusions In the present study, building on the previous research on the photocatalytic degradation \n [36] \n of β‐5 lignin model compounds by TiO 2 , a typical non‐metallic semiconductor CN was successfully synthesised and used as a photocatalyst for the degradation of this material. The results of this work demonstrated that CN effectively degraded the β‐5 model compounds under both UV and visible light irradiation, while generating different decomposition by‐products compared to those produced when a P25 TiO 2 material was used as the photocatalyst. The reaction solution was analyzed by HPLC, and the products identified by GC‐MS with the results showing that the possible initial fracture site was between the α‐C−C bond between the benzofuran and the aromatic ring. Furthermore, the ROS involved in the photocatalytic reaction were also investigated. Using the CN photocatalyst the degradation of β‐5 was believed to result from reaction with ⋅ O 2 \n − as indicated from radical quenching studies. The role and the mechanism of how acetonitrile participated in the reaction, however, requires further investigation. This reaction demonstrates that CN photocatalysts may be highly selective materials for the conversion of β‐5 compounds into monomeric high‐value compounds.",
"introduction": "Introduction With increasing CO 2 emissions each year, it is imperative that there is a transition from fossil fuels to meet Net Zero targets, \n [1] \n which have been set to ensure the rise in global average temperature does not exceed 2 °C above pre‐industrial levels as agreed in the 2015 Paris agreement. \n [2] \n For example, in 2019 the UK was one of the first countries to enshrine into law the objective of ensuring greenhouse gas emissions would be net zero, compared to 1990 levels, by 2050. \n [3] \n \n With abundant renewable biomass sources, bioenergy could provide sustainable and low‐carbon energy which supports the Net Zero energy transition. \n [4] \n Bioenergy is already an important part of the energy economy. In 2024 biomass supplied (on average) 6.3 % of the UKs electricity generation, which equates to approx. 1.9 GW. Overall, in 2024, renewables (including biomass) provided 44.4 % of electricity to the grid, which is just over 13 GW of production.[ \n 5 \n , \n 6 \n ] To meet the carbon emission reduction targets, the proportion of modern bioenergy which is considered as “biomass used alongside modern heating technologies, power generation and transport fuels as opposed to traditional wood‐burning methods commonly used for heating and cooking in developing countries ” \n [7] \n needs to increase. Biomass conversion technology is also one of the few systems that could offer ‘negative emissions’ if it is coupled with Carbon Capture Utilization and Storage (CCUS) processes. Based on hydrogen production, wood gasification with CCS could generate ~‐150 g CO 2 e/MJ H 2 (LHV). By comparison, fossil fuel SMR produces ~90 g CO 2 e/MJ H 2 (LHV). \n [8] \n Consequently, research into converting biomass into fuels and high‐value chemicals has received a significant amount of attention in recent years.[ \n 9 \n , \n 10 \n , \n 11 \n , \n 12 \n , \n 13 \n , \n 14 \n , \n 15 \n , \n 16 \n ] Over the past decade there has been a substantial level of research interest in the application of photocatalytic reforming of lignocellulosic biomass.[ \n 17 \n , \n 18 \n , \n 19 \n , \n 20 \n , \n 21 \n ] This abundant sustainable feedstock is a promising substrate for clean energy production with cellulose and hemicellulose having been widely investigated and demonstrated to be potential substrates for the generation of hydrogen and value‐added chemicals.[ \n 22 \n , \n 23 \n ] Using sunlight as the energy source under ambient temperature and pressure, photocatalytic reforming of biomass to produce hydrogen or high‐value chemicals is recognized as a potential low‐carbon and more sustainable technology \n [24] \n compared to conventional thermochemical methods. \n [25] \n Lignin, a complex and abundant biopolymer found in plant cell walls, is one of the most promising renewable resources as the largest natural large scale source of aromatics. \n [12] \n Historically considered a by‐product of the pulp and paper industry, lignin has gained increasing attention in recent years due to its potential as a valuable feedstock for the production of high‐value chemicals and materials. \n [26] \n Lignin can be depolymerized to produce various phenolic compounds, such as vanillin, syringaldehyde, guaiacol, and catechol. \n [27] \n These compounds have applications in the manufacturing of food flavorings, pharmaceuticals and fragrances, or as chemical intermediates for the synthesis of polymers and resins. \n [28] \n \n The reforming of lignin, however, remains challenging due to its complicated three‐dimensional amorphous polymer structure which consists of three aromatic main units: syringyl, guaiacyl, and p‐hydroxyphenyl units.[ \n 29 \n , \n 30 \n ] The monomers of lignin are linked mainly by ether or carbon‐carbon bonds, with more than two thirds ether bonds and the rest carbon‐carbon bonds. The linkages of the lignin monomers are mainly β‐O‐4 (40–60 %), β‐5 (4–12 %), and 5–5 (4–25 %). There are also small amounts of α‐O‐4 (4–8 %), β‐β (2–7 %), 4‐O‐5 (4–7 %), β‐1 (3–7 %) and α‐O‐γ, etc.[ \n 31 \n , \n 32 \n ] (Figure 1 ). β‐O‐4 linkages being the most abundant linkage found in native lignin has been the first and most extensively studied[ \n 33 \n , \n 34 \n , \n 35 \n , \n 36 \n , \n 37 \n , \n 38 \n ] as model compounds, while the other linkages have not yet been investigated in as much detail.[ \n 39 \n , \n 40 \n , \n 41 \n ]\n Figure 1 \n S chematic showing linkages present between lignin monomers. Our group has previously reported the photocatalytic degradation of β‐5 linkages using a TiO 2 photocatalyst activated with UV‐light. \n [42] \n Under low power UV‐light emitting diode (LED) irradiation at 370 nm, complete conversion (to the limit of detection) of the β‐5 compounds (6.3×10 −3 mg ml −1 min −1 ) was achieved together with the formation of a number of reaction intermediates. While a number of intermediates in the TiO 2 ‐catalysed degradation of β‐5‐linked model compounds were identified, guaiacol or guaiacol‐based products were not detected at the end of the reaction period. \n [42] \n This may have been due to non‐selective oxidation as a result of the strong oxidising capacity of the OH radical (2.8 eV vs NHE) generated on the photocatalyst surface. Therefore, there is a need for a photocatalyst that can selectively convert β‐5 model compounds under visible light to obtain the targeted high‐value monomers. In other studies, the selective conversion of lignin model compounds on graphite like carbon nitride (CN) and derived materials has been demonstrated. In previous studies it was confirmed that modified CN can selectively convert β‐O‐4 and β‐1 lignin model compounds.[ \n 37 \n , \n 43 \n ] Consequently, in this work CN,[ \n 44 \n , \n 45 \n , \n 46 \n , \n 47 \n ] a common non‐metallic semiconductor photocatalytic material, has been used for the photocatalytic conversion of the β‐5 model compound. CN, which consists of carbon and nitrogen elements in a layered graphite‐like structure, is stable, and the threat of secondary pollution to the environment is lower than that with nano‐TiO 2 materials.[ \n 48 \n , \n 49 \n ] CN has a lower bandgap (Eg ≈2.8 eV) and conduction band position than TiO 2 (Eg ≈3.2 eV), and as a result is not believed to generate hydroxyl radicals directly from water. Consequently, this should inhibit degradation of monomeric products. An additional key advantage of CN photocatalysts is their ability to absorb visible light, which constitutes a significant portion of the solar spectrum. Moreover, CN has been used by other researchers to verify the feasibility of photocatalytic cleavage of β‐O‐4 model compounds and to demonstrate that CN can selectively break these ether linkages.[ \n 37 \n , \n 50 \n ] In light of these results with the β‐O‐4 model compounds, CN was synthesized and investigated for the cleavage of the β‐5 linkage and compared to TiO 2 with identification of monomeric products. This paper details an investigation of the performance, potential monomer products and cleavage mechanism for degradation of β‐5 lignin model compound using a CN photocatalyst under visible light irradiation.",
"discussion": "Results and Discussion Characterisation of the CN Photocatalyst The CN photocatalyst was initially characterized using X‐ray diffraction (XRD), Fourier transform infra‐red spectroscopy (FTIR), Brunauer‐Emmett‐Telle surface area (BET) analysis, transmission electron microscopy (TEM) and scanning electron microscopy (SEM). As shown in Figure 2 , a series of characteristic peaks for CN were observed in the FT‐IR spectrum. \n [51] \n A series of absorption bands from 1200–1600 cm −1 which correspond to the stretching vibration modes of C−N(−C) −C or bridging C−NH−C units were observed. The wide band between 3000–3500 cm −1 is attributed to the N−H vibration. There was a sharp peak located at 808 cm −1 in the fingerprint region that has been attributed to the breathing mode of tri‐s‐triazine units \n [52] \n . The XRD spectrum and TEM images for the CN photocatalyst are displayed in Figure 3 . In the XRD diffractogram the two peaks located at 13° and 27° represent the interlayer long range order and stacking of carbon and nitrogen under van der Waals forces. \n [53] \n TEM analysis confirmed that the CN material consists of multiple layers stacked on top of one another. Based on the above results, it is clear that the synthesized material has a layered graphite‐like structure. BET analysis (Table 1 ) revealed a surface area of 48.92 m 2 g −1 which was comparable to the surface area (48.2 m 2 g −1 ) reported previously for this material. \n [37] \n \n Figure 2 FT‐IR spectrum of CN. Figure 3 \n A) XRD pattern and B) TEM images of prepared CN. Table 1 BET surface area and pore volume of CN. \n Sample \n \n Surface area \n (m 2 /g) \n \n Pore volume \n (cm 3 /g) \n \n Pore size \n (Å) \n \n CN \n \n 48.92 \n \n 0.13 \n \n 82.50 \n \n C3N4‐U \n [33] \n \n \n \n 48.20 \n \n 0.15 \n \n 29.00 \n Wiley‐VCH GmbH The CN bandgap width was determined using UV‐visible diffuse reflectance spectroscopy (UV‐vis DRS). In Figure 4 the UV‐vis DRS spectra of the CN and TiO 2 P25 photocatalyst materials shows the adsorption wavelength edges located at 451 nm and 391 nm respectively. This spectrum implies that the prepared CN is photo‐responsive in the visible light region of 400–450 nm whereas P25 TiO 2 is not. Estimated from the tangent intercept of the (αhν)1/2 vs photon energy curve, the band gap energy (Eg) of CN is 2.88 eV which is consistent with a previous report. \n [37] \n \n Figure 4 \n A) UV‐vis DRS spectra and B) plots of (αhυ)1/2 versus photon energy of CN and P25. Comparison of Photocatalytic Performance of CN and P25 TiO 2 Photocatalysts To investigate the performance of the prepared CN photocatalyst for the degradation of the β‐5 lignin model, photocatalytic degradation experiments were carried out with both this material and an Evonik P25 TiO 2 photocatalyst. For comparison with previous work, and to investigate the performance under visible light in this experiment, 370 nm, 440 nm and 470 nm LEDs were used as irradiation sources. As shown in Figure 5 , in an acetonitrile solution, 96.4 % of β‐5 was degraded by the CN material within 90 min under 370 nm irradiation. In comparison while using a P25 TiO 2 photocatalyst, only 37.2 % degradation was achieved under the same reaction conditions. When the irradiation source was changed to visible light (440 nm), CN still demonstrated a good level of degradation, with 66.6 % of β‐5 degraded over the same reaction time period. When a 470 nm LED was utilized as the illumination source, the degradation rate decreased rapidly, with only 17.5 % degradation following 90 min of irradiation. Interestingly, however, in an acetonitrile/water solution as used in our previous work, CN′s degradation capacity was greatly inhibited. Only 24.8 % degradation was obtained over 90 min under irradiation with the 440 nm illumination. The P25 material was able to achieve a 42.0 % degradation using the mixed acetonitrile/water solvent under identical experimental conditions. This indicates that CN has demonstrated photocatalytic activity for β‐5 degradation in the UV−A and near UV‐visible regions. The bandgap of CN is narrower than that of P25, allowing it to obtain a wider light absorption region than that of P25, but at the same time reducing the redox capacity of CN. Therefore, this may lead to different reaction mechanisms for CN and P25. In order to obtain a reasonable level of degradation, it was necessary to use only acetonitrile as a solvent.\n Figure 5 β‐5 photocatalytic degradation with CN and P25 TiO 2 under different light wavelengths and solvents(a) and plot ln(C/C0) verse time(b). (AN: pure acetonitrile, AW: acetonitrile/water mixture (50/50 v/v)). In addition to investigating the performance of CN for the degradation of β‐5, it was important to examine the by‐products generated as part of this degradation process. The HPLC chromatograms shown in Figure 6 are taken from substrate analysis at different time periods under CN and P25 photocatalytic degradation reactions. The peak located at around 9.8 min (R T 9.8) is characteristic of the β‐5 substrate. During 0–30 min, P25 TiO 2 and CN simultaneously produced products that peaked at 13 min (R T 13). After 30 min as the intensity of the R T 13 peak decreased, the product peaks R T 5.1 and R T 7.3 were produced under P25 TiO 2 photocatalysis, whereas the product peaks R T 8.1, 9.1 and 9.3 were produced under photocatalysis using the CN material. This implies that P25 TiO 2 and CN appeared to follow the same reaction pathway in the first 30 min of photocatalysis. After 30 min, however, the two photocatalysts followed different pathways for the β‐5 substrate resulting in different degradation products.\n Figure 6 Stack HPLC chromatograms of extracted substrate across reaction time periods with P25 (a) and CN (b). Investigation of Photocatalytic Conversion Mechanism of β‐5 Model Compound on CN For a better understanding of the products generated and the mechanism of CN photocatalytic conversion of the β‐5 model compound, GC‐MS was utilized for the analysis of reaction solutions from different reaction times and irradiations. From the results shown in Figure S1–S5 and, Table S1 and S2 in the supplementary information file, it can be noted that the products obtained from the photocatalytic conversion of β‐5 by CN under different peak wavelengths (e. g.) are consistent. In order to compare with P25 TiO 2 under the same conditions, the reaction solution after 1 and 4 under 370 nm irradiation were used for analysis. Based on previous studies, the identified products are shown in Figure 7 . Notably, a guaiacol‐based monomeric compound which was not observed in previous work was detected as a product of the CN photocatalytic process, labelled as products P4 and P5 (Figure 7 ). Products P1 , P2 and P3 are proposed to have been produced by cleavage of the α−C−C bond (labelled on β‐5 dimer in Figure 7 ) between the benzofuran and the aromatic ring, with such an elimination being proposed to result in the formation of an alkene in the furan ring. The above postulated products are consistent with the previous work, where the α−C−C bond of the model compound was cleaved. Furthermore, product P2 is proposed to undergo oxidation of the propenyl sidechain to furnish an aldehyde, while P3 is further oxidised to a carboxylic acid. Unusually, however, product P2 was not further degraded but remained stable after cleavage under CN photocatalysis, as it was detected in both the 4 h reaction samples, with the signal intensity remaining stable with only about 6.5 % loss. These findings suggest that P2 has not undergone oxidation to the carboxylic acid within the reaction solution and that P3 was formed by a potentially different mechanism. Vanillin ( P5 ) was identified by HPLC and confirmed by GC‐MS (Figure S3 and S4). P5 is thought to be produced via cleavage of side‐chain olefin and subsequent bond cleavages within the benzofuran ring.\n Figure 7 Potential products from β‐5 conversion on CN analyzed by GC‐MS and detailed structure of β‐5 substrate. In order to gain a deeper understanding of the mechanism of photocatalytic conversion of β‐5 model compounds by the CN photocatalyst, free radical experiments were carried out to elucidate which Reactive Oxygen Species (ROS) were generated as part of the photocatalytic process. Isopropanol (IPA), p‐benzoquinone (BQ) and ammonium oxalate (AO)/EDTA Na were utilized as hydroxyl radical ( ⋅ OH), superoxide radical ( ⋅ O 2 \n − ) and photogenerated holes (h + ) quenchers, respectively, and the results are shown in Figure 8 (a) and (b). It is noteworthy that the reaction was significantly inhibited after the addition of both IPA and BQ. According to other reports, IPA is photocatalytically oxidised to acetone by CN photocatalysts \n [53] \n , hence it is likely that IPA is the preferred substrate leading to the inhibition of the β‐5 conversion reaction. To exclude the interference of IPA and to verify whether ⋅OH was generated and involved in the reaction, coumarin was used as a hydroxyl radical probe in this study.[ \n 54 \n , \n 55 \n ]\n Figure 8 Results of radical quenching experiments for P25 and CN (a) in acetonitrile/water (v/v 50 : 50) and acetonitrile (b). Hydroxyl radical yield comparison between P25 and CN in acetonitrile/water (v/v 50 : 50) and CN in acetonitrile with and without β‐5 (c). To detect the production of ⋅ OH radicals, coumarin was used as a probe. Coumarin can trap ⋅ OH and produce hydroxycoumarin compounds with 7‐hydroxycoumarin fluorescing at 405 nm, and the presence and intensity of fluorescence can be used to determine the generation and the concentration of ⋅ OH radicals, as shown in Figure 8 (c). In mixed acetonitrile/water solutions no 7‐hydroxycoumarin was generated by the CN photocatalyst in acetonitrile and acetonitrile/water solutions in the absence of β‐5. Interestingly, 7‐hydroxycoumarin was, however, detected when β‐5 was added to the reactor. Since CN cannot produce ⋅OH radicals in pure acetonitrile, 7‐hydroxycoumarin could only result from hydrogen peroxide produced by the reduction of oxygen by CN, which subsequently produces hydroxyl radicals. Moreover, in the process of generating hydrogen peroxide, protons were likely supplied by β‐5. These results would suggest that ⋅O 2 \n − are the dominant ROS in the photocatalytic degradation reaction of β‐5 using a CN photocatalyst. \n (1) \n \n \n \n (2) \n \n \n Additionally, it was observed that the dosage of the photogenerated hole scavenger (ammonium oxalate and EDTA Na) accelerated the photocatalytic degradation of β‐5 both in pure acetonitrile and in the mixed acetonitrile/water solution. The fact that the reaction was accelerated by the depletion of holes suggests that the hole scavenger promoted the separation of photogenerated hole‐electron pairs, thus enhancing the generation of ⋅O 2 \n − . Considered alongside the results of other radical quenching experiments, superoxide radicals appear to be the predominant ROS involved in the CN photocatalytic degradation reaction of β‐5. Combining all the above results, a schematic of the mechanism of the photocatalytic degradation of the β‐5 substrate is proposed in Figure 9 . It is proposed that following light excitation of the CN photocatalyst, charge separation occurs in the CN catalyst then photogenerated electrons and holes are generated on the surface. Photogenerated electrons on the CN surface subsequently reduce oxygen into a ⋅ O 2 \n − . Meanwhile, photo‐generated holes extract protons from β‐5, ultimately combining with ⋅ O 2 \n − to generate hydrogen peroxide. Product P2 was derived from P1 and accounts for the cleavage in side‐chain olefin. According to Zhang et. al., \n [56] \n The double bond of the olefin is cleaved to form an aldehyde with CN, under blue LED light conditions with acetonitrile as the solvent, as a result of direct hole oxidation and reaction via superoxide radical anions generated via the valence band reduction of oxygen. Therefore, the side‐chain olefin P1 is oxidized into P2 . The aldehyde group is subsequently further oxidized to a carboxyl group, resulting in product P3 . As for P5 , the introduction of the aldehyde group may be generated by the involvement of superoxide radicals in the reaction.\n Figure 9 Schematic of proposed β‐5 conversion mechanism on CN. Additionally, acetonitrile not only functions as a solvent but is also proposed to participate in the transformation reaction of β‐5. The cyano group introduced in product P4 is likely derived from the solvent acetonitrile. According to Addamo et al., \n [57] \n the reaction of ⋅ OH with acetonitrile has the ability to produce cyano radicals. It was however not possible to confirm the presence of cyano radicals in this study, but this will be investigated later using isotopically labelled acetonitrile solvents. In the absence of such a protocol for cyano radical detection, it is still our belief that the generation of product P4 arises from the solvent participating in the reaction. With this in mind, it is thought that CH 3 CN is likely to also have a significant impact on the degradation rate and pathways of product formation"
} | 5,739 |
39746978 | PMC11695677 | pmc | 6,915 | {
"abstract": "Polymer gels have been widely used in flexible electronics, soft machines and impact protection materials. Conventional gels usually suffer from the inherent conflict between stiffness and toughness, severely hampering their applications. This work proposes a facile yet versatile strategy to break through this trade-off via the synergistic effect of crystal-domain cross-linking and chelation cross-linking, without the need for specific structure design or adding other reinforcements. Both effects are proven to boost the mechanical performance of the originally weak gel, and result in a stiff and tough conductive gel, achieving significant enhancements in elastic modulus and toughness by up to 366-, and 104-folds, respectively. The resultant gel achieves coordinatively enhanced stiffness (110.26 MPa) and toughness (219.93 MJ m −3 ), reconciling the challenging trade-off between them. In addition, the presented strategy is found generalizable to a variety of metal ions and polymers, offering a promising way to expand the applicability of gels.",
"introduction": "Introduction The design and construction of polymeric gels for promising applications in the fields of tissue engineering 1 , energy storage 2 , soft robots 3 and flexible electronics 4 have attracted intensive attention. Because of the inherent softness and stretchability of conventional synthetic polymer, gels based on poly(vinyl alcohol) (PVA) 5 , polyacrylamides (PAM) 6 , poly(acrylic acid) (PAA) 7 , and polydimethylsiloxane (PDMS) 8 are paving the way toward versatile application scenario. For example, Zhang et al. reported a class of ultra-stretching, soft, and self-adhesive polymeric gel to wearable electronics for high-quality physiological signal monitoring 9 . Among these, little attention has been paid to impact-resistant protective materials from polymeric gels, when it comes to discussing practical applications (head impact protection device, ballistic armor, back protectors, safety barriers, etc.) where long service life, high loading capability and/or impact tolerance are highly demanded, because they are normally soft and fragile 10 . Effective strategies to improve their mechanical properties are urgently needed. Meanwhile, gels also need to fulfill further requirements, such as a wide working temperature range (−80°C–25°C), fracture resistance (> 250 kJ m –2 ), good ionic conductivity (> 0.1 S m −1 ), etc., in order to function under critical conditions 11 . Therefore, the preparation of stiff and tough conductive polymeric gels with multifunctional properties still represents a challenge for existing material systems. In recent years significant breakthroughs have been achieved in the stiffening and toughening of gel materials mainly include molecular engineering and micro/nano structural engineering. The former includes the preparation of homogeneous gel materials by multiple networks (double-network and interpenetrating polymer network) 12 , dynamic interactions (hydrogen bonds, coordination bonds) 13 , designed molecular structures 14 , developing novel cross-linkers (chemical and physical) 15 , and others 16 , resulting in the sufficiently enhanced mechanical behavior of gel materials. However, the irreversibility of covalent networks and the poor mechanical strength of dynamic bonds limit the molecular engineering of strong gels. The latter includes the construction of anisotropic micro–nano structures by in situ methods such as directional freezing 17 , pre-stretching 18 and mechanical training 19 , whereas these methods always involve complicated and multi-step preparation routes. In addition, enhanced microregions of gels are introduced by phase-separation and nanocomposites 20 , salting-out treatment 21 , which improve the strength and toughness in the counterpart of a homogeneous network. However, the deformability of gels is sacrificed. The method is unfavorable in terms of cost and energy consumption. The stiffness and toughness of gels prepared by the above strategies fails to fully trigger their deformation-resisting and energy-dissipating capacities, leaving it still challenging to reconcile the inherent trade-off between stiffness and toughness. Therefore, the development of a universal strategy for constructing both high stiff and tough with multifunctional gels is imperative. The advent of deep eutectic solvents (DESs) offers a great opportunity to the above challenges 22 . In general, DESs is a mixture of multiple components formed under hydrogen bonding and possess low freezing point (Supplementary Table 1 ), low vapor pressure and high-electrical conductivity (Supplementary Table 2 ). Moreover, DESs have some characteristics similar to ionic liquids (ILs) such as tunability, thermal/chemical stability and wide electrochemical window 23 . Additionally, DESs have significant advantages over ILs such as low cost, green, and facile preparation 24 . Therefore, DESs is becoming a favorable medium for building gels in extreme environments. However, the defects in mechanical properties, such as low stiffness (< 10 MPa) and poor toughness (< 20 MJ m −3 ), remain current obstacles for functional improvements and applications. The stiffness and toughness of networked gel materials largely depend on the characteristics, number density, and spatial distribution of cross-linking points 25 . Based on this, designing robust yet dynamic cross-linking domains (e.g., crystalline domains, micellar microspheres, and folded protein domains) and incorporating dynamic energy dissipation mechanisms are keys to creating gels with both high stiffness and toughness 26 . In detail, at the initial deformation stage, all robust cross-linking domains behave as high-energy barriers to minimize the network mobility, bonds in dynamic energy dissipation domains randomly broke to absorb energy, significantly increasing the initial elastic modulus; at the larger deformation stage, the adaptive disentanglement of all robust cross-linking domains within homogeneous networks effectively dissipates energy until sample failure, and bonds in dynamic and reversible energy dissipation domains can withstand strain attributed to the reduction of cross-linking density, both serving as a continuous source of toughening. Based on the above considerations, here we demonstrate a synergy of homogeneous distribution yet high number density of robust crystalline domains by phase separation and dynamic cross-linked network created by a large number of metal coordination sites, simultaneously tuning network structure, enhancing stiffness and toughness, and offering functions. In detail, a simple solvent exchange method was applied to convert weak PVA-carboxymethyl cellulose (CMC) composite gels to both stiff and tough gels in a deep eutectic solvent containing metal ions (DESs-M). Crucially, the rational regulation of noncovalent interactions and, thereby, the conformation of polymer chains via solvent exchange can lead to homogeneous distribution and robust crystalline domains for meanwhile dynamic and reversible physical cross-linked domains induced by the metal coordination sites, achieving rigid phase and soft phase. As expected, the as-prepared gel collectively renders good stiffness (110.26 MPa) and toughness (219.93 MJ m −3 ) to balance the inverse relation of those two. In addition, it is universal and directly applicable to DESs-M liquid containing different metal ions and polymers, offering a promising direction for stiff and tough polymeric gels or other soft materials.",
"discussion": "Discussion In this study, polymer network structure was adjusted to prepare both stiff and tough gels (FTD-M) via the synergy of homogeneous distribution and high number density of robust crystalline domains by phase separation and dynamic cross-linked network created by metal coordination sites. In detail, we first have synthesized a deep eutectic solvent containing various metal ions (DESs-M), and then a facile and universal solvent-exchange method was used to achieved an unusual combination of mechanical properties. The obtained FTD-C gel exhibited impressive stiffness (110.26 MPa), high toughness (219.93 MJ m −3 ) and strength (43.51 MPa), which effectively circumvented the long-standing incompatibility between stiffness and toughness. Furthermore, the gel also demonstrated good cushioning effect, impact resistance, wide temperature-tolerance and conductivity. Considering that the aforementioned design principle is based on generic polymer networks, it should be applicable to imparting competitive mechanical performance to other soft materials comprised of polymer networks, thereby extending their scope of applications, such as tissue engineering, vibration absorbers, soft robotics, and smart wearable devices."
} | 2,194 |
37671025 | PMC10475793 | pmc | 6,917 | {
"abstract": "Summary The metabolic “handshake” between the microbiota and its mammalian host is a complex, dynamic process with major influences on health. Dissecting the interaction between microbial species and metabolites found in host tissues has been a challenge due to the requirement for invasive sampling. Here, we demonstrate that secondary electrospray ionization-mass spectrometry (SESI-MS) can be used to non-invasively monitor metabolic activity of the intestinal microbiome of a live, awake mouse. By comparing the headspace metabolome of individual gut bacterial culture with the “volatilome” (metabolites released to the atmosphere) of gnotobiotic mice, we demonstrate that the volatilome is characteristic of the dominant colonizing bacteria. Combining SESI-MS with feeding heavy-isotope-labeled microbiota-accessible sugars reveals the presence of microbial cross-feeding within the animal intestine. The microbiota is, therefore, a major contributor to the volatilome of a living animal, and it is possible to capture inter-species interaction within the gut microbiota using volatilome monitoring.",
"introduction": "Introduction The gastrointestinal tract is colonized by a complex ecosystem of microbes, referred to as the gut microbiota, which plays a major role in modulating host metabolism and immune function. 1 Exchange of metabolites between the microbes and their host is an integral mechanism in this interaction. 2 , 3 Correspondingly, our ability to measure the production of small molecules produced by a host and its microbiota has the potential to advance our understanding of this highly complex relationship. 4 , 5 Studies on plasma have shown microbiota-dependent effects on a variety of host metabolites, including on amino acids, 3 , 6 fatty acids, 7 , 8 lipids, bile acids, 9 and xenobiotics. 10 When conventionally colonized mice are compared with germ-free (GF) mice, 10% of identified features in common between these mice differ in signal intensity. 11 Furthermore, it was previously suggested that plasma metabolites in the host are predictive of human gut microbiota diversity, and metabolites in the host can be used as predictors of microbial phyla abundance. 12 To date, most metabolomics studies have focused on quantification of metabolites in body fluids, often requiring invasive sampling. Moreover, the easily accessible fecal metabolites are often used as a proxy for microbial metabolism that is most active in the upper large intestine. Acquiring meaningful, time-resolved data on microbial activity in live animals is thus challenging. However, this can be done by monitoring small-molecule metabolites released by the experimental animals to the gas phase, including volatile, semi-volatile, and low-volatility species (hereafter, for simplicity, referred to as the “volatilome”). Time-resolved monitoring of host metabolism is possible using sensors measuring light gases released by the organism (e.g., H 2 , CH 4 , O 2 , and CO 2 7 , 13 , 14 , 15 ). These data allow monitoring of some aspects of host and microbiota metabolism, such as host energy expenditure, respiratory quotient, and microbial hydrogen production. However, due to the limited available sensors, only a small number of simple gases or volatiles can currently be studied using these methods. Ambient mass spectrometry (MS) methods are ionization methods that operate under ambient conditions and require minimal sample preparation. 16 Progress in ambient MS now renders it possible to non-invasively monitor the volatilome in breath 17 , 18 , 19 and exhaust gases. 20 , 21 Notably, this includes the ability to detect short-chain fatty acids (SCFAs), which are of known interest in host-microbe interactions. SCFAs have already been extensively studied, for example in epithelial gene expression and metabolism. 22 This suggested the possibility to apply ambient MS technology to the study of microbiome function. Secondary electrospray ionization-mass spectrometry (SESI-MS), which has been developed in the past decades, 23 is compatible with high-resolution MS and has a relatively fast response and broad detection range 24 , 25 when compared with conventional MS-based detection methods for volatiles. For example, SESI-MS allows direct detection without pre-concentration and separation (compared with GC-MS, 26 which requires pre-concentration and separation), it can detect volatiles ionized in both positive and negative modes (compared with proton-transfer reaction MS, 27 which only allows positive ion mode), and it can be easily mounted on a high-resolution MS (compared with selected-ion flow-tube MS, 28 which only provides unit mass resolution). SESI-MS allows detection of molecules in the gas phase down to sub-ppt concentrations with minimum detectable vapor pressure down to 10 −7 bar. 29 , 30 It is therefore the method of choice for efficient, real-time detection of volatile metabolites in biological systems. Previously, SESI-MS has been used for identifying biomarkers for various lung pathogens in human breath 31 and to capture the metabolic responses of Staphylococcus aureus to antibiotic treatment in the headspace of microbial batch cultures. 32 However, due to the high complexity of the gut microbial community, it is usually difficult to assign metabolic signals to a specific origin and to decipher causal relations between changes in bacterial and host metabolism; thus, the application of SESI-MS in gut microbiota research is yet to be explored. Here, we present a SESI-MS-based approach to simultaneously study microbiome and host metabolism by monitoring volatilome released from live animals, in which we can manipulate the microbial composition: ranging from GF via a simplified 3-species microbiota to a complete “specific-pathogen-free” (SPF) microbiota. By first identifying volatile markers for each member of the reduced microbiota, we were able to track bacterial metabolism inside a host non-invasively while at the same time monitoring the changes in host metabolism. Feeding of microbiota-available heavy-isotope-labeled sugars allowed us to unambiguously identify the microbial origin of volatilome and to directly observe microbial cross-feeding in live animals. Applying this approach to living animals expanded our ability to understand the host-microbiota interaction and could potentially enable us to study the interaction in a continuous way.",
"discussion": "Discussion In this study, we have shown that SESI-MS of the mouse volatilome is a powerful, non-invasive tool to analyze host-microbiota interactions. By comparing the SESI-MS signatures of the host-microbiota system in gnotobiotic animals with those of the colonizing microbes in isolation, we were able to identify the metabolite footprint of the most abundant microbes ( Figure 3 D). By feeding with isotope-labeled sugars that can only be used by the microbes and not by the host, we were able to conclusively show the presence of microbial metabolites in the volatilome and could use these metabolites to demonstrate within-host cross-feeding between intestinal bacteria. It should be noted that SESI-MS analysis of air around an animal clearly does not, and is not expected to, detect all produced metabolites. Very small metabolites, such as H 2 and CO 2 , have too low a molecular weight to be measured by our mass spectrometer, and high-molecular-weight metabolites present in the gas phase with extremely low concentration fall below the detection limit of SESI-MS. However, within the accessible molecular weight window for observing host-microbiota metabolism, we were able to recapitulate metabolic changes upon microbial colonization of a GF mouse that have previously been measured in host tissues and plasma 7 , 56 (e.g., amino acid metabolism, xenobiotic metabolism; Table S2 ). In colonized mice, bacterial numbers are always very high (around 10 11 cells/g; Figure 2 A), and we have shown that we can measure abundant metabolic interactions non-invasively in this system. Whether it would also be possible to measure less abundant metabolic products in a host-microbiota system, e.g., metabolic interactions on a lower trophic level, remains to be tested. Given that SESI-MS is non-invasive and does not preclude the later analysis of, e.g., host tissue and that many metabolic pathways have profound circadian rhythms, a potential application is therefore the identification of the optimal time points for endpoint analysis. 20 We envisage one of the most powerful features of the SESI-MS approach to be the possibility of collecting time-resolved data on metabolic processes of a host-microbiota system in an undisturbed way. This has, up to now, only been possible by collecting fecal samples, which comes with a set of limitations. First, fecal samples are not representative of active microbial metabolism, 57 which happens largely in the upper part of the large intestine. 58 , 59 , 60 Second, repeated fecal sampling, even though minimally invasive, will disturb an animal’s routine if conducted over longer time periods and will therefore affect its behavior and potentially the experimental outcomes. And third, sampling of fecal material is limited in its time resolution, which can lead to gaps in information when measuring the effects of experimental interventions that work with a time delay, which is true for most experimental interventions with an effect inside the host. The SESI-MS approach presented here is not affected by any of these issues. Given the availability of the equipment, the workflow as presented here is relatively easy to implement and allows room for customization. SESI sources are commercially available and can be interfaced with commercial high-resolution mass spectrometers. 24 , 25 , 61 The data generated from SESI-MS are compatible with standard software for direct-injection MS-based metabolomics, 61 , 62 and online MS2 fragmentation for compound identification can be performed. 63 , 64 Development is still needed for collecting and processing time series data since most current MS-based metabolomics datasets have low time resolution or do not consider time-dependent changes at all. One example where time series of SESI-MS data at a minute-level resolution has been successfully used was for continuous monitoring of human metabolism during sleep. 61 In conclusion, we describe a method for tracking metabolic changes in host-microbiota systems non-invasively. This method has the potential to generate metabolomic data for such systems at an unprecedented time resolution and with minimal to no system perturbation. Applying this system to gnotobiotic and GF mice has considerable potential to reveal relevant microbiota functions and to improve our understanding and diagnosis of microbiota-associated conditions. Limitations of the study Here, we have validated the SESI-MS-based method using gnotobiotic mice, for which we have extensive control over microbiota composition and extensive knowledge of the species present. For animals with an undefined microbiota, it will require considerable further research to link microbial metabolites to individual gut bacterial species. Additionally, comparing animals with widely differing microbiota may be associated with the typical confounders in direct-injection MS analysis relating to matrix effects on ionization efficiency of different compounds. 65 , 66 , 67 Future developments of gas-phase standards of the most relevant molecules would allow tracking of these effects and absolute quantification of volatile-molecule concentrations. It should also be noted that the current setup does not allow the animal to be monitored continuously, first because the mice tube is a confined space and second because the accumulation of waste (e.g., feces, urine) could accentuate the ion suppression and ion competition of SESI 68 and therefore interfere with data acquisition."
} | 2,982 |
24885047 | PMC4055215 | pmc | 6,919 | {
"abstract": "Background Trans- 4-hydroxy-L-proline ( trans -Hyp), one of the hydroxyproline (Hyp) isomers, is a useful chiral building block in the production of many pharmaceuticals. Although there are some natural biosynthetic pathways of trans -Hyp existing in microorganisms, the yield is still too low to be scaled up for industrial applications. Until now the production of trans -Hyp is mainly from the acid hydrolysis of collagen. Due to the increasing environmental concerns on those severe chemical processes and complicated downstream separation, it is essential to explore some environment-friendly processes such as constructing new recombinant strains to develop efficient process for trans -Hyp production. Result In this study, the genes of trans- proline 4-hydroxylase ( trans- P4H) from diverse resources were cloned and expressed in Corynebacterium glutamicum and Escherichia coli , respectively. The trans -Hyp production by these recombinant strains was investigated. The results showed that all the genes from different resources had been expressed actively. Both the recombinant C. glutamicum and E. coli strains could produce trans -Hyp in the absence of proline and 2-oxoglutarate. Conclusions The whole cell microbial systems for trans -Hyp production have been successfully constructed by introducing trans- P4H into C. glutamicum and E. coli . Although the highest yield was obtained in recombinant E. coli , using recombinant C. glutamicum strains to produce trans -Hyp was a new attempt.",
"conclusion": "Conclusions In this study, two new and a modified trans -P4Hs were expressed in C. glutamicum and E. coli successfully. Different amount of trans- Hyp were produced by these recombinant strains detected. Although the yield in recombinant C. glutamicum was less than that in recombinant E. coli , C. glutamicum as a native proline producing strain was worthy of further optimizing. This is the first report of producing trans- Hyp by introducing L-proline 4-hydroxylase into C. glutamicum.",
"discussion": "Results and discussion Construction of trans -Hyp producing recombinant strains There are several genes being speculated as putative L-proline 4-hydroxylase gene in the database, including genes in Pseudomonas stutzeri \n , \n Janthinobacterium sp ., Bordetella bronchiseptica RB50, Bradyrhizobium japonicum, Achromobacter xylosoxidans C54 and Dactylosporangium . sp. Using PCR, we cloned and obtained the putative genes of P4H from P. stutzer and B. bronchiseptica RB50 , named p4hP and p4hB. They were ligated to the corresponding plasmids after digestion and converted to C. glutamicum and E. coli , respectively. The length of p4hP was 918 bps while p4hB was 924 bps. These sequences were 100% identical to the reported genes in NCBI. The gene of trans -P4H from Dactylosporangium sp. ( p4hD ) had been expressed in E. coli successfully and can transform L-proline with good enzymatic properties [ 11 - 13 , 24 - 26 ]. The length of p4hD was 816 bps encoding a 272-amino-acid polypeptide with the molecular weight of 29,715 daltons [ 11 , 26 ]. In this study, p4hD was applied with some modifications on the nuclear bases. The original gene sequence of p4hD was analyzed ( http://www.kazusa.or.jp/codon/ ) and the results showed there were some rare codons for both C. glutamicum and E. coli. It has been reported that rare codons are strongly associated with low level of protein expression [ 27 ]. Codon optimization for heterologous protein expression has often been shown to drastically increase protein expression [ 28 ]. Thus, the rare codons of p4hD gene were substituted for those used with high frequency in C. glutamicum and the GC content was adjusted from 73% to 61% through synonymous conversion, which was close to that of C glutamicum . The modified gene of p4hD was synthesized according to the above modifications (Additional file 1 ). The expression of P4H is one of the important aspects on the construction of trans-Hyp biosynthetic pathway. Figure 2 shows the SDS-PAGE of trans -P4Hs expressed in recombinant C. glutamicum and E. coli. All the recombinant trans -P4Hs were expressed as soluble proteins without inclusion bodies. It was obviously that the recombinant trans- P4Hs in E. coli were expressed much more than those in C. glutamicum (Figure 2 ) . Many factors have influences on the expression of foreign proteins including promoters, the host-vector system and cultural conditions etc. [ 29 , 30 ]. Since it was the first to express trans- P4Hs in C. glutamicum , more comprehensive studies such as promoter selection and culture condition optimization will be considered in our future work. Figure 2 Expression of trans -P4Hs in different strains. A1: E. coli BL21/pET28a- p4hD ; A2: E. coli BL21/pET28a- p4hP ; A3: E. coli BL21/pET28a- p4hB . B1: C. glutamicum ATCC15940/pECXK99E- p4hD ; B2: C. glutamicum ATCC21355/pECXK99E- p4hD ; B3: C. glutamicum ATCC21157/pECXK99E- p4hD ; B4: C. glutamicum 49-1/ pECXK99E- p4hD . Comparison of P4H activities Oxygenases are widely applied in industry since they can catalyze the highly specific oxyfunctionalization of unactivated C-H bonds under mild conditions, especially transferring molecular oxygen to a substrate [ 31 ]. P4H belongs to a family of 2-oxoacid-dependent dioxygenase, which is a monomeric protein and utilizes the monomeric rather than polymeric substrates [ 10 ]. The activities of trans- P4Hs using recombinant whole cells in this study were measured (Table 1 ). Our data indicated that the expressed protein level and enzymatic activity was higher at 30°C. The results also showed that the plasmids were very stable as the plasmid stabilities of recombinant E. coli and C. glutamicum strains were all more than 98% at the end of fermentation. Table 1 Comparison of \n trans \n -P4Hs activities and \n trans \n -Hyp production by different recombinant \n C. glutamicum \n and \n E. coli \n strains Strains Specific activities (U/mg · wet cell weight) \n Trans \n -Hyp (g/L) OD600 (OD620)* C. glutamicum ATCC13032/pEC-XK99E- p4hD 37.4 ± 1.4 0.072 ± 0.001 5.5 ± 0.7 C. glutamicum ATCC13032/pEC-XK99E- p4hP 20.7 ± 1.1 0.106 ± 0.002 7.3 ± 0.5 C. glutamicum ATCC13032/pEC-XK99E- p4hB 40.7 ± 0.8 0.079 ± 0.016 5.4 ± 0.03 C. glutamicum ATCC15940/pEC-XK99E- p4hD 12.9 ± 0.5 0.103 ± 0.001 14.0 ± 0.2 C. glutamicum ATCC21355/pEC-XK99E- p4hD 35.9 ± 0.1 0.087 ± 0.005 6.6 ± 0.2 C. glutamicum ATCC21157/pEC-XK99E- p4hD 12.3 ± 0.9 0.112 ± 0.004 13.3 ± 0.1 C. glutamicum 49-1/pEC-XK99E- p4hD 12.4 ± 0.6 0.113 ± 0.001 13.8 ± 0.5 E. coli BL21/pET-28a - p4hD 60.4 ± 1.8 0.470 ± 0.028 6.5 ± 0.2 E. coli BL21/pET-28a - p4hP 22.2 ± 0.5 0.126 ± 0.007 7.3 ± 0.05 E. coli BL21/pET-28a - p4hB 50.0 ± 2.2 0.115 ± 0.006 6.9 ± 0.1 *Optical density at 600 nm for E. coli and 620 nm for C. glutamicum. The recombinant cells with expressing of different genes showed different levels of catalytic activities toward L-proline. The activity of trans- P4H expressed by E. coli BL21/ pET28a- p4hD was the highest among all the constructed recombinant strains. The new cloned and expressed genes from P. stutzeri and B. bronchiseptica also showed interested activities. As for different host strains, E. coli represented better than C. glutamicum , which may be related to the performance of corresponding plasmid. Four L-proline producing strains of C. glutamicum were used as expression host strains and the resulted recombinant strains showed different enzymatic activities. The highest specific enzymatic activity among C. glutamicum strains was 40.7 U/mg · wet cell by C. glutamicum ATCC13032/pEC-XK99E- p4hB . However, the specific enzymatic activity of recombinant E. coli /pET28a - p4hD was up to 60.4 U/mg · wet cell. The growth of three recombinant E. coli strains was similar. But there was significant difference among the recombinant C. glutamicum strains. The recombinant C. glutamicum strains with higher specific enzymatic activities grew less than those with lower specific enzymatic activities. Additionally, the enzymatic activity of E. coli BL21 /pET28a - p4hD was similar to that of E. coli W1485/pWFH1 and higher than that of E. coli BL21/pET24- p4h 1 of [ 12 , 13 ]. The p4hD in E. coli W1485/pWFH1 was the original one in Dactylosporangium sp ., while p4hD in E. coli BL21/pET24- p4h 1 was modified. Although the codon optimization in this study was designed for C. glutamicum , the results indicated that it was also successfully in E. coli. Trans -Hyp production in flasks The production of trans- Hyp by different recombinant C. glutamicum and E. coli strains was also shown in Table 1 . The yields of trans- Hyp by these recombinant strains depended both on the enzymatic activity of P4H and cell growth. E. coli BL21/ pET28a- p4hD had the highest yield, which was coincided of its specific enzymatic activity. Although the recombinant E. coli strains grew similarly in the production medium, there was significant difference in the production of trans- Hyp which did not keep the same level with the specific enzymatic activities. The productions of trans- Hyp by recombinant C. glutamicum strains were also much less than that of E. coli BL21/pET28a- p4hD. It was due to both the less expression of trans- P4H and less cell growth in C. glutamicum . The L-proline production of four C. glutamicum strains was also less than 1 g/L. There was little difference of trans- Hyp production among the recombinant strains of C. glutamicum with same gene p4hD , despite that some strains had better enzymatic performance and proline production. Using the recombinant strains to directly synthesize trans- Hyp from glucose via fermentation was achieved since both the enzymes and precursors needed in the process were available. The over expressed foreign trans- P4Hs catalyzed the hydroxylation of L-proline at the trans -4 position, while 2– ketoglutarate was supplied by glucose through TCA cycle and then oxidatively decarboxylated to succinate (Figure 1 ). It was reported that proline was demanded in the production of Hyp by recombinant E. coli only with p4h gene. The carbon in proline added during the fermentation only flowed into amino acids synthesized from TCA cycle intermediates and not into gluconeogenesis [ 13 ]. However, the accumulated Hyp was at a relative high level even without the addition of proline in this study. It could be understood that Corynebacterium had the powerful biosynthetic pathway of proline [ 32 , 33 ]. The biosynthetic pathway of proline has also been identified in E. coli , which may contribute to the synthesis of trans -Hyp by recombinant E. coli strains. The modification of proline pathway in E. coli enhanced the yield of Hyp, whereas the formation of Hyp can also relieve the feedback inhibition of proline [ 24 ]. The amount of proline (0-4 mM) promoted the production of trans- Hyp. However, continuously addition of L-proline didn’t improve the production yield significantly (Table 2 ). In this study, the time of cultivation was significantly less than those reported in the literatures, which might attribute to the different media used and also indicated there was great potential with optimization. In fact, 2.28 g/L of trans- Hyp was produced by recombinant E. coli without adding L-proline in flasks with a little modification of media and 6.72 g/L was achieved by supplement only 4 mM L-proline. Table 2 Hyp production under different L-proline supplementation Supplementary addition of L- proline (mM) 0 1 2 4 8 12 Hyp (g/L) 2.28 2.81 3.25 6.72 5.56 6.32 In order to further increase the biosynthesis of trans -Hyp by recombinant C. glutamicum and E. coli , alternative approaches should be considered as well. In E. coli , the degradation of proline should be overcome. Although the trans -Hyp production by a putA mutant of E. coli was not improved furthermore, the yield based on the proline utilized was enhanced greatly. In both C. glutamicum and E. coli , the expression of recombinant P4H as one of the oxygenases is involved in the physiological metabolism of host cells including the cofactor, co-substrate and oxygen. Moreover, without a powerful proline synthetic pathway in E. coli, the availability and transportation of substrate will limit the transformation seriously."
} | 3,115 |
39835267 | PMC11743900 | pmc | 6,921 | {
"abstract": "Highlights • Plant growth-promoting microorganisms (PGPM) as microbial inoculants and biofertilizers. • PGPM mechanisms in enhancing plant growth, nutrient uptake, and soil health. • The interactions between plants and PGPMs, emphasizing their role in crop productivity. • PGPM-mediated stress resistance and soil health improvement.",
"conclusion": "9 Conclusion This review paper aims to provide a comprehensive understanding of plant-microbe interactions and their significance in sustaining crop productivity and soil fertility. The underlying mechanisms, exploring influencing factors, and discussing practical implications and future directions, it seeks to inform researchers for advancing sustainable agricultural practices. Promising approaches for ecologically sustainable farming may result from beneficial microbial-plant interactions, in sustainable agriculture, the development of biofertilizer, biocontrol, and bioremediation agents has been greatly aided by the interaction between plants and microbes. And to improving plant nutrition and production, PGPM is crucial for maintaining ecological stability, an enhancing plant health, a variety of interacting microorganisms shield plants from biotic and abiotic stress. It is clear from this that the interacting microbes work to maximize a variety of biological processes occurring in the soil in order to provide a thriving, healthy environment that guarantees the crop will receive enough nutrients. Nonetheless, public education on PGPM's application in agriculture and its wider use is imperative.",
"introduction": "1 Introduction Sustainable agricultural methods are essential in the face of growing environmental pressures, both biotic and abiotic, that have an impact on plant productivity worldwide ( Khan and Lazali, 2023 ). For the goal to assure food security, agricultural output must be intensified in order to attain higher crop yields and overall production due to the world population's rapid expansion ( FAO, 2017 ). However, due to the overuse of synthetic chemical pesticides and fertilizers, which worsen environmental degradation and pose health hazards to humans ( Babin et al., 2019 ), agriculture is one of the human activities that most strongly contributes to the rise in chemical pollutants. On the other hand, the implications of climate change on interactions between plant pathogens are complex, with the ability to modify pathogen biology, host development, and disease severity. These changes have the potential to increase or decrease the incidence of diseases. Crop yields are negatively influenced by climate change-related issues like flooding, droughts, and extremely high temperatures. Low soil moisture levels and changes in plant physiology have been brought about by heat waves, severe droughts, and a shortage of water. Crops must have increased nutritional value, be resistant to disease, and be able to withstand stress from heavy metals, salt, and drought in order to achieve sustainable agriculture. Microbes, which can be parasitic or mutualistic organisms, colonize the plant's above- and below-ground parts ( Nadarajah and Rahman, 2021 ). These microbes may affect plants' growth and well-being in a beneficial, neutral, or detrimental way ( Smith and Goodman, 1999 ; Berg et al., 2005 ). Microorganisms in the soil play a crucial role in protecting plants from stress by regulating phytohormones and improving the absorption of nutrients, among other processes that promote crop development and yield ( Tkacz and Poole, 2015 ). Moreover, by inducing systemic resistance mechanisms in plants, these microbes increase their resilience to biotic stressors. The microbial interaction with plants acts as a catalyst in agriculture, enhancing output on its own ( Nelson, 2004 ; Bhattacharyya et al., 2016 ). Plant growth-promoting microorganisms (PGPMs), particularly bacteria and fungi, present a practical means of achieving these goals ( Rasool et al., 2021 ). These microorganisms can enhance a plant's nutrient-absorption capacity ( Larsen et al., 2015 ) and water-use efficiency ( Armada et al., 2014 ), as well as foster resistance against plant diseases ( Turner et al., 2013 ; Kumar and Verma, 2018 ). Plant-microbe interactions are an achievable path towards agricultural sustainability since they are essential to maintaining soil fertility and crop productivity ( Oldroyd, 2013 ). Long-term soil fertility and agricultural crop yield depend on interactions between plants and microbes ( Parniske, 2008 ). Nevertheless, there are a lot of intricate relationships that are formed in the soil, especially in the rhizosphere, between the microorganisms, the crop, and the environment ( Agrahari et al., 2020 ). In order to enhance farming techniques, beneficial microorganisms can be added to the soil or inoculated. To improve crop health and yield and lessen the harmful effects of agrochemicals, microbial inoculants are administered into the soil or plants. It is an effective substitute for chemical treatment and may stabilize soil structure, manage diseases and pests, and encourage plant growth ( Pandey et al., 2019 ). These inputs can be used as biopesticides, biocontrol agents, bioherbicides, and biofertilizers ( Nadarajah and Rahman, 2021 and Rasool et al., 2024 ). This study examines the complex processes and modes of action underlying these interactions, as well as their bidirectional relationship and the particular areas in which they take place. In instance, beneficial bacteria play a variety of roles that are vital to the growth and well-being of plants. Since these PGPMs have both positive and antagonistic features that affect crop productivity and health, it is imperative to understand their method of action, the function of PGPMs in preserving soil fertility is investigated, focusing their contributions to the cycling of nutrients, the enhancement of soil structure, and general soil health. This work underscores how crucial it is to use plant-microbe interactions to address the problems of contemporary agriculture through a thorough investigation of these subjects."
} | 1,522 |
30886426 | PMC6636631 | pmc | 6,922 | {
"abstract": "Photosynthetic organisms provide food and energy for nearly all life on Earth, yet half of their protein-coding genes remain uncharacterized 1 , 2 . Characterization of these genes could be greatly accelerated by new genetic resources for unicellular organisms. Here, we generated a genome-wide, indexed library of mapped insertion mutants for the unicellular alga Chlamydomonas reinhardtii . The 62,389 mutants in the library, covering 83% of nuclear, protein-coding genes, are available to the community. Each mutant contains unique DNA barcodes, allowing the collection to be screened as a pool. We performed a genome-wide survey of genes required for photosynthesis, which identified 303 candidate genes. Characterization of one of these genes, the conserved predicted phosphatase-encoding gene CPL3 , showed it is important for accumulation of multiple photosynthetic protein complexes. Notably, 21 of the 43 highest-confidence genes are novel, opening new opportunities for advances in our understanding of this biogeochemically fundamental process. This library will accelerate the characterization of thousands of genes in algae, plants and animals."
} | 289 |
33910891 | PMC8081363 | pmc | 6,923 | {
"abstract": "An unusual and beautiful phenomenon in which life-like crystal structures spontaneously grow from nano-engineered materials.",
"introduction": "INTRODUCTION Many of the uses for water are intimately familiar to us. Drinking water, wash water, water for agriculture, and even water used for recreation have an omnipresent and essential impact on our lives. However, water’s impact and importance extend far beyond these everyday uses. In many developed countries, thermoelectric power production is one of the largest sources of water consumption ( 1 ), where it is used to cool reactors and transport heat. In 2015, 41% of all surface water withdrawals in the United States went toward cooling in thermoelectric power plants ( 2 ). Thermoelectric power accounts for 90% of all electricity generated within the United States and encompasses many forms of power production, including nuclear, coal, natural gas, and oil. In its role as a coolant, water is sprayed on, flown through, or otherwise placed in contact with hot equipment (pipes, tanks, reactors, etc.). Many cooling processes use evaporation as a vital part of heat exchange due to the large heat transfer associated with phase change. However, when water is evaporated, contaminants within the water (including minerals) will precipitate at the point of evaporation. Over time, accumulation of these impurities reduces heat transfer performance, blocks pipes, and generally causes material corrosion and deterioration ( 3 ). Mineral fouling, in particular, is a leading cause of equipment degradation and failure in heat exchange processes ( 4 ). To prevent mineral fouling, considerable effort and monetary investment goes toward pretreatment of coolant water using technologies such as ion exchange and reverse osmosis ( 2 ). Because of the ever increasing importance of water conservation ( 5 ), more and more water for thermoelectric cooling is being sourced from saline surface waters or from desalination waste brines rather than from freshwater sources ( 6 ), despite the associated increases in pretreatment costs. Surface engineering for control of salt-substrate interactions is therefore becoming increasingly attractive as a cost-effective alternative to pretreatment for preventing mineral fouling ( 7 ). The wetting properties of a material (also known as hydrophobicity and superhydrophobicity) have been the major focus of investigations seeking to eliminate fouling via surface engineering ( 8 ). However, mineral fouling is a multiphase problem, and interactions between the crystal and substrate and between the crystal and liquid are equally important for determining fouling propensity as are the interactions between the liquid and substrate (typically characterized by the contact angle) ( 9 – 11 ). One method of examining interactions between a liquid, surface, and crystallizing solute is via drop evaporation. Traditionally, drop evaporation experiments are motivated by applications in self-assembly, inkjet printing, and sensing/diagnostics ( 12 , 13 ). Previous investigations have demonstrated that this technique can also be used as a method of exploring crystal adhesion and interfacial properties ( 11 ) and can inform on how damage to surfaces caused by crystallization occurs ( 14 ). Interfacial properties ( 15 ) and nucleation barriers ( 16 ) associated with different crystal chemistries control deposit morphologies, allowing us to draw inferences regarding the antifouling properties of a material from drop evaporation experiments. Evaporating a drop of a volatile liquid containing a nonvolatile solute will induce crystallization of said solute due to rising concentrations, which eventually exceed the solubility limit. For solutes of low solubility, the patterns left by evaporative crystallization are similar to “coffee-ring” patterns formed by evaporation of a particle-laden drops ( 11 , 12 , 16 ). However, when the dissolved mass is excessive, three-dimensional crystal structures may arise ( 17 , 18 ). In particular, when a drop containing saturated sodium chloride is evaporated on a hydrophobic surface, “salt globes” form because of the propensity of crystals to nucleate at the air/water interface ( 18 ). These globes replicate the shape of the evaporating drop after reaching the solubility limit. Here, we report an unexpected and unusual phenomenon in which crystal structures formed from evaporating drops of water saturated with sodium chloride self-eject from heated, superhydrophobic surfaces ( 19 , 20 ). This self-ejection occurs via the growth of crystalline “legs” during the end phase of evaporation, which cause the entire crystal structure to eject from the surface ( 21 ). We term the resulting structures composed of the salt globe and legs “crystal critters” due to the eerie motions produced during self-ejection and to the resemblance of the crystal structures to biological forms ( 19 , 20 ). This remarkable effect could enable design of extreme antifouling systems for spray cooling of hot surfaces using concentrated brines produced during desalination. This effect is also of interest for drop levitation/transport applications, which have traditionally been accomplished by heating surfaces to temperatures far in excess of the fluid boiling point ( 22 ). In such Leidenfrost levitation, evaporative flows create a lubricating vapor cushion between the drop and surface ( 22 ). In contrast, the critter effect occurs at much lower temperatures (60° to 100°C) than previously observed in both the traditional Leidenfrost effect (200°C) and even for cold-regime Leidenfrost on superhydrophobic materials (~130°C) ( 23 – 25 ). We demonstrate that this low-temperature ejection is accomplished via cooperative effects of crystallization, evaporative flows, and nanoscale phenomena.",
"discussion": "DISCUSSION We have introduced and explained an unusual behavior exhibited by evaporating drops of saturated sodium chloride on heated, nanotextured superhydrophobic surfaces, in which salt structures self-eject via growth of legs. This effect is trigged by the dewetting of water from the low solid fraction substrate in favor of wetting salt crystals previously grown at the air/water interface. This dewetting event is only possible due to confinement of both crystallization and evaporative flows induced by nanotexture and will not occur for superhydrophobic surfaces composed of microscale features ( 18 , 21 ). Confinement and a dearth of adhesion point also enable ejection at lower temperatures than previously observed for evaporation-induced drop levitation ( 23 – 25 ). In addition to being innately interesting, the crystal critter effect has potential application for improving sustainability in spray cooling heat exchange by introducing a strategy for eliminating crystal adhesion on hot surfaces. By taking advantage of the effect, it may be possible to directly use saline waters as a heat transfer fluid while also avoiding mineral fouling of the heat exchange surface. Using saltwater rather than fresh water for cooling applications simultaneously reduces costs associated with water treatment while also preserving fresh water for other vital purposes. Furthermore, one might also imagine a new type of cogeneration plant in which desalinated seawater is produced as a by-product of heat exchange by recovering water vapor produced during critter formation. Another potential use for the effect could be for treatment of waste brines in zero liquid discharge systems, where complete recovery of water from very salty reverse osmosis reject water is challenging due to the difficulties in working with high salinity brines. Self-ejection of salt foulants is also of particular importance for marine vessels and coastal structures ( 42 ), where deposition of salt and subsequent crystal growth from seaspray is a leading cause of damage ( 43 ). More broadly, these insights on confinement-driven evaporative crystallization could also be applied for novel drop-based fluidic machines ( 44 ) or for self-propulsion ( 24 , 45 – 47 )."
} | 2,015 |
35017303 | PMC8784112 | pmc | 6,924 | {
"abstract": "Significance Under anoxic conditions, various microorganisms couple the oxidation of organic carbon to the reduction of solid ferric iron oxide phases using extracellular electron shuttles (EES). Determining the contribution of this widespread terminal electron accepting process to total anaerobic respiration has proven challenging because of large variations in observed ferric iron reduction rates. This study demonstrates that rates of ferric iron oxide reduction by EES can be rationalized based on a unifying relationship that links rates to the thermodynamic driving force for the least favorable electron transfer from the EES to ferric iron. The relationship derived herein allows for a generalized and precise assessment of the contribution of EES-facilitated ferric iron oxide reduction to organic matter decomposition in anoxic environments.",
"discussion": "Results and Discussion Free Energies of Fe(III) Reduction by EES. We illustrate the differences of reaction free energies for the reduction of Fe(III) oxides by the first vs. second electron transferred from EES for two widely used EES, anthraquinone-2,6-disulfonate (AQDS) and riboflavin. Fig. 1 A shows the E H –pH diagram for AQDS with colored lines depicting the reduction potentials, E H, x 0 ′ , for the half-reactions specified in Eqs. 1 – 3 at pH 4 to 8. The reduction potential of the first electron transferred to AQDS 2 − to form the semiquinone radical anion species AQDS • 3 − , E H, 1 0 ′ ( Eq. 1 ; orange line in Fig. 1 A ) is low and pH-independent given that the formed semiquinone is deprotonated over the environmentally relevant pH range considered here [i.e., p K a AQDSH • 2 − = 3.0 (31); SI Appendix , Fig. S1 ]. By comparison, the reduction potential of the second electron transferred to the semiquinone, E H, 2 0 ′ , is higher and has a slope of –0.118 V · pH - 1 , resulting from the stoichiometric transfer of two protons with this electron ( Eq. 2 ; blue line in Fig. 1 A ). Under these conditions, the thermodynamics of the single-electron transfer steps result in only very small concentrations of the transient semiquinone species. The reduction potential for the transfer of both electrons, E H, 1 , 2 0 ′ , is the average of E H, 1 0 ′ and E H, 2 0 ′ and has a slope of –0.059 V · pH - 1 ( Eq. 3 ; red line in Fig. 1 A ). [1] AQDS 2 − + e − ⇌ AQDS • 3 − [2] AQDS • 3 − + e − + 2 H + ⇌ AQDSH 2 2 − [3] AQDS 2 − + 2 e − + 2H + ⇌ AQDSH 2 2 − Fig. 1. Thermodynamic analysis of the reduction of goethite ( α -FeOOH) by AQDS and riboflavin over the pH range examined in this work. ( A and C ) Pourbaix diagrams for redox couples quinone/hydroquinone ( E H,1,2 0 ′ , red), semiquinone/hydroquinone ( E H,2 0 ′ , blue), and quinone/semiquinone ( E H,1 0 ′ , orange) of AQDS and riboflavin. The redox couple α -FeOOH/Fe 2 + ( E H (oxide)) is shown as a gray line. E H 0 ′ –pH diagrams were drawn using published standard reduction potentials [AQDS ( 31 , 34 ), riboflavin ( 33 , 35 , 36 ), SI Appendix, Table S1 and 0.768 V ( 37 ) for α -FeOOH]. Molecular structures of redox-active species of AQDS and riboflavin are shown in SI Appendix , Figs. S1 and S2 , respectively. ( B and D ) Free energies, Δ r G x 0 ′ , of the two-electron transfer from reduced AQDS and riboflavin (x = 1,2: red lines) to α -FeOOH, and of the individual one-electron transfers; x = 2 (blue lines) stands for the first one-electron transfer from the fully reduced hydroquinone to α -FeOOH; x = 1 (orange lines) for the second one-electron transfer from the semiquinone to α -FeOOH. Δ r G x 0 ′ values were calculated from the difference in reduction potentials between the α -FeOOH/Fe 2 + redox couple and the redox couples of AQDS and riboflavin in A and C according to Eq. 18 and as detailed in Materials and Methods . The conditions at which Δ r G x 0 ′ = 0 kJ · mol −1 are marked in A–D by circles ( ° ) labeled “i,ii” for x = 1,2 (red), “ii” for x = 2 (blue), and “i” for x = 1 (orange). Vertical lines in B and D denote the p K a of the reduced riboflavin species. Fig. 1 A also shows the reduction potential of the goethite ( α -FeOOH)/Fe 2 + redox couple with a slope of –0.177 V · pH - 1 , reflecting the transfer of three protons per electron according to Eq. 4 . [4] α -FeOOH + e − + 3H + ⇌ Fe 2 + + 2H 2 O . The resulting complete redox reactions for both the single-electron transfers ( Eqs. 5 and 6 for the first and second electron transferred, respectively) as well as average combined two-electron transfer ( Eq. 7 ) are [5] α -FeOOH + AQDSH 2 2 − + H + ⇌ Fe 2 + + AQDS • 3 − + 2H 2 O [6] α -FeOOH + AQDS • 3 − + 3 H + ⇌ Fe 2 + + AQDS 2 − + 2H 2 O [7] 2 α -FeOOH + AQDSH 2 2 − + 4H + ⇌ 2Fe 2 + + AQDS 2 − + 4H 2 O . We calculated the Gibbs free energies, Δ r G x 0 ′ \n = − n F · Δ E H, x 0 ′ (in kilojoules per mole of transferred electrons), for α -FeOOH reduction by AQDSH 2 2 − ( Eqs. 5 – 7 ) from the differences in reduction potentials at any given pH ( Δ E H, x 0 ′ ) between the α -FeOOH/Fe 2 + redox couple and the semiquinone/hydro- quinone, quinone/semiquinone, and quinone/hydroquinone redox couples of the EES as in Eqs. 8 – 10 . [8] Δ E H, 2 0 ′ = E H 0 ( α -FeOOH ) − 2.303 R T F ( log { Fe aq 2 + } + 3 pH ) − ( E H, 2 0 − 2.303 R T F ( log { AQDSH 2 2 − } { AQDS • 3 − } + 2 pH ) ) [9] Δ E H, 1 0 ′ = E H 0 ( α -FeOOH ) − 2.303 R T F ( log { Fe aq 2 + } + 3 pH ) − ( E H, 1 0 − 2.303 R T F log { AQDS • 3 − } { AQDS 2 − } ) [10] Δ E H, 1 , 2 0 ′ = E H 0 ( α -FeOOH ) − 2.303 R T F ( log { Fe aq 2 + } + 3 pH ) − ( E H, 1 , 2 0 − 2.303 R T 2 F ( log { AQDSH 2 2 − } { AQDS 2 − } + 2 pH ) ) . In these equations, F is the Faraday constant, E H 0 ( α -FeOOH) is 0.768 V ( 37 ), R is the gas constant, T is the absolute temperature, { Fe aq 2 + } is the activity of dissolved Fe 2 + , and E H, x 0 is the standard reduction potential of the redox couples semiquinone/hydroquinone ( x = 2; Eq. 8 ), quinone/semiquinone ( x = 1 Eq. 9 ), and quinone/hydroquinone ( x = 1, 2; Eq. 10 ). Note that E H, x 0 ′ values were calculated using the experimental proton activity and assumed equal activities of the hydroquinone, quinone, and semiquinone species of the EES (hence the superscript 0 ′ ). This assumption was necessary because we could not experimentally quantify concentrations of semiquinones due to their transient nature. Fig. 1 B shows that Δ r G x 0 ′ increases with increasing pH with slopes reflecting the proton stoichiometries in Eqs. 5 – 7 . More importantly, Δ r G 2 0 ′ is less negative than Δ r G 1 0 ′ over the entire pH range shown. Therefore, the transfer of the first electron from the hydroquinone to α -FeOOH is less exergonic (and even endergonic above pH 5.6) than the transfer of the second electron from the semiquinone to α -FeOOH. The difference in free energies of α -FeOOH reduction by the first and the second electron transferred from the EES increases with decreasing pH from 2 kJ · mol −1 at pH 8.0 to as much as 36 kJ · mol −1 at pH 5.0. We therefore expect that rates of α -FeOOH reduction by a fully reduced EES relate to the driving force of the first electron transferred from the ESS to Fe(III) on the basis of the general notion that rates of electron transfer reactions scale with reaction free energies in the normal Marcus region ( 29 ). We conducted the same thermodynamic analysis for the reduction of α -FeOOH by riboflavin ( Fig. 1 C and D , see species in SI Appendix , Fig. S2 ). Calculated E H, x 0 ′ and Δ r G x 0 ′ values again show that the transfer of the first electron from fully reduced RBFH 2 ( x = 2, blue lines) to α -FeOOH is thermodynamically less favorable than that of the second electron from the semiquinone species RBFH • ( x = 1, orange lines) to α -FeOOH. As for AQDS, we stipulate that the less exergonic, and above pH 6.6 even endergonic, first electron transfer from RBFH 2 and RBFH − to α -FeOOH controls the overall rates of electron transfer from reduced riboflavin to α -FeOOH. Relating Rates of Goethite Reduction by Reduced EES to Free Energies. We correlated the initial rates of goethite reduction that we determined experimentally for reactions with AQDSH 2 2 − and RBFH 2 over the pH range 4.50 to 7.25 and 6.25 to 7.25, respectively, at 0.25-pH intervals with the corresponding Δ r G x 0 ′ values calculated as described above. Surface area-normalized initial rates, r SA , of goethite reduction were determined by spectrophotometrically following EES oxidation over time (examples for AQDSH 2 2 − in Fig. 2 A–C and SI Appendix , Figs. S12 A and S13 , and examples for RBFH 2 in SI Appendix , Fig. S12 B ). We first relate r SA values for AQDSH 2 2 − and RBFH 2 to Δ r G 1,2 0 ′ values calculated by using the reduction potential averaged over both electrons transferred, E H, 1 , 2 0 ′ ( Fig. 2 D ). While r SA values for AQDSH 2 2 − and RBFH 2 decreased with increasing (i.e., less negative) Δ r G 1,2 0 ′ ( Fig. 2 D ), we observed disparate trends between the two EES, consistent with the results of Shi et al. ( 25 ) According to our hypothesis, these disparate trends are artificial, as they result from relating rates to free energy measures of both electrons transferred. Indeed, when replotting r SA versus Δ r G 2 0 ′ for the less exergonic of the two electron transfers from the reduced EES to Fe(III), all rate data for AQDSH 2 2 − and RBFH 2 coalesce into one single relationship ( Fig. 2 E ). Consequently, a disparity between the two EES datasets is again observed when plotting r SA values against the free energy of the more exergonic, second electron transfer from the semiquinone species to Fe(III), Δ r G 1 0 ′ ( Fig. 2 F ). Fig. 2. Selected data from goethite reduction experiments with reduced AQDS and riboflavin (RBFH 2 ). ( A and B ) Absorbance spectra of mixtures of oxidized (AQDS 2 − ) and reduced AQDS (AQDSH 2 2 − ) collected during experiments at pH 4.50 and 7.25. ( C ) Changes in Fe(III) concentrations normalized to initial goethite surface area over time in the two experiments. Experiments were designed such that equal electron equivalents of Fe(III) and AQDSH 2 2 − were present at all times. Fe(III) concentrations were therefore determined directly from the changes in the absorption spectra shown in A and B by deconvolution as described in Materials and Methods . Surface area-normalized initial rates of Fe(III) reduction, r SA , for goethite were derived from changes of Fe(III) concentrations according to Eq. 12 ( SI Appendix , Fig. S13 ). ( D – F ) r SA values measured at pH 4.50 to 7.25 for AQDSH 2 2 − and pH 6.25 to 7.25 for RBFH 2 versus Δ r G x 0 ′ , where x refers to the free energy of ( D ) the two-electron transfer from the reduced EES to Fe(III) (x = 1,2), ( E ) the first one-electron transfer from the hydroquinone species of the EES to Fe(III) (x = 2), and ( F ) the second one-electron transfer from the semiquinone species of the EES to Fe(III) (x = 1). Error bars represent deviations from the mean of duplicate measurements. Kendall’s τ B values from rank correlation analysis are reported for the number of data points (n) used in the statistical analysis (–1 = perfect negative correlation, 0 = no correlation). SI Appendix , Fig. S6 A and B shows the same r SA values versus pH. Reconciling Iron Oxide Reduction Rates by EES across Datasets. We demonstrate a broad applicability of the r SA – Δ r G 2 0 ′ relationship by extending our evaluation to experimental systems with hematite and two additional EES, as well as by including literature-reported rates of goethite and hematite reduction by various EES ( 25 , 38 – 40 ). Taken together, the resulting datasets include experiments with three two-electron EES (5-hydroxy-1,4-naphthalenedione [juglone], anthrahydroquinone-2,6-disulfonate, and riboflavin) and four viologen EES. We note that, contrary to the three two-electron EES, viologens exhibit an E H,1 0 above E H,2 0 ( 41 ), thus allowing formation of stable semiquinone species that act as one-electron transfer reductants of Fe(III). Rates of Fe(III) reduction by one-electron reduced viologens are therefore expected to follow the above free energy relationship for two-electron EES, but for viologens on the basis of relating rates to Δ r G 1 0 ′ . Fig. 3 A and B show initial rates of goethite and hematite reduction by the two-electron EES anthrahydroquinone-2,6-disulfonate, riboflavin, and juglone plotted versus Δ r G 2 0 ′ . The figures also include rates of reduction by the four viologens (i.e., cyanomethylviologen, methylviologen, benzylviologen, and diquat) versus the free energy of the one-electron transfer from the viologen semiquinone to Fe(III), Δ r G 1 0 ′ . Rates determined in this work are shown in colored symbols, whereas data from the literature ( 25 , 38 – 40 ) are displayed in gray symbols. Values of r SA for both goethite and hematite by the two-electron EES fell into well-constrained relationships with Δ r G 2 0 ′ with only few exceptions of some literature-reported data. The viologen-based r SA data extended these relationships for the two-electron EES toward higher rates and more exergonic Δ r G 1 0 ′ . The relationships not only hold for experiments which commonly are set up with fully reduced EES and in the absence of Fe(II) but also for systems in which the iron oxides are partially reduced, as typically found in the natural environment: Reduction rates that we measured at varying initial Fe(II) concentrations and thereby modulated thermodynamic driving force also fell into these relationships ( SI Appendix , Figs. S8 and S9 ). By comparison, plotting the reduction rates by the two-electron EES versus Δ r G 1,2 0 ′ resulted in scattered data without a consistent trend for the different EES ( Fig. 3 C and D ). We also observed a pronounced offset of the r SA values toward more negative Δ r G 1,2 0 ′ relative to r SA values obtained for the viologen single-electron transfer reductants. Our finding that r SA values collected over a wide range of experimental conditions and mineral morphologies (i.e., different EES, varying ratios of Fe(III) to EES concentration, iron oxide crystal sizes and shapes, and solution pH) exhibited the same dependence on Δ r G 2 0 ′ confirms that rates of Fe(III) reduction by EES correlate with thermodynamic descriptors of the first, one-electron transfer from the fully reduced EES to Fe(III). Fig. 3. Free energy relationships for the rates of goethite and hematite reduction by EES. ( A and B ) Surface area-normalized initial reduction rates, r SA , for goethite and hematite with reduced one- and two-electron EES versus the free energy of the one-electron transfer from the semiquinone species of one-electron EES ( Δ r G 1 0 ′ ) and first one-electron transfer from reduced two-electron EES ( Δ r G 2 0 ′ ) to the iron oxide. The r SA values from the literature are shown in gray (see SI Appendix , sections S1 and S5 for thermodynamic calculations) ( 25 , 38 – 40 ). The three literature r SA values in A which are smaller than expected based on the free energy relationship were determined from measurements of dissolved Fe(II) at near-neutral pH and therefore may have underestimated the rates of electron transfer ( SI Appendix , section S1 ). ( C and D ) The identical r SA values vs. the free energy of the two-electron transfer from the EES to the iron oxide ( Δ r G 1,2 0 ′ ). Data for one-electron EES are replotted from A and B . SI Appendix , Figs. S6 and S7 shows r SA values versus pH. Blue and red areas serve as visual guides for the quality of the correlation of kinetic and thermodynamic data. Error bars represent deviations from the mean of duplicate measurements. Kendall’s τ B values from rank correlation analysis for the two-electron EES data are reported for the number of data points (n) used in the statistical analysis (–1 = perfect negative correlation, 0 = no correlation). While we present these principles with data for pH 4 to 8, we expect that the established free energy relationship also applies outside this pH range and for other potentially relevant two-electron transfer EES such as pyocyanines ( 17 ). Note that, at pH > 8 for quinones and flavins and at pH < 5 for pyocyanines, semiquinone species are stable because these EES have higher E H, 1 0 than E H, 2 0 ( 34 ). It follows from the above reasoning that, under such conditions, rates of iron oxide reduction are expected to correlate with Δ r G 1 0 instead of Δ r G 2 0 , a behavior that we observed in experiments with viologens. Free Energy Relationships for Fe(III) Oxide Reduction in a Biogeochemical Context. Comparing the kinetics of Fe(III) reduction by EES on a thermodynamic basis allows one to systematically assess the relative importance of this process to anaerobic respiration pathways across systems with different biogeochemical conditions. In fact, the free energy relationships in Fig. 3 A and B integrate over a range of elementary processes and iron oxide thermodynamic properties that all contribute to the observable reduction rates. The potential processes and properties that cause higher reduction rates of hematite than goethite at Δ r G 2 0 ′ \n > − 20 kJ · mol −1 include the following: 1) higher density of reactive surface hydroxyl sites ( 42 , 43 ) on hematite ( 44 ) than goethite ( 42 ) crystal faces at which interfacial electron transfer to Fe(III) occurs, 2) larger reactive surface area of hematite than goethite resulting from differences in particle aggregation ( 45 ), 3) larger number of surface defects and hence more efficient electron transfer to structural Fe(III) in hematite than goethite ( 46 ), 4) smaller band gap and hence more efficient electron transfer inside hematite than goethite crystals ( 47 , 48 ), and 5) variations in the distribution and reactivity toward reduction of different crystal faces in hematite and goethite ( 49 , 50 ). Despite differences in the free energy relationships between hematite and goethite, reduction rates of both iron oxides asymptotically approach a maximum value of r SA \n ≈ 5 · 10 3 μ mol · h −1 \n · m −2 at Δ r G 2 0 ′ \n < − 20 kJ · mol −1 . This rate is approximately three orders of magnitude lower than estimated rates of EES diffusion to the oxide surface (estimated at 10 6 μ mol · h − 1 · m −2 for our reactors; SI Appendix , Fig. S10 ) but falls into the range of reported rates of electron transfer from Fe(II) complexes on oxide surfaces to Fe(III) in the crystal lattice ( 51 , 52 ). This comparison suggests that the electron transfer into the crystal lattice rather than rates of EES diffusion toward the oxide surface governed the maximum reduction rates measured herein. The detachment of Fe(II) from the iron oxide surface, which has previously been considered the rate-limiting step for iron oxide reductive dissolution ( 53 – 55 ), proceeds at rates of ∼ 10 − 1 μ mol · h − 1 · m −2 (51). We measured similarly small r SA values under the least favorable thermodynamic conditions (i.e., Δ r G 2 0 ′ > 10 kJ · mol −1 ). Implications. Our work presents a conceptual framework that allows to relate measured rates of EES-mediated microbial respiration using crystalline iron oxides to the underlying thermodynamics of the electron transfer from the EES to Fe(III)-bearing solids. The framework advances our capabilities to assess the efficacy of this microbial respiration pathway, particularly when applied to (laboratory) model systems that are well characterized with regards to iron-reducing microorganisms, EES, iron oxide mineralogy, and pH. In such systems, measured reduction rates may provide estimates for the reaction free energy and thus the reduction potential to which microbes reduced the EES. If the oxidized and reduced EES species are analytically accessible, the reduction potential of the EES can be computed and compared to the E H of the EES inferred from the experimental reduction rates. Agreement between measured and predicted rates would help to elucidate which fundamental step in Fe(III) reduction controls the overall rate (see above), whereas disagreement would point at either kinetic limitations (e.g., rates of microbial EES reduction or EES diffusion) or enhanced reactivities (e.g., by favorable pH in the microbial biofilm on iron oxide surfaces that deviate from bulk solution pH). Such deviations would therefore be highly informative to aid in the identification and interpretation of potentially limiting and enhancing factors to EES-mediated iron oxide respiration. When extending free energy relationships from goethite and hematite to other iron-bearing minerals (e.g., lepidocrocite, magnetite, six-line ferrihydrite, and clay minerals), relative microbial reduction rates of these minerals in the presence of EES in mixed mineral systems can be interpreted by considering the underlying reaction thermodynamics. Similarly, in systems that contain iron-reducing microorganisms which utilize electron transfer pathways other than EES (e.g., nanowires or direct electron transfer), reaction thermodynamics allow assessing rates of EES-mediated iron oxide reduction and thus the relative competitiveness of this respiration pathway. Finally, in engineered applications, such as microbial fuel cells, the framework developed herein lays the foundation to identify EES that are highly effective to catalyze electron transfer from the microbial cells to the electrode. We anticipate that the identical driving-force dependence also applies for the kinetics of EES reduction at microbial cell surfaces, for example, by membrane-bound multiheme cytrochrome proteins ( 15 , 22 , 23 , 56 , 57 ). In analogy to the above consideration for electron transfers from the EES to Fe(III) oxides, electron transfer rates from such proteins to the EES would be thermodynamically controlled by the one-electron reduction potential of the oxidized quinone species, E H, 1 . In fact, metal-reducing bacteria such as Shewanella sp. sustain different combinations of outer-membrane c-type cytochromes with bound cell-secreted flavins poised to increase the E H of flavin quinone/semiquinone redox couples for one-electron transfer reactions with solid and dissolved electron donors and acceptors ( 15 , 22 , 23 ). From a thermodynamic perspective, the ability of microorganisms to elevate the E H,1 of the quinone/semiquinone redox couple either by binding the EES or by modulation of the pH in their near cell-surface microenvironments might constitute a competitive advantage for effective respiration onto two-electron transfer EES and thus Fe(III) oxides."
} | 5,734 |
34601665 | PMC8818023 | pmc | 6,927 | {
"abstract": "Recurrent neural networks of spiking neurons can exhibit long lasting and even persistent activity. Such networks are often not robust and exhibit spike and firing rate statistics that are inconsistent with experimental observations. In order to overcome this problem most previous models had to assume that recurrent connections are dominated by slower NMDA type excitatory receptors. Usually, the single neurons within these networks are very simple leaky integrate and fire neurons or other low dimensional model neurons. However real neurons are much more complex, and exhibit a plethora of active conductances which are recruited both at the sub and supra threshold regimes. Here we show that by including a small number of additional active conductances we can produce recurrent networks that are both more robust and exhibit firing-rate statistics that are more consistent with experimental results. We show that this holds both for bi-stable recurrent networks, which are thought to underlie working memory and for slowly decaying networks which might underlie the estimation of interval timing. We also show that by including these conductances, such networks can be trained to using a simple learning rule to predict temporal intervals that are an order of magnitude larger than those that can be trained in networks of leaky integrate and fire neurons.",
"introduction": "Introduction Neurons in the Brain exhibit long-lasting activity that outlasts the typical intrinsic time constants of single neurons by orders of magnitude ( Fuster & Alexander, 1971 ; Goldman-Rakic, 1995 ). In some experimental settings, recorded neurons also exhibit long-lasting activity that terminates at intervals with a behavioral significance such as the expected timing of reward ( Huertas et al., 2015 ; Shuler & Bear, 2006 ). Such experimentally observed behaviors can be accounted for by networks of interacting neurons, and reverberations within these networks can account for the long-lasting time constant of neuronal activity. Such patterns of behaviorally relevant neural dynamics can be learned from examples in experimental settings. Various models have been proposed over the years to demonstrate how such recurrent networks can account for long lasting activity ( Compte et al., 2000 )( Renart et al., 2004 ), and for learning temporal intervals ( Gavornik & Shouval, 2011 ; Gavornik et al., 2009 ). Working memory models have often relied on synapses with slow time constants such as NMDA receptors ( Wang, 1999 ). Such slow synapses were assumed because networks with faster, AMPA like synapses typically exhibit very high firing rates in the self-sustaining persistent activity state ( Gavornik & Shouval, 2011 ; Wang, 1999 ), and these activity levels are much higher than those observed experimentally. If the network activity is not self-sustained, but receives external input it can be bi-stable and sustain realistic firing rate statistics in the active state even with fast time constants ( Renart et al., 2006 ). There is some evidence that there is a high concentration of NMDA receptors in prefrontal cortex, where many experimental results of persistent activity have originated ( Wang et al., 2013 ). However, even if there is a high concentration of NMDA receptors, it still needs to be shown that these receptors, and not the faster AMPA receptors are the ones that are modified in order to generate these plastic recurrent networks. Similarly, in networks that learn to predict interval timing, slow synaptic conductances have been used as well ( Gavornik & Shouval, 2011 ; Gavornik et al., 2009 ), in order to avoid unrealistically high firing rates. Additionally, networks with fast, AMPA-type, receptors with realistic variability are hard in practice to train in order to generate temporal intervals that last for more than a few hundred milliseconds. These prior observations and the impact of AMPA-type receptors on network dynamics are explained in more detail below and in figure 1 . Although recurrent networks are the most prominent theory for long-lasting neural activity, an alternative theory with experimental support is that positive activity feedback loops of intrinsic conductances within single cells are able to generate persistent activity ( Egorov et al., 2002 ; Fransén et al., 2006 ), and such mechanisms can also be generalized to neurons that can learn to predict interval timing. The primary experimental support for such active intrinsic conductances, and their contribution to persistent activity arises from Entorhinal slices, although similar channels are shown to exist in other regions including thalamus ( O’Malley et al., 2020 ). Currently, most evidence that intrinsic conductances play a role in persistent activity arises from in vitro studies. In this paper, we set up to show that a recurrent network of neurons with active intrinsic channels( Fransén et al., 2006 ; Tegnér et al., 2002 ), and with fast synapses, is able to generate persistent activity with low firing rates, and to robustly learn temporal intervals that last more than 10 seconds. In a sense this is a hybrid of the two previous approaches, the positive feedback loop observed in single cells is embedded within each neuron of a network model. Single cells within this network are unable to generate sufficient persistent activity alone, but the intrinsic mechanism contributes to long-lasting activity in combination with the recurrent connections. In such a network, the plasticity that generates these ensembles with long-lasting activity is synaptic plasticity rather than plasticity of the intrinsic channels themselves. In this model, the intrinsic activity feedback loop, acts as a conditional slow time constant; this mechanism is typically turned off at rest, but gets activated by sufficient feedforward input or recurrent network activity. With this hybrid model, networks with fast synapses are able to generate persistent activity while exhibiting biologically plausible firing rates. Also, the intrinsic mechanism allows recurrent networks to be trained robustly to predict interval timing over larger temporal intervals, while exhibiting biologically observed firing rates. The active intrinsic conductnaces generate a conditional slow time constant, which is turned on only when the neuronal activity is sufficiently high. This conditional slow time constant allows the network to have a fast on rate for these states together with persistent or very slowly decaying activity. In contrast, in network models with slow synapses, the convergence to the persistent state is also slowed down when synaptic time constants are long.",
"discussion": "5. Discussion Single neurons are highly complex and they possess many intrinsic active conductances that contribute significantly to the function of neural circuits. In contrast, many theoretical circuit models ignore single neuron complexity and use instead highly simplified models of the single neurons. This simplified approach is justified because it helps understand the role of the circuit itself, but it might not faithfully represent the properties of a circuit composed of more complex neurons. Generally, intrinsic properties of single neurons can and do affect circuit dynamics ( Jin et al., 2007 ; Marder et al., 1996 ). In this paper we demonstrate how a specific set of intrinsic conductances can affect the dynamics of bi-stable and slowly decaying networks. Recurrent networks can exhibit bi-stability, in which the network activity can be either in a low or high activity state which lasts indefinitely ( Compte et al., 2000 ). Networks with the same type of structure, but at parameters that are subcritical for bi-stability can exhibit slow transient dynamics( Gavornik et al., 2009 ). For both of these cases slow synaptic dynamics, on the order of 100ms are typically assumed for the networks to quantitively approach physiological measurements of firing rates and possible decay times, and in some systems such long time-constants might be justified ( Wang et al., 2013 ). In this paper, we examined if the addition of specific active conductances to the single neuron model can improve the circuit behavior, in the absence of slow synaptic conductances. We chose a combination of I CAN and voltage gated calcium channels that form a subthreshold positive feedback loop, which acts as a conditional slow intrinsic time constant. We show that by including these channels, we improve significantly the agreement between the network performance and experimental results. With active intrinsic conductances, the bi-stable network achieves bi-stability at much lower firing rates than obtained by a network with fast conductances, and even lower than the networks with NMDA-like slow synaptic time constants. These results are in much better agreement with firing rates observed experimentally ( Fuster and Alexander, 1971 ; Goldman-Rakic, 1995 ). We have also shown that the slowly decaying networks have plateaus at much lower firing rates, consistent with experimental results ( Namboodiri et al., 2015 ; Shuler & Bear, 2006 ). In this subthreshold mode, the network can realistically exhibit decays of up to 16 seconds, much larger than can be accomplished with networks of IAF neurons with fast or even slow synaptic time constants alone. This network can also be trained, with a biophysically plausible learning rule, to decay at short or long intervals over a much larger range than networks with relatively slow synaptic time constants ( Gavornik & Shouval, 2011 ; Gavornik et al., 2009 ). We have also shown that these networks with AIF neurons are robust to network size, degree of sparseness, and randomness in the recurrent connectivity matrix. Moreover, they exhibit biologically plausible spike rasters. The single cell mechanisms assumed here are inspired by previous experimental papers that observed persistent activity in single cells in various brain regions ( Egorov et al., 2002 ; Fransén et al., 2006 ; O’Malley et al., 2020 ; Rahman & Berger, 2011 ) and by the dependance of this persistent activity on non-specific cationic channels and calcium currents, as identified in those papers. This work is also based on previous single cell models of such observations ( Egorov et al., 2002 ; Fransén et al., 2006 ; Shouval & Gavornik, 2011 ). However, other experiments in slices ( Winograd et al., 2008 ) and cultures ( Volman et al., 2007 ) have indicated alternative mechanisms that can lead to slow time constants and to persistent or reverberating synaptic plasticity. It is quite feasible that such alternative mechanisms that generate effective slow time constants in single neurons or single synapses would produce qualitatively similar results to those described here. Indeed, it is quite likely that any mechanism that generates a conditional slow time constant in single neurons or synapses will have a similar effect on circuit dynamics. Such a mechanism is conditional in the sense that the slow time constant are turned on only when cellular activity exceeds a threshold, such that onset dynamics are still rapid, but the decay dynamics are slowed down. revious work ( Tegnér et al., 2002 ) has simulated recurrent networks with using more realistic and complex single cell models, and in that case as well a large NMDA/AMPA ratio is typically required. However, this paper also explored a similar mechanism to the one proposed here, in which I CAN channels were added to the single neurons which also had voltage gated calcium channels. The Tengér et al. (2002) paper has shown that the addition of I CAN channels lowers the minimal NMDA/AMPA ratio that is required for attaining bi-stability. However, this previous publication did not explicitly investigate how such intrinsic active condutances affect the firing rates in the active state, it did examine how it affects the spike statistics of the slowly decaying network, how it extends the range of decay times of a slowly decaying networks by an order of magnitude or how it enables a learning rule based on reward dependent synaptic plasticity ( Gavornik et al., 2009 ) to learn decay times of up to 16 seconds. In order to obtain bi-stability with realistic firing rates in the UP state, simply adding a recurrently connected inhibitory population is not a solution. Adding a population of recurrent inhibitory neurons without changing other parameters will indeed reduce firing rates, but it will also destabilize the UP state. In order to restabilize the UP state excitatory conductances can be increased resulting in an increase in firing rates. Networks that receive external input, even with fast intrinsic time constants can exhibit bi-stability with lower firing rates in the UP state ( Renart et al., 2006 ; Shouval & Gavornik, 2011 ). Such networks are not self-sustained, since attaining bi-stability depends on this external input ( Renart et al., 2004 , 2006 ). When such networks include balanced excitatory and inhibitory conductances they can also attain bi-stability in which spike count variability is high in both the UP and DOWN states, consistent with experimental observations ( Renart et al., 2006 ). This fluctuation driven bi-stability requires fine tuning of the ratio between excitatory and inhibitory weights. In addition, networks that can sustain an UP state with experimentally realistic firing rates, due to an external current still have very steep T vs W curves, similar to those in figure 1c . Therefore, it is not simple to use such a model in combination with synaptic plasticity of excitatory weights, which alone will easily move the network out of the balanced, fluctuation driven state, resulting in high firing rates, and low variability. Moreover, such networks could not be trained to generate long-duration transients that are longer than those that can be learned by a self-sustaining network of LIF neurons with fast conductances. Another use of recurrent networks is to produce integrator-like networks. Such networks have a continuum of fixed points and the activity level at each fixed point is proportional to the integral of an external signal. At the fixed points of such networks, the leak term is exactly equal to the feedback term that results from the recurrent network. The fixed points of such integrator networks are highly sensitive to their parameters, and very small variability in such parameters can result either in a decay or an explosion in network activity. Several approaches to overcome such ultra-sensitivity of been proposed ( Goldman et al., 2003 ; Koulakov et al., 2002 ). Robustness in these models arises from the networks being composed of robust hysteretic sub-networks ( Koulakov et al., 2002 ), or the existence of hysteretic subunits in dendrites ( Goldman et al., 2003 ). Interestingly the hysteretic sub-networks have also been assumed to required NMDAR like receptors, either for their slow dynamics, or because of the voltage dependence of the NMDAR receptors ( Koulakov et al., 2002 ). Similarly, the hysteretic dendritic compartments are also assumed to have slow time constants which are assumed to arise from slow calcium channels or NMDA receptors ( Goldman et al., 2003 ). Models of sensory integration or of decision making also employ recurrent networks. Such models might be multi-stable and the different states represent decisions or sensory processing. In such models, activity in the network depends on a persistent external input, and they do not maintain the firing of the network solely due to feedback in the recurrent network, and therefore do not need to maintain as high a firing rate while persistently active. However, in practice such models also typically assume that excitatory recurrent connections are dominated by slow, NMDAR-like, synaptic transmission ( Wang, 2002 ; Wimmer et al., 2015 ). Although we have added some biologically realistic complexity to our neural model, real neurons in the brain are much more complex, they include active sodium and potassium conductances that are necessary for spiking and a slew of other active conductances, which are differentially expressed in different types of neurons. Neurons also have a complex spatial structure with different types of compartments that also express different channel types. The neuron used here is still very simple, it is a single compartment model with only two additional channels expressed. Action potentials, in the model, are still simply generated by threshold crossing. Obviously, such a simple model is also not a faithful representation of real cortical neurons. We adopt the approach in order to understand what role such channels can play, and demonstrate that with such channels, firing statistics in networks have more realistic properties and the networks are more robust. By using this conservative approach for adding complexity, we can interpret the model and understand the possible role of such channels, at the possible cost of reduced biological realism. The networks used here are simplified in other respects as well, for example they do not include any inhibitory neurons. Although these recurrent networks either with LIF or AIF neurons, are composed of only excitatory neurons the simple addition of an unstructured, randomly connected, population of inhibitory neurons does not qualitatively change the network behavior. An addition of recurrently connected inhibitory neurons, without any other parameter changes, will clearly reduce the firing rates of the network, destabilize bi-stability and eliminate the slow decay. However, an increase of the recurrent excitatory efficacies can reestablish both these behaviors, without significant qualitative differences in firing rates in the UP state, or the shapes of the decay times vs. recurrent weight curves. In contrast, an addition of structured inhibitory connections can have a more profound effect on network dynamics. Structured connections can for example be used to generate competitive networks that can be used for decision making ( Wang, 2002 ; Wimmer et al., 2015 ), or to generate different classes of neuronal dynamics within the network ( Huertas et al., 2015 ). The analysis of such network dynamics is beyond the scope of the current paper."
} | 4,575 |
28066403 | PMC5179597 | pmc | 6,928 | {
"abstract": "Sponges have a significant impact on marine benthic communities, they are of biotechnological interest owing to their production of bioactive natural compounds, and they promise to provide insights into conserved mechanisms of host–microbe interactions in basal metazoans. The natural variability of sponge-microbe associations across species and environments provides a meaningful ecological and evolutionary framework to investigate animal-microbial symbiosis through experimentation in the field and also in aquaria. In addition, next-generation sequencing technologies have shed light on the genomic repertoire of the sponge host and revealed metabolic capacities and symbiotic lifestyle features of their microbiota. However, our understanding of symbiotic mechanisms is still in its infancy. Here, we discuss the potential and limitations of the sponge-microbe symbiosis as emerging models for animal-associated microbiota.",
"conclusion": "Summary and Conclusion Research on sponge-associated microbiota adds valuable insights to our understanding of animal-microbiota symbiosis, mainly because of the natural range of symbiosis within this early-divergent phylum. However, the field needs to move from exploratory to mechanistic projects, where state-of-the-art techniques are applied to meaningful experimental design and the symbiosis is manipulated. Although certain infrastructure is required, aquaculture conditions are described for several species and the regeneration capacity of sponges could serve for keeping clone lines in laboratory. Further research and resources should focus on the physiology and microbiology of cultured sponges over the long term. The new techniques for targeted genome editing appear to be the most promising method for investigating the sponge-microbiota symbiosis through host manipulation. Finally, the most suitable sponge species for experimental models will depend on the specific focus of the study. In terms of host manipulation, Tethya wilhelma and Ephydatia muelleri seem the most advanced model species, together with Amphimedon queenslandica , with a well annotated genome and possibility of aquaria maintenance. Based on adequate performance in aquaculture, Clathria prolifera, Dysidea avara, Halichondria panicea, Ianthella basta, Ircinia spp., or Mycale laxissima are valuable candidates but they still require comprehensive genomic data on the symbiosis. The cumulative genomic information on their symbiotic communities and possibilities for in situ manipulation or aquaculture of other species such as Aplysina aerophoba, R. odorabile , or Xestospongia sp., ( Table 2 ), necessitate further studies to enhance their tractability. Developing more than one model species will produce a more comprehensive view of the mechanisms of symbiosis. The amenability of laboratory sponge models to manipulation would certainly help to identify key players and key functions of the interaction, and their relevance can be further validate by in situ studies in different species and environmental conditions. Thus, researchers take advantage of the insights from an experimentally tractable model in combination with the holistic view of the sponge symbiosis in its natural ecological context.",
"introduction": "Introduction Each model of animal-associated microbiota offers unique opportunities to address questions related to symbiosis ( Ruby, 2008 ; Bosch and McFall-Ngai, 2011 ; Kostic et al., 2013 ). Existing laboratory models, such as fly ( Drosophila melanogaster ), worm ( Caenorhabditis elegans ), zebra fish or mice, enable to test the influence of genetic or environmental factors on symbioses. However, it remains unclear to which extent the findings in lab models apply to natural systems and to other taxa. Natural models provide a relevant ecological and evolutionary framework, but are frequently restricted to systems in which the number of players and interactions is reduced (e.g., the squid Euprymna scolopes and the bacterium Vibrio fischeri , or the mussels Bathymodiolus spp. and their microbiota). However, owing to deep-sequencing technologies, it has become clear that many animals are associated with complex microbial consortia and the implications of such symbioses are just beginning to be unraveled ( McFall-Ngai et al., 2013 ). The study of complex consortia is challenging because multiple interactions take place simultaneously, making it difficult to decipher the specific roles of each symbiont. Additionally, methodologies are limiting and microbes are frequently recalcitrant to cultivation. Enhancing the tractability of the symbiosis within different animal phyla would contribute to our understanding on animal–microbe interactions. Marine sponges (phylum Porifera) represent prominent examples for such complex symbioses. Many sponge species contain diverse microbial consortia within their mesohyl matrix that can reach densities of up to 10 9 microbial cells/cm 3 of sponge ( Hentschel et al., 2006 ). Sponges belong to a phylum that originated ca. 600 million years ago ( Li et al., 1998 ). Their porous body plan contains a highly ramified aquiferous canal system through which seawater is pumped. Specific cells lining the choanocyte chambers (termed “choanocytes”) take up particles, such as bacterioplankton, from the seawater and transfer them into the mesohyl interior where these are digested by phagocytosis. Sponges lack organs, muscles, and a nervous system. In spite of their simple anatomy, recent studies have revealed an unexpected genomic complexity in sponges ( Srivastava et al., 2010 ; Riesgo et al., 2014a ). For example, they express homologs of genes involved in the animal nervous system ( Ludeman et al., 2014 ) and possess central elements of the Toll-like receptor signaling cascade, potentially involved in innate immunity ( Hentschel et al., 2012 ; Riesgo et al., 2014a ). In addition, sponges and their associated microorganisms yield secondary metabolites that are of relevance to biotechnological and medical applications ( Mehbub et al., 2014 ; Indraningrat et al., 2016 ). The sponge-associated microbiota is exceedingly complex with thousands of symbiont lineages reported per sponge individual ( Thomas et al., 2016 ). The most abundant phyla are Proteobacteria, Chloroflexi , and Crenarchaeota , among others. Altogether, more than 40 microbial phyla, including several candidate phyla (e.g., Tectomicrobia, Poribacteria ) were recovered from sponges. In contrast, most animal-associated microbiota belong within 3–5 phyla and diversification is found at species and strain level ( Kostic et al., 2013 ). Some members of the sponge-associated consortia are vertically transmitted to the next generation by larval stages ( De Caralt et al., 2007b ; Schmitt et al., 2008 ), but horizontal acquisition also appears likely ( Taylor et al., 2007 ). Sponge symbiont OTUs have also been recovered from seawater, albeit at very low abundances, and their activity in the free-living state still needs to be explored ( Webster et al., 2010 ). The diversity patterns, metabolic repertoire and genomic features of sponge-associated microbiota have been reviewed in detail elsewhere ( Taylor et al., 2007 ; Hentschel et al., 2012 ; Webster and Thomas, 2016 ). Despite a species-specific composition, similar functions are detected in the microbiomes of distantly related sponge species, suggesting convergent evolution ( Fan et al., 2012 ; Ribes et al., 2012 ; Thomas et al., 2016 ; Horn et al., 2016 ). These functions relate to nutritional interactions (e.g., nitrification, vitamin B synthesis), host–microbe recognition [e.g., eukaryotic-like proteins (ELPs)] and adaptation to host’s internal environment (e.g., CRISPR-Cas defense system). A significant body of information has been accrued from analyzing the natural variability of sponge microbiomes in different host species and environments. Host species appears to be the main factor driving microbial diversity ( Erwin et al., 2012 ; Easson and Thacker, 2014 ), although environmental factors (e.g., intertidal vs. subtidal habitat, Weigel and Erwin, 2016 ; depth, Steinert et al., 2016 ; location, Pita et al., 2013b ) can also cause intraspecific variability. Also, differences in symbiont density within the mesohyl (i.e., high microbial abundance HMA sponges vs. low microbial abundance LMA sponges; Gloeckner et al., 2014 ) have an impact on microbial community composition as well as on the host pumping rate and other metabolic parameters ( Weisz et al., 2008 ; Ribes et al., 2012 ). Monitoring microbiota changes over seasons, bleaching episodes, natural gradients, or upon transplantation showed the plasticity of the symbiosis at scales that are difficult to mimic in laboratory ( Steindler et al., 2007 ; López-Legentil et al., 2010 ; Erwin et al., 2015 ; Morrow et al., 2015 ). As similar past and present environment has been also faced by other benthic invertebrates, the features and mechanisms of sponge-microbe symbioses contribute to a more comprehensive view of marine symbiotic systems. The need for sponge models for symbiosis has been debated within the sponge microbiology community (i.e., at the 1st International Symposium of Sponge Microbiology, Taylor et al., 2011 ). While the benefits of pooling resources, developing standardized protocols and limiting redundancy were clearly acknowledged, the dangers of developing too narrow a view of the natural diversity were also voiced ( Taylor et al., 2011 ; Webster and Taylor, 2012 ). An experimental sponge model would, however, be immensely useful to put the large amount of sequence data into functional context. Even though some advances toward an experimental sponge model have been made, such as the generation of protocols and procedures for sponge aquaculture ( Schippers et al., 2012 ) as well as the silencing of sponge genes for functional studies ( Rivera et al., 2011 ), the overall efforts are still in its infancy. Here we discuss the current status and future directions toward establishing an experimental sponge model for symbiosis."
} | 2,528 |
35520686 | PMC9056393 | pmc | 6,929 | {
"abstract": "This study proposes new optical roughness parameters that can be objectively quantified using image processing techniques, and presents an analysis of how these parameters are correlated with the degree of superhydrophobicity. To this end, photolithography and dry etching processes were used to form regular square pillars with different heights and spacings with a length of tens of micro-meters on silicon wafers. Optical roughness parameters of the specimens were obtained using image processing, and surface wettability was characterized using static contact angle and sliding angle measurements for water droplets of volume V D = 3.5 μl or 12 μl. As a result, seven optical roughness parameters were derived to describe the surface roughness topography in a multi-faceted way. Between the Cassie–Baxter state and the Wenzel state, two distinct wetting states intermediate state I, and intermediate state II were observed. Multiple linear regression of optical roughness parameters and superhydrophobicity demonstrated that in the stable Cassie–Baxter state, the contact angle can be increased or sliding angle decreased more effectively by adjusting the spacing between pillars than by just tuning the solid area fraction. However, in the metastable state where the Cassie–Baxter state can be changed to intermediate state I and vice versa by adjusting V D or surface geometry, reducing the solid area fraction is a priority to ensure a stable Cassie–Baxter state. Horizontal-perspective roughness parameters had a great effect on dynamic wettability in the Cassie–Baxter state. The results confirmed that the proposed optical roughness parameters may be useful for quantitative analysis of the complex effects of roughness on superhydrophobic surfaces.",
"conclusion": "4. Conclusions This study was intended to propose objective optical parameters to quantify surface roughness and to establish the relationship between these parameters and the superhydrophobicity of the surface. For this purpose, microscale pillars were formed on silicon wafers by photolithography followed by etching, and their roughness topographies were evaluated using image processing techniques. Six specimens with various roughness were vapor deposited with FAS-17. Their wetting behaviors were affected by pillar height H , spacing between pillars S and water droplet volume V D . The wetting states of the specimens by water droplets of two volumes were divided into four states: Cassie–Baxter state, intermediate state I, intermediate state II and Wenzel state. Increase in water droplet volume yielded enhanced sliding of the droplet only in the Cassie–Baxter state and intermediate state I. For quantitative evaluation of the surface roughness, the developed image processing program yielded seven optical roughness parameters. The values were very similar to the geometric characteristic values of specimens. Optical roughness parameters A *, n * and ρ * provided additional information to existing roughness factors. Therefore, these optical roughness parameters provide reliable and complementary methods to describe surface topography of silicon wafers that have micro-scale roughness. This is the first step to eliminate the need for subjective assessment of the morphological aspects of superhydrophobic surfaces. The relationship between optical roughness parameters and superhydrophobicity was analyzed by linear regression. In samples with H = 50 μm and V D = 3.5 μl, the wetting state was stable Cassie–Baxter state; increase of SCA and decrease of SA can be achieved more effectively by increasing s * than by reducing f *. However, with H = 10 μm and V D = 3.5 μl, or H = 50 μm and V D = 12 μl, the wetting state was switchable from Cassie–Baxter state to intermediate I, so increase in SCA or decrease in SA can be accomplished best by reducing f * to make the wetting state enter a stable Cassie–Baxter state. In the stable Cassie–Baxter state, horizontal parameters had a strong effect on dynamic wettability. Therefore, to obtain extreme superhydrophobic phenomena, e.g. lotus effect, horizontal parameters should be considered. The validation of these relationships improves over earlier studies of superhydrophobic surface, which focused mainly on the relationship between f and superhydrophobicity. The relationships identified in this study can be used to optimize superhydrophobic surface by controlling an appropriate optical roughness parameter according to the surfaces wetting states and V D . However, as an early attempt to objectively quantify superhydrophobic surface roughness, this study only focuses on the regular micro-scale roughness parameters. Most industrial superhydrophobic surfaces have uncontrolled roughness scale less than several micrometers, or even nanometers, so further study should be conducted to expand objective characterization of irregular nano-scaled roughness and its effect on superhydrophobicity. The proposed optical roughness parameters have potential applications for comprehensive quantitative analysis of the effects of roughness on superhydrophobic surfaces.",
"introduction": "1. Introduction Superhydrophobic surfaces have many uses, including self-cleaning textiles, anti-fogging/anti-icing coatings, and micro-fluidic systems. 1–6 For water droplets, the maximum static contact angle can be increased to 120° ( ref. 1 ) on a flat surface by decreasing the surface energy without roughness, whereas a superhydrophobic surface has a high contact angle ≥150° and also a low sliding angle <10°. Thus, to create superhydrophobicity, surface roughness must be formed and controlled to minimize the contact area and interaction between water and the solid surface. Many studies 1–9 have been conducted to understand the effect of surface roughness on the surface wettability, with the goal of increasing the ability to obtain superhydrophobic surfaces. Theoretical models and experimental studies have determined how surface roughness affects surface wettability. 7,9–13 The most basic models for describing wettability on rough surfaces are the Wenzel model 10 and the Cassie–Baxter model. 11 The Wenzel model 10 is 1 cos( θ W ) = r cos( θ 0 ) where θ W is Wenzel static contact angle, r is a roughness factor (ratio of real area of the solid surface to its projection area) of the solid surface, and θ 0 is the apparent static contact angle on a surface. The Cassie–Baxter model 11 is 2 cos( θ CB ) = f cos( θ 0 ) + f − 1 where θ CB is the Cassie static contact angle, f is the fraction of the projection area of the tops of the solid surface that are in contact with water. The two models consider different behaviors of a water droplet on a surface. The Wenzel model considers a water droplet that seeps in between the irregularities of a rough surface ( Fig. 1A ). The Cassie–Baxter model considers a water droplet that sits on top of the irregularities, with a layer of air trapped between the irregularities beneath it ( Fig. 1B ). Fig. 1 Illustrative two basic wetting models: (A) Wenzel model; (B) Cassie–Baxter model. The Wenzel and Cassie–Baxter models are widely used to describe and predict the static contact angle of rough surfaces. However, some wetting phenomena ( e.g. , transition state, 5 rose-petal effect 8 ) cannot be described by either of the two models. A droplet's attachment to, and rolling off from a surface are mainly associated with the dynamics of the three-phase contact line. 19 The balance of adhesion, shear, gravitational and air drag forces along the contact line governs acceleration of droplet rolling (or the droplet can be pinned on the surface by the balance). 19 Surface energy, roughness topography, water droplet volume and environmental conditions all affect each of these force. In the petal effect, water contact angle is high but the droplets stick to the surface even in the case of reversing. As the structure is in a Cassie impregnating wetting state, the droplets on the surface penetrate through the microstructures but cannot penetrate through the nanostructures, with a strong negative pressure owing to the small volume of sealed air between the nanostructures. 25 However, the Wenzel and Cassie–Baxter models predict water contact angle by using only surface energy (cos( θ 0 )) and simplified roughness factors ( r , f ) and therefore disregarded important factors. Furthermore, they only consider situations in which the droplet completely wets the surface (Wenzel) or touches the tops of surface structure without much penetration (Cassie–Baxter). For these reasons, these models are valid only in some range of wetting states 21 so numerous follow-up studies 20–23 have been conducted to complement these models. Especially, regarding the effect of surface roughness, it was experimentally proven that surface superhydrophobicity is considerably affected not only by the vertical-perspective roughness factors ( r in the Wenzel model, f in the Cassie–Baxter model) but also by the horizontal-perspective roughness factors, including spacing between surface irregularities and the shape of surface irregularities. 7,9,12,13 Rahmawan et al. 12 explained the effect of surface roughness on superhydrophobicity by combining the Wenzel and Cassie–Baxter models for the surface that has dual micro-nanoscale roughness. The superhydrophobicity greatly decreases when the spacing between micro-pillars becomes wider than a certain threshold. 12 Han and Gao 13 evaluated the wettability of the film surfaces having hexagonal ZnO nano-rods, and found that the contact angle depends on ZnO nano-rods' length, density and diameter. 13 To represent these results, the Cassie–Baxter f is modified to 13 3 where A is the projection area of the water droplet on the surface, n is the number of nanorods in the area A, h ′ is the depth to which water seeps between adjacent rods, a is length of one side of the rod top, and h is rod height. 13 According to Yoshimitsu et al. , 7 to increase the shedding of water droplets, adjustment of the shape and scale of the solid–liquid–gas triple phase boundary is better than simply reducing the solid area fraction f . Heavy droplets can easily overcome sliding resistance because of the effect of gravity, so they have low sliding angle, and their weight has a larger effect on sliding angle than on static contact angle. 7 Zheng et al. 9,14,15 mathematically analyzed the wetting transition state between Wenzel state and Cassie–Baxter state as a function of hydraulic pressure P . They proposed a critical hydraulic pressure P c , which is the maximum at which the Cassie–Baxter state can be maintained ( eqn (4) ). 9 P c can be increased by reducing the droplets surface tension ( γ ), the solid fraction ( f ), and the solid surface energy (cos ∅ 0 ), while also adjusting the shape of the pillar tops ( λ ): 4 These studies showed how horizontal-perspective roughness factors such as pillar density and shapes affect superhydrophobicity. However, the results reached conclusions that have limited applications for given surfaces or are partially contradictory. Moreover, surface roughness topography affects superhydrophobicity of surfaces in ways that have not been clarified. 16 Until now, it was only possible to quantitatively obtain vertical-perspective surface roughness parameters such as average deviation R a and standard deviation R q due to the limitations of the roughness evaluation methods. Even though the horizontal-perspective has importance in the superhydrophobic properties on surfaces, there are not universal and easy-to-handle characterization methods of horizontal-perspective roughness. Thus, quite frequently, the visual observation of microscopic images of superhydrophobic surfaces is reduced to expressions such as “the grain sizes varied” 13 or “similar rod density”, 13 which estimate the number, shape, density and spacing of irregularities on surfaces. The reader might be then led to the subjective point of view of the writer. 17 On the other hand, descriptions such as “at least ten measurements were made manually for analysis of the geometry of the nano-rods” 6 can also be found, but ranges are normally limited to the selected area and cannot represent the overall characteristics of the surface. As such, we consider that an effort on quantification of various roughness parameters and their effects on superhydrophobicity should be made. Therefore, the goal of this study was to obtain quantitative roughness parameters that describe surface topography by using simple image-processing techniques, and to examine the effect of each parameter on superhydrophobicity to establish an understanding of how surface roughness affects surface wettability at surface roughness range of several tens of micro-meters. For this purpose, Si wafers with pillars of different heights and spacing on the surface were prepared, and the surface roughness was quantitatively evaluated from binarized grayscale digital images. The reliability and accuracy of the proposed parameters were verified by comparison with actual values. To estimate superhydrophobicity, contact angle and sliding angle were measured, then the relationship between these optical roughness parameters and superhydrophobicity was quantified using multiple linear regression analysis.",
"discussion": "3. Results and discussion 3.1 Superhydrophobicity of the micro-roughness samples The flat sample had SCA = 120° after hydrophobization. This result confirms that a FAS-17 monolayer was deposited on the sample surface. 16 The tilted flat sample showed no shedding effect. \n Fig. 3 and 4 shows the effect of pillar spacing S and water droplet volume V D on the SCA and SA of the hydrophobized specimens with pillars that had height H = 50 μm or 10 μm, respectively. At H = 50 μm ( Fig. 3 ), SCA of 3.5 μl water droplet increased as S increased; the reason is that the contact area between water and solid decreases as S increases. At S = 30 μm, the droplet could not remain on the surface, and fell off immediately (SCA = 180°). These measured SCA were higher than the predictions θ CB of the Cassie–Baxter model. However, at V D = 12 μl, SCA increased in the range 10 ≤ S ≤ 20 μm, but maintained similar values at 20 ≤ S ≤ 30 μm. Droplets of V D = 12 μl on S = 30 μm had SCA < θ CB . At both V D , SA decreased as S increased. Here we found two surface-wetting cases: (1) a stable Cassie–Baxter state in which the water droplet has SCA > θ CB and the water slides off the surface; and (2) an intermediate state I in which SCA < θ CB and the water also slides off the surface. The 12 μl droplet on H50–S30 had SCA < θ CB and was therefore in intermediate state I, because the shedding effect of the droplet was maintained (SA < 10°). Fig. 3 Static contact angle (red lines, left axis) and sliding angle (blue lines, right axis) of samples with pillars with height = 50 μm. Orange line: prediction of Cassie–Baxter model. Fig. 4 Static contact angle (red lines, left axis) and sliding angle (blue lines, right axis) of samples with pillars height = 10 μm. Orange line: prediction of Cassie–Baxter model. When the H was 10 μm ( Fig. 4 ), SCA of 3.5 μl and 12 μl droplets increased as S increased from 10 μm to 20 μm. At S = 30 μm, SCA decreased slightly to < θ CB . However, the effect of S on SA was different for the two V D . For the 3.5 μl droplet, SA decreased as S increased. In contrast, for the 12 μl droplets, SA decreased in as S increased from 10 μm to 20 μm, but the droplet became pinned at S = 30 μm. In the H10–S30 specimen, a 12 μl droplet showed another wetting case: intermediate state II in which SCA < θ CB and the water does not run off the surface. Intermediate state II is distinct from the Wenzel state because the SCA of intermediate state II does not correspond to the static contact angles θ W predicted by Wenzel model ( Table 5 ). Therefore, we observed four wetting states of water droplet on solid surface ( Fig. 5 ): the Cassie–Baxter and Wenzel states, and two distinct intermediate states between them. During intermediate state I, the water starts moving down and entering the valleys between the pillars, 18 but this movement does not affect the ability to shed water. During intermediate state II the water has filled the grooves but does not completely penetrate into the valleys; in this state, shedding of water is impaired. Both intermediate states occur when S is sufficiently wide or the water droplet is sufficiently heavy, or H is sufficiently low. Fig. 5 Side view of contact between the water drop and pillars at four different wetting states. The predicted contact angles by Cassie–Baxter and Wenzel model, measured contact angles and wetting states of the specimens are summarized in Table 5 . The measured contact angles were generally closer to the predictions of the Cassie–Baxter model than of the Wenzel model. However, sometimes the measured values are way off than theoretical values. The large gaps might result from the limitations of the Wenzel and Cassie–Baxter models, which cannot perfectly incorporate the effects of water droplet volume and surface geometry on SCA. Furthermore, some small gaps might be due to fine surface scratches that are inevitably created on the pillar tops during sample fabrication. \n V \n D affected surface wettability. SCA was higher for 3.5 μl droplets than for 12 μl droplets. The inner pressures ( Table 4 ) of the droplets were calculated as 9 5 P = (4/ D ) × γ where D (cm) is the diameter of the water droplet when the droplet is assumed to be perfectly spherical, and γ = 72.8 dyne cm −1 , is the surface tension of water at 20 °C. Calculated P is higher for a 3.5 μl droplet than for a 12 μl droplet, so the adhesion with the solid surface might be further reduced for a 3.5 μl droplet. Inner pressure of water droplets Water volume (μl) 3.5 12 1000 Surface tension (dyne cm −1 ) 72.8 Droplet diameter (mm) 1.9 2.9 12.4 Inner pressure (Pa) 1531.58 1003.45 234.68 Predicted/measured contact angles and wetting states of samples with different micro-roughness a Sample Predicted contact angle Measured contact angle and wetting state \n θ \n W \n \n θ \n CB \n Droplet volume 3.5 μl Droplet volume 12 μl \n θ \n Mea \n Wetting state \n θ \n Mea \n Wetting state H10–S10 180 151 161 Cassie–Baxter 159 Cassie–Baxter H10–S20 136 160 164 Cassie–Baxter 162 Cassie–Baxter H10–S30 128 165 164 Intermediate I 161 Intermediate II H50–S10 NA 151 160 Cassie–Baxter 159 Cassie–Baxter H50–S20 NA 160 165 Cassie–Baxter 162 Cassie–Baxter H50–S30 NA 165 >180 Cassie–Baxter 161 Intermediate I a \n θ \n W : predicted contact angle by Wenzel; 10 θ CB : predicted contact angle by Cassie–Baxter; 11 θ Mea : averaged actual measured contact angle by experiment. \n V \n D also affected dynamic wettability. SA tended to be smaller for the 12 μl droplets than for the 3.5 μl droplets. This was possibly because the weight influenced more strongly on SA for large-volume drops and thus made the water droplet overcome the sliding resistance more easily. 7 However, this tendency with V D was applicable only for the Cassie–Baxter state and intermediate state I. Therefore, within a range of V D , the Cassie–Baxter state can be maintained at a given surface roughness; from the opposite perspective, a suitable range of roughness values can maintain a stable Cassie–Baxter state according to the V D and pressure. In this study, SA was measured while the droplet is at rest on the level surface, so that measurements exclude the effect of outer hydraulic pressure that is exerted when the droplet strikes a surface. \n Fig. 6 shows the photographic images of a 3.5 μl droplet and a 12 μl droplet on the sample surface. We can see that the 12 μl droplet has a distorted appearance compared to the 3.5 μl droplet due to its weight. Fig. 6 Shapes of water droplets (left: 3.5 μl droplet, right: 12 μl droplet). 3.2 Optical roughness parameters of the micro-roughness samples \n Tables 6 and 7 show the original optical microscope images of the samples and binary images obtained using the OTSU thresholding algorithm. When the H was 50 μm ( Table 6 ), the contrast between the pillar tops and bottom was clear in all specimens, although the contrast decreased slightly as S increased. The high pillars can cast shadows by blocking the incident light from the microscope, and can increase the amount of diffused reflection between the pillars ( Fig. 7A ); this effect causes a difference in light intensity collected by the sensor from the pillar tops and bottom. Thus, in the grayscale image, the pixels that represent pillar tops were significantly whiter than those of bottom, so OTSU extracted the tops successfully. The optimal threshold values of each sample image were set automatically; 102 (H50–S10), 107 (H50–S20) and 166 (H50–S30) ( Table 6 ). Original optical microscope images and binary images of samples with pillars of 50 μm height Sample H50–S10 H50–S20 H50–S30 Original images \n \n \n \n \n \n Binary images (OTSU) \n \n \n \n \n \n Original optical microscope images and binary images of samples with pillars of 10 μm height Sample H10–S10 H10–S20 H10–S30 Original images \n \n \n \n \n \n Binary images (OTSU) \n \n \n \n \n \n Binary images (revised) \n \n \n \n \n \n Fig. 7 Simple schematic illustration of reflection of light rays: (A) pillars height H = 50 μm; (B) pillars height H = 10 μm. However, in specimens that had pillars H = 10 μm, the contrast between the pillar tops and bottom was not clear. The distinction between the pillar tops and bottom became difficult as S increased. As shown in Fig. 7B , the H was not sufficient to cast a shadow or to scatter the reflected light between pillars; the result was a weak difference in light intensity reflected from the pillar tops and bottom. As a result, image binarization could not be performed successfully with these samples. Therefore, to extract pillar tops from the images, we converted the white pixels of the bottom except the pillar tops into black after OTSU binarization ( Table 7 ). Further considerations must be taken into account to develop an objective binarization method to extract meaningful features in this case. \n Table 8 shows several geometric characteristics of the specimens and the optical roughness parameters by binary images. Geometric characteristics are calculated as: 6 7 8 where L is the sampling length of the specimen roughness profile, y ( x ) is the profile ordinates of roughness profile, a is the length of one side of the pillar top, s is the spacing between the pillars. Geometric characteristics of specimens and obtained optical roughness parameters by binary images a Sample Geometric characteristics Optical roughness parameters Horizontal Vertical Horizontal \n s \n \n R \n a \n \n f \n \n 11 \n (%) \n λ \n \n 9 \n \n \n s * (μm) \n s * | Δ | \n \n \n | Δ | \n f * (%) \n f * | Δ | \n λ * \n λ * | Δ | \n n * \n ρ * ( n */mm 2 ) \n ρ * (μm 2 ) H10–S10 10 5.0 25 2.5 10.2 (±0.03) 0.2 3.66 1.34 24.1 0.9 2.65 (±0.04) 0.15 20 1851.9 2601.1 H10–S20 20 4.8 11 2.5 20.3 (±0.1) 0.3 2.17 2.63 12.4 1.4 2.61 (±0.07) 0.11 12 1111.1 1334.5 H10–S30 30 4.2 6 2.5 30.7 (±0.2) 0.7 1.13 3.07 6.0 0 2.44 (±0.20) 0.06 6 555.6 649.6 H50–S10 10 25.0 25 2.5 11.2 (±0.08) 1.2 18.25 6.75 24.0 1.0 2.42 (±0.02) 0.08 24 2222.2 2587.7 H50–S20 20 24.0 11 2.5 21.5 (±0.05) 1.5 8.93 15.07 9.9 1.1 2.30 (±0.02) 0.20 12 1111.1 1067.7 H50–S30 30 21.0 6 2.5 33.3 (±0.06) 3.3 5.15 15.85 5.5 0.5 2.47 (±0.01) 0.03 6 555.6 587.3 a \n s : spacing between pillars; R a : arithmetical average roughness; f : area fraction by Cassie–Baxter; 11 λ : pillar characteristic parameter by Zheng et al. ; 9 A *: total top area*; λ *: pillar character*; f *: solid area fraction*; n *: number of pillars*; s *: spacing between pillars*; ρ *: pillar density*; : arithmetical average roughness* (asterisk denotes optical roughness parameters), Δ : uncertainties of optical roughness parameters ( s *, f *, λ *). Versatile explanation of surface topography was possible by using the optical roughness parameters A *, n * or ρ *. The common roughness factors R a , f and λ were also obtainable from the optical roughness parameters f * and λ *. Comparison of optical roughness parameter values with geometric characteristic values of specimens revealed little difference, so the results of image processing were considered reliable and applicable. 3.3 Relationship between superhydrophobicity and the optical roughness parameters The relationship between the optical roughness parameters and superhydrophobicity of specimens was quantified by linear regression. First, the optical roughness parameters A * and n * were adjusted to ad. A * and ad. n * by normalization to the droplets contact area with the solid surface ( Table 9 ). This area was calculated using the actual length l of droplet baseline that touches the solid surface. Adjusted A * (ad. A *) and n * (ad. n *) values reflecting l values of the two water droplet volumes on specimens Sample \n l (mm) Adjusted A * (mm 2 ) Adjusted n * 3.5 μl 12 μl 3.5 μl 12 μl 3.5 μl 12 μl H10–S10 0.84 1.52 0.13 0.44 1026 3360 H10–S20 0.85 1.52 0.07 0.22 631 2016 H10–S30 0.85 1.62 0.03 0.12 315 1145 H50–S10 0.85 1.54 0.14 0.45 1261 4139 H50–S20 0.85 1.53 0.06 0.18 631 2043 H50–S30 0 1.56 0 0.10 0 1062 Linear regressions were performed to determine the relationships between optical parameters and SCA/SA for each H and V D ( Table 10 , ESI Fig. 1a–h † ). At H = 10 μm and V D = 3.5 μl, f * and SCA showed the highest correlation (correlation coefficient >0.95, R 2 > 0.90). For SA, f * and ad. A * showed the highest correlation (correlation coefficient >0.95, R 2 > 0.90). At H = 50 μm and V D = 3.5 μl, the optical roughness parameter s * was the most influential factor on SCA, and s * × ρ * × ad. A * × ad. n * were the most influential factors on SA. The R 2 values of parameters were high enough to prove the validity of the regression models. However, at H = 50 μm pillars and V D = 12 μl, f * and ad. A * had the greatest correlation with SCA ( R 2 > 0.80), and f * × ρ * × × ad. A * × ad. n * had the most significant relationship with SA. At H = 10 μm and V D = 12 μl, none of the optical roughness parameters had a significant correlation with either SCA or SA. Another notable point is that at H = 50 μm and V D = 3.5 μl, the dynamic wettability SA was more affected by horizontal-perspective roughness parameters ( s * × ρ * × ad. A * × ad. n *) than by vertical-perspective roughness parameters Linear regression results of optical roughness parameters and contact/sliding angle of specimens Sample Water droplet volume Static contact angle Sliding angle Correlation coefficient \n R \n 2 \n Correlation coefficient \n R \n 2 \n >0.95 >0.9 >0.8 >0.95 >0.9 >0.8 >0.95 >0.9 >0.8 >0.95 >0.9 >0.8 H10 3.5 μl \n f * \n ρ *, ad. A *, ad. n * \n s * — \n f * \n ρ *, ad. A *, ad. n * \n f *, ad. A * \n ρ *, ad. A *, ad. n * \n s * — \n ρ *, ad. A *, ad. n *, s * — 12 μl — — \n f *, ad. A * — — — — — \n s * — — — H50 3.5 μl \n s * ad. A *, ad. n * \n f *, ρ *, — \n s *, ad. n * ad. A * \n s *, ρ *, ad. A *, ad. n * \n f *, — \n s *, ad. A *, ad. n * \n ρ * \n f *, 12 μl — \n f *, ad. A * \n ρ *, ad. n * — — \n f *, ad. A \n f *, ρ *, ad. A *, ad. n * \n s * — \n f *, ρ *, ad. A *, ad. n * \n s * — These results lead to the following conclusions. First, in the range of the stable Cassie–Baxter state ( H = 50 μm V D = 3.5 μl), the best way to increase SCA or decrease SA is to increase s * rather than f *. Second, at a metastable state that the Cassie–Baxter state and intermediate state I can be switched depending on V D or surface geometry ( H = 10 μm V D = 3.5 μl; H = 50 μm V D = 12 μl), the best way to increase SCA and SA is to tune f *; the result is a stable Cassie–Baxter state. Third, to increase the dynamic wettability in the Cassie–Baxter state, horizontal parameters are more important than vertical ones. Consequently, depending on the wetting state, the influence of each optical roughness parameter on superhydrophobicity changes. Therefore, strategies to control static or dynamic hydrophobicity by coordinating surface roughness must consider the wetting state. The roughness factor λ was the same in all specimens, influence of λ * on superhydrophobicity could not be analyzed in this study. According to previous study, 9 λ depends on the shape and size of the pillar tops. Smaller λ can be driven by more complicated or multi-connected pillar tops, and lead higher superhydrophobicity. 9 The influence of the optical roughness parameter λ * should be explored in further study with various samples of different λ *."
} | 7,225 |
39277779 | PMC11523051 | pmc | 6,930 | {
"abstract": "Abstract A hydrogen (H 2 )-based membrane biofilm reactor (H 2 -MBfR) can reduce electron acceptors nitrate (NO 3 − ), selenate (SeO 4 2− ), selenite (HSeO 3 − ), and sulfate (SO 4 2− ), which are in wastewaters from coal mining and combustion. This work presents a model to describe a H 2 -driven microbial community comprised of hydrogenotrophic and heterotrophic bacteria that respire NO 3 − , SeO 4 2− , HSeO 3 − , and SO 4 2− . The model provides mechanistic insights into the interactions between autotrophic and heterotrophic bacteria in a microbial community that is founded on H 2 -based autotrophy. Simulations were carried out for a range of relevant solids retention times (SRT; 0.1–20 days) and with adequate H 2 -delivery capacity to reduce all electron acceptors. Bacterial activity began at an ∼0.6-day SRT, when hydrogenotrophic denitrifiers began to accumulate. Selenate-reducing and selenite-reducing hydrogenotrophs became established next, at SRTs of ∼1.2 and 2 days, respectively. Full NO 3 − , SeO 4 2− , and HSeO 3 − reductions were complete by an SRT of ∼5 days. SO 4 2− reduction began at an SRT of ∼10 days and was complete by ∼15 days. The desired goal of reducing NO 3 − , SeO 4 2− , and HSeO 3 − , but not SO 4 2− , was achievable within an SRT window of 5–10 days. Autotrophic hydrogenotrophs dominated the active biomass, but nonactive solids were a major portion of the solids, especially for an SRT ≥ 5 days.",
"conclusion": "Conclusion A mechanistic model was used to simulate the performance of a CFSTR whose influent contained NO 3 − , SeO 4 2− , and SO 4 2− in concentrations relevant to coal mine and combustion wastewater. The unique feature of the model is that it simulated a bioreactor in which the only supplied electron donor was H 2 , which means that the electron-acceptor reductions were founded on respirations by H 2 -oxidizing autotrophic bacteria. The model included H 2 -oxidizing autotrophs the respire NO 3 − , SeO 4 2− , HSeO 3 − , or SO 4 2− . It also included heterotrophic bacteria that could respire the same electron acceptor by oxidizing organic products generated by the hydrogenotrophic autotrophs. The model also included bacterial productions of protein and carbohydrate EPS, soluble microbial products, and hydrolytic enzymes that can break down both types of EPS to simple fermentable organics, represented as acetate. Simulations were carried out for SRTs from 0.1 to 20 days for a system in which the H 2 -delivery capacity was not limiting. Bacterial activity began at an ∼0.6-day SRT, when hydrogenotrophic denitrifiers were able to accumulate. Selenate-reducing and selenite-reducing hydrogenotrophs became established at SRTs of ∼1.2 and 2 days, respectively. Full NO 3 − , SeO 4 2− , and HSeO 3 − reductions were complete by an SRT of ∼5 days. SO 4 2− reduction began at an SRT of ∼10 days and was complete by ∼15 days. Thus, the desired goal of full reductions of NO 3 − , SeO 4 2− , and HSeO 3 − , but no SO 4 2− reduction, was possible in an SRT window of 5–10 days when the H 2 supply was not limiting. Due to H 2 being the only input electron donor, autotrophic hydrogenotrophs dominated the active biomass, but nonactive solids (EPS, endogenous decay products, and Se 0 ) became the largest portion of the solids for SRT ≥ 10 days, and they were >46% for an SRT of 5 days.",
"introduction": "Introduction A hydrogen (H 2 )-based membrane biofilm reactor (H 2 -MBfR) can be used to reduce electron acceptors such as nitrate (NO 3 − ), selenate (SeO 4 2− ), selenite (HSeO 3 − ), and sulfate (SO 4 2− ), which are in wastewaters from coal mining and combustion (Zhou et al. 2019 ). The H 2 -MBfR opens up the advantages of using H 2 as the electron donor: on-demand and nearly 100% efficient delivery of H 2 , minimum waste-biomass production, microbial reduction of all bioavailable electron donors, and low cost per electron equivalent. The microbial communities that develop in an H 2 -MBfR commonly have distinct hydrogenotrophic and heterotrophic bacteria populations, despite treating waste-water that does not contain organic carbon (Ontiveros-Valencia et al. 2018 ). Here, we present a model that describes a H 2 -driven microbial community comprised of hydrogenotrophic and heterotrophic bacteria that respire NO 3 − , SeO 4 2− , HSeO 3 − , and SO 4 2− , and we apply it to (1) evaluate the microbial ecology of hydrogenotrophic and heterotrophic bacteria in a H 2 -based wastewater treatment system and (2) describe strategies to control the accumulations of target bacteria. The model describes the impacts of solids retention time (SRT) on the reductions of the four electron acceptors in a reactor in which the oxidation of H 2 by hydrogenotrophic bacteria, which are autotrophs, is the foundation of a microbial community that also contains heterotrophic bacteria able to respire the same electron acceptors. While the model does not directly represent the biofilm of an H 2 -MBfR, it provides mechanistic insight into the interactions between autotrophic and heterotrophic bacteria as their specific growth rate is systematically decreased. In a biofilm process, the average specific growth rate of the bacteria is equal to the first-order detachment rate (Rittmann and McCarty 2020 ), which can be manipulated by changes to shear stress or turbulence induced by the normal liquid flow regime (Rittmann 1982 , Laspidou and Rittmann 2004a , b , Wanner et al. 2006 ) or by intentional biofilm-management strategies to periodically remove some of the biofilm. All bacteria carry out oxidation–reduction reactions to obtain energy for synthesis and cell maintenance, and they undergo endogenous processes (Rittmann and McCarty 2020 ). Figure 1 illustrates the simulated metabolic processes carried out by all of the bacteria in the model: electron-donor oxidation, electron-acceptor respiration, synthesis of biomass and extracellular polymeric substances (EPS), and endogenous decay. Figure 2 is an electron-flow diagram for selenite-reducing hydrogenotrophic bacteria (denoted X H2 ). Figures SI.1–SI.8 are the analogous electron-flow diagrams for all of the other bacteria in the model. Figure 1. Simulated metabolic processes of bacteria: respiration, synthesis, and endogenous decay. Oxidized nitrogen, selenium, and sulfur may be reduced as respiratory electron acceptors by one or more of the bacteria. Bacteria synthesize nitrogen as ammonium (NH 4 + ), selenium as hydrogen selenide (HSe − ), and sulfur as hydrogen sulfide (H 2 S), creating an intracellular pool of synthesizable nutrients. EPS and hydrolytic enzymes exist outside the cell wall, which is where most elemental selenium (Se 0 ) accumulates. Heterotrophic bacteria utilize soluble endogenous decay products or the fermentation product acetate as an electron donor, while hydrogenotrophs utilize H 2 as an electron donor. Figure 2. Electron-flow diagram for selenite-reducing heterotrophic bacteria (denoted X SeO3 ). Analogous electron-flow diagrams for other bacteria in the multispecies model are presented in Figures SI.1–SI.8 . Hydrolysis is an enzymatic process that is required to initiate the biodegradation of complex particulate organics. Hydrolytic enzymes are produced by heterotrophic bacteria, and we model the transformation of biodegradable particulate endogenous decay products (denoted X EDP ) into biodegradable soluble endogenous decay products (denoted S EDP ) by hydrolytic enzymes that are bound to biomass. An important aspect of a microbial community that is founded on an inorganic electron donor and autotrophy is that organic carbon (C) and nutrient-available forms of nitrogen (N), sulfur (S), and selenium (Se) become available through endogenous decay of the autotrophic biomass (Okabe et al. 2005 , Zevin et al. 2016 , Ontiveros-Valencia et al. 2018 ). In particular, the organic carbon released from autotrophic biomass decay can be used by nitrate-, selenate-, and selenite-reducing heterotrophic bacteria. We model endogenous decay that transforms active bacteria into biodegradable particulate endogenous decay products and nonbiodegradable particulate endogenous decay products (denoted X EDP, U ). Biodegradable and nonbiodegradable particulate endogenous decay products are assigned the chemical formula C 5 H 7 O 2 N, which is commonly used to describe bacterial biomass (Rittmann and McCarty 2020 ). Soluble endogenous decay products can be fermented, a process in which complex dissolved organics are transformed to simple dissolved organics (e.g. acetate). Nitrate-reducing heterotrophic bacteria and, we assume, selenium-reducing heterotrophic bacteria can utilize soluble endogenous decay products (i.e. C 10 H 19 O 3 N) and acetate (CH 3 COO − ) as electron donors. We model sulfate-reducing heterotrophic bacteria as able to utilize only acetate as an electron donor (Noguera et al. 1998 ).",
"discussion": "Results and discussion The model was run for the influent concentrations relevant for the treatment of flue-gas-desulfurization wastewater from a coal-fired power plant (Boltz and Rittmann 2021 ): NO 3 − = 50 g N/m 3 , SeO 4 2− = 1 g Se/m 3 , SeO 3 2− = 0 mg Se/l, SO 4 2− = 500 g S/m 3 , elemental Se 0 = 0 g/m 3 , and COD = 0 g/m 3 . Three things are important to note about these influent concentrations. First, the input concentrations of SeO 4 2− and HSeO 3 − are much smaller than the input concentrations of NO 3 − and SO 4 2− . This means that selenate- and selenite-reducing bacteria can be only a very small fraction of the total biomass and can oxidize only a small fraction of the electron donors. Second, the SO 4 2− concentration is substantial and will lead to a large production of sulfide if SO 4 2− reduction is not suppressed by lack of electron-donor supply, a low SRT, or a combination. Third, influent organic matter (i.e. COD) is absent, which means that the growth and accumulation of heterotrophic bacteria can occur only through the generation of organic material produced by the H 2 -oxidizing autotrophs. Model simulations were carried out for an SRT range of 0.1–20 days. Based on the kinetic parameters in Table 1 , this range should span from washout of all bacteria at the lowest SRTs (i.e., a value greater than ∼0.6 day) to accumulation of all bacteria at the higher SRTs (i.e. greater than ∼7 day). The H 2 delivery rate is represented as a maximum H 2 concentration in the reactor’s liquid phase. The H 2 -delivery capacity in an H 2 -MBfR is controlled by the H 2 pressure in the membranes’ lumen, and it is given by the H 2 flux, or J H2 in g COD/m 2 -day (Zhou et al. 2019 ). For the model outputs, actual H 2 -delivery flux is related to the reported H 2 concentration by Equation ( 30 ). \n (30) \n \\documentclass[12pt]{minimal}\n\\usepackage{amsmath}\n\\usepackage{wasysym}\n\\usepackage{amsfonts}\n\\usepackage{amssymb}\n\\usepackage{amsbsy}\n\\usepackage{upgreek}\n\\usepackage{mathrsfs}\n\\setlength{\\oddsidemargin}{-69pt}\n\\begin{document}\n\\begin{eqnarray*}\n{{\\mathrm{J}}_{{\\mathrm{H}}2}} = \\frac{{{\\mathrm{Q\\,\\,}} \\cdot \\Delta {{\\mathrm{H}}_2}}}{{{{\\mathrm{A}}_{\\mathrm{F}}}}}.\n\\end{eqnarray*}\\end{document} \n Here, A F is the biofilm surface area in the H 2 -MBfR (m 2 ), Q is the volumetric flow rate (m 3 /day), and ∆H 2 (g COD/m 3 ) is the difference between the maximum (1,220 mg COD/m 3 ) and reported H 2 concentrations (mg COD/m 3 ). The maximum H 2 -delivery capacity used in the modeling simulations was sufficient for complete reductions of NO 3 − , SeO 4 2− , and SO 4 2− ; thus, the available control parameter was the SRT. Figure 3 shows that denitrification began at an SRT of ∼0.6 day and rapidly increased to have almost 100% removal by SRT ∼2 day. This corresponds to the H 2 -oxidizing denitrifiers, which are the autotrophic foundation of the process, having 1.8/d µ max , which corresponds to a minimum SRT of ∼0.6 day. The next bacteria to accumulate were the selenate-reducers, and Fig. 4 shows that they began to appear at an SRT of ∼1.2 day and had full SeO 4 2− reduction by an SRT of ∼2 day. Next to appear were the HSeO 3 − reducers (Fig. 4 ), which initiated HSeO 3 − reduction at an SRT ∼2 days and had full HSeO 3 − reduction by an SRT of ∼5 days. The onset of HSeO 3 − reduction depended on having substantial SeO 4 2− reduction to HSeO 3 − , which was in place with an SRT of ∼2 days. Figure 3. Nitrogen concentrations as a function of solids residence time. Figure 4. Selenium concentrations as a function of solids residence time. SO 4 2− reduction required a much longer SRT. Figure 5 shows that the onset of SO 4 2− reduction was at an SRT of ∼10 days, or about the minimum SRT for H 2 -oxidizing SO 4 2− reducers, i.e. 3 × 3.3 days ∼10 days. Full SO 4 2− reduction was in place by an SRT of ∼15 days. The results in Figs 3 – 5 mean that the desired reductions of NO 3 − , SeO 4 2− , and HSeO 3 − could be achieved with an SRT > ∼5 days, but suppression of SO 4 2− reduction meant keeping the SRT < ∼10 days. Figure 5. Sulfur concentrations as a function of solids residence time. Figure 6 documents that the consumption of H 2 paralleled the onsets of electron acceptor reductions. The H 2 concentration began to decline at an SRT ∼0.6 day, the onset of denitrification, and its major decline was in place at ∼1 day, when denitrification was more than 95% complete (Fig. 3 ). The onsets and build ups of SeO 4 2− and HSeO 3 − reductions (from 1.2 to 5 days, Fig. 4 ) are not perceptible in Fig. 6 because of the small influent concentration of SeO 4 2− . However, the steady demand for H 2 for SO 4 2− reduction is obvious from SRT ∼10 days to ∼15 days. Figure 6. H 2 concentration as a function of solids residence time. The difference between the maximum concentration (1220 mg COD/l) and the plotted value is a measure of the actual H 2 -delivery flux: J H2 = Q(∆H 2 )/A F , in which A F is the biofilm surface area in the H 2 -MBfR (m 2 ), Q is the volumetric flow rate (m 3 /day), ∆H 2 is the concentration difference (g COD/m 3 ), and J is the H 2 -delivery flux (g COD/m 2 -day). Figure 7 shows that organic products accumulated and became more important with longer SRT. By far the dominant organic component was protein-based BAP, which has slow biodegradation kinetics. However, its peak concentration was only 20 g COD/m 3 (at a 15-day SRT), a concentration only ∼1.6% of the COD of the delivered H 2 . Due to their faster biodegradation kinetics, all other organic components had very low concentrations, < ∼2 g COD/m 3 . Figure 7. COD concentrations as a function of solids residence time. Figure 8 illustrates how increasing SRT changed the distribution of biomass types in the microbial community. For the lowest SRTs (0.75 and 1.0 day), the only active bacteria were the denitrifying hydrogenotrophs. Also present were EPS and endogenous-decay products generated from the denitrifying hydrogenotrophs. Other active bacteria begin to appear at an SRT of 1.25 days. Selenate-reducing hydrogenotrophs were present, but cannot be seen due to their small biomass. While selenite-reducing hydrogenotrophs cannot be visualized, they are present for SRT ≥ 2 days, and elemental Se 0 becomes evident for SRT ≥ 5 day. A large shift takes place for SRT ≥ 15 days, when the sulfate-reducing hydrogenotrophs become a the large component of active biomass due to the onset of complete SO 4 2− reduction. Figure 8. Changes in the biomass distribution with increasing SRT. Figure 8 present two other important trends. The first is that the nonactive fractions (EPS, endogenous-decay products, and Se 0 ) became more important for longer SRT. By 15 days, they comprised ∼56% of the total solids mass, and they were about 46% at 5 days. The advent of SO 4 2− reduction increased the active fraction, but the nonactive solids still were ∼52% for an SRT of 20 days. Second, the hydrogenotrophic bacteria completely dominated the active biomass. The only type of heterotrophic biomass that can be visualized in Fig. 8 is the denitrifying heterotrophs, and they are at most 4% of the solids (at SRT = 10 days). Thus, the H 2 -driven biomass was dominated by H 2 -oxidizing autotrophs, although some heterotrophs were present due to their ability to oxidize organic components generated by the autotrophs. This paper describes two means to manage the extent to which nitrate-, selenate-, selenite-, and sulfate-reducing hydrogenotrophic and heterotrophic bacteria accumulate in a H 2 -based MBfR: SRT and H 2 -feed control. In practice, SRT control readily applied to suspended-growth processes, which use controlled solids wasting to establish the SRT (Rittmann and McCarty 2020 ). The SRT can be applied directly for evaluating the ecology of hydrogenotrophic and heterotrophic bacteria in a H 2 -based suspended-growth process. For the H 2 -based MBfR, in which the bacteria accumulate in biofilms, the average SRT is the reciprocal of the specific detachment rate, or b det (d −1 ) (Rittmann and McCarty 2020 ). Because H 2 is delivered directly the biofilm through the gas-transfer membranes in an H 2 -MBfR, the H 2 -delivery capacity, which is controlled by the H 2 pressure in the membrane lumen, is the more direct metric for ecological control (Ontiveros-Valencia et al. 2018 )."
} | 4,328 |
37447571 | PMC10346438 | pmc | 6,931 | {
"abstract": "Over the past few decades, the enhancement of polymer thermal conductivity has attracted considerable attention in the scientific community due to its potential for the development of new thermal interface materials (TIM) for both electronic and electrical devices. The mechanical elongation of polymers may be considered as an appropriate tool for the improvement of heat transport through polymers without the necessary addition of nanofillers. Polyimides (PIs) in particular have some of the best thermal, dielectric, and mechanical properties, as well as radiation and chemical resistance. They can therefore be used as polymer binders in TIM without compromising their dielectric properties. In the present study, the effects of uniaxial deformation on the thermal conductivity of thermoplastic PIs were examined for the first time using atomistic computer simulations. We believe that this approach will be important for the development of thermal interface materials based on thermoplastic PIs with improved thermal conductivity properties. Current research has focused on the analysis of three thermoplastic PIs: two semicrystalline, namely BPDA-P3 and R-BAPB; and one amorphous, ULTEM TM . To evaluate the impact of uniaxial deformation on the thermal conductivity, samples of these PIs were deformed up to 200% at a temperature of 600 K, slightly above the melting temperatures of BPDA-P3 and R-BAPB. The thermal conductivity coefficients of these PIs increased in the glassy state and above the glass transition point. Notably, some improvement in the thermal conductivity of the amorphous polyimide ULTEM TM was achieved. Our study demonstrates that the thermal conductivity coefficient is anisotropic in different directions with respect to the deformation axis and shows a significant increase in both semicrystalline and amorphous PIs in the direction parallel to the deformation. Both types of structural ordering (self-ordering of semicrystalline PI and mechanical elongation) led to the same significant increase in thermal conductivity coefficient.",
"conclusion": "4. Conclusions The development of new thermal interface materials that combine necessary dielectric properties with improved conductive properties is an important industrial task. To provide an overview of this problem, microsecond-scale computer simulations were performed to investigate the thermal conductivity of TIMs based on three thermoplastic PIs. The influence of mechanical deformation performed in melting state on the thermal conductivity coefficient κ of three thermostable PI including two semicrystalline BPDA-P3 and R-BAPB as well as amorphous ULTEM TM was considered at different temperatures. To study the thermal conductivity properties of the thermoplastic polyimides, the choice of the partial charge calculation method for their model based on the GAFF force field as well as the choice of thermal conductivity coefficient calculation method were performed. The best agreement with thermophysical properties of considered PIs was obtained for the HF/6-31G* (RESP) method using EMD. The influence of mechanical deformation of polymer samples on the thermal conductivity coefficient of semicrystalline and amorphous polyimides was studied. The samples stretched up to 200% of their initial size, demonstrating the initiation of nematic ordering of the polymer chains along the deformation direction. Semicrystalline PIs were ordered upon deformation, which was significantly more substantial than that of the amorphous PIs. The enhancement in the thermal conductivity of the ordered PI samples accounts for the appearance of structural ordering in one direction in the polyimide chains. The anisotropy of the heat transfer properties was found: the thermal conductivity coefficient along the deformation was essentially higher than that computed in the lateral direction. The impact of two different structural orderings on the thermal conductivity coefficient of PI chains was studied. With regard to the semicrystalline PI BPDA-P3, the thermal conductive properties of the samples ordered over tens of microseconds by computer simulations were compared with the thermal conductive properties of the samples ordered by mechanical elongation. When cooling started from T = 800 K, a higher enhancement in the thermal conductive properties of the samples ordered by mechanical elongation was observed compared to the properties of the unordered samples. However, when cooling started at 600 K, the results reveal that both structural orderings caused a similar enhancement in thermal conductivity in the glassy state. For this case, the EMD method showed an almost 40% increase in the thermal conductivity coefficient of semicrystalline PI. This might be useful for the creation of the BPDA-P3 samples by applying mechanical elongation, which would reduce the computational time for the creation of BPDA-P3 self-ordered 30 µs long computer simulation samples.",
"introduction": "1. Introduction The development of new thermal interface materials (TIM) with improved thermal conductivity properties is important in order to address various issues in modern electronic and electrical industries. These issues primarily relate to the need to minimize the size of and lighten electronic components [ 1 , 2 ], including computers, consumer devices, telecommunication infrastructure, LED lighting products, renewable energy, automotive engineering, various types of industrial and medical equipment, wireless systems, and solutions for 5G technology [ 3 ]. These materials allow for the removal of heat from the heat source, extend the lifecycle of devices, and reduce the cost of maintenance and emergency repair [ 3 , 4 ]. Recently, the use of polymers has become critical in the manufacture of new devices due to their lower specific mass fraction, convenient recycling and usage capabilities, and good dielectric characteristics [ 1 , 2 ]. The thermal conductivity coefficient ( κ ) of polymers, however, is still rather low, lying in the range from ~0.1 to ~0.5 W/m·K [ 5 , 6 ]. In particular, as noted above, the ability not only to remove heat from heat sources but also to preserve excellent dielectric properties is extremely desirable in electronic and electrical industries when producing new thermal interface materials. Among conjugated polymers [ 7 ], thermoplastic polyimides (PIs) [ 8 , 9 , 10 ] can be used for polymer binding, since they are heterocyclic polymers with one of the best thermal and dielectric properties and are characterized by good thermal stability, mechanical strength, and chemical resistance. These polymers are used as films, fibers, nanofibers, membranes, foams, adhesives, and coatings in various industries, including electronics, aerospace, automotives, and medicine. PIs are often used as heat-control coatings, as well as protective layers for electronic equipment [ 11 ] due to their low coefficient of thermal expansion. The total global market for polyimides in 2022 was approximately USD 2.31 billion, with the compound annual growth rate (CAGR) predicted to be 7.8% between 2023 and 2030 [ 12 ]. The thermal conductivity of PIs is quite low, however, in comparison with that of other polymers [ 13 , 14 ], a fact which significantly hinders the use of these compounds at the heat–dissipation interface in microelectronic devices. Improving the thermal conductivity of PIs is therefore of great importance. One possible way to improve the thermal conductivity coefficient of polymers [ 15 , 16 ] is the addition of various organic [ 17 , 18 , 19 ] or inorganic [ 20 , 21 ] nanoparticles with a high thermal conductivity coefficient. However, the desired thermal conductivity is often paired with a growth in electrical conductivity, which ultimately degrades the dielectric properties [ 22 ]. For example, the addition of graphene derivatives can improve the thermal conductivity of materials [ 17 , 23 ]; however, since graphene has a high electrical conductivity, this could impair the excellent dielectric characteristics of PIs [ 22 , 24 ]. Metal-based nanoparticles (Al 2 O 3 , Fe 2 O 3 , etc.) [ 25 , 26 , 27 ] are also widely used to improve the thermal conductivity properties of polymers; nevertheless, similar to graphene derivatives, the incorporation of metal nanofillers into PI also causes a significant deterioration of their dielectric properties. Boron nitride particles and their derivatives [ 20 , 28 , 29 , 30 , 31 , 32 ] can act in some instances as a substitute for graphene particles to improve the thermal conductivity properties without losing the dielectric properties of PIs. However, the addition of nanofillers to multicomponent systems has many technological problems related to the dispersion of nanoparticles, heat transfer resistance at the nanofiller–polymer interface, and decrease in the fragility of materials, all of which affect the mechanical properties of polymer nanocomposite materials [ 33 ]. Therefore, in some instances, it is necessary to change the heat transfer properties of polymers without the addition of nanoparticles [ 34 , 35 ]. On the one hand, an alternative to adding nanoparticles to polymer binders is to achieve structural order in polymer sample. With respect to semicrystalline polymers, an enhancement in the thermal conductivity of these substances can be achieved by isothermal (structural ordering) [ 36 , 37 ] or nonisothermal crystallization [ 38 , 39 , 40 ]. The thermal conductivity of the ordered polymer chains may depend on the crystallinity, crystallite orientation, and size, as well as the orientation of the polymer chain in the amorphous region [ 33 ]. The structural change in the polymer chains induced by the transition from a disordered amorphous state to a partially or fully ordered state could cause an increase in the phonon-free path, thereby reducing the number of phonon-scattering centers [ 41 ]. The improvement in the thermal conductivity coefficient of semicrystalline polyethylene (PE) was found to depend on the crystallinity degree of the samples [ 42 ]. Ruan et al. [ 43 ] studied the thermal conductivity properties of liquid crystal PI films and found that the orientation of polymer chains reduces phonon scattering between polymer chains, which improves the intrinsic thermal conductivity. Kurabayashi et al. [ 44 ] investigated the anisotropy of the PI films and established that enhancement in thermal conductivity was in the lateral direction. On the other hand, stretching the polymer sample might lead [ 34 , 45 , 46 ] to an improvement not only in thermal conductivity, but also in other properties [ 22 , 47 , 48 ] of both polymers and nanocomposites on their bases. Using different experimental techniques, Yoon et al. [ 34 ] studied the influence of orientation of amorphous BAPP-ODPA PI on the heat transfer properties. They found that oriented samples, even of amorphous PIs, significantly enhanced the thermal conductivity coefficient. The authors suggested that this improvement could be attributed to the orientation of the molecular chain and the appearance of π–π interactions between the aromatic fragments of the PI chains. Lin et al. [ 48 ] investigated the influence of draw ratio on the structural and mechanical properties of amorphous PIs and showed that the uniaxial deformation of amorphous PIs leads to a change in their thermophysical properties. Xiang et al. [ 49 ] studied amorphous/low-crystallized PI composite fibers and found an increase in the κ value in these systems, owing to the orientation of polymer chain and the formation of interchain hydrogen bonds from the wet spinning or low-ratio thermal drawing process. A significant increase in the thermal conductivity of the drawn PE nanofibers was observed after mechanical deformation was applied [ 50 ]. Furthermore, for crystallizable polymers, the additional deformation of ordered samples might further enhance their thermal conductivity [ 51 ]. He et al. [ 52 ] studied anisotropic thermal transport in crystalline PE. They found that the thermal conductivity increased in the axial direction with an increase in strain, while the thermal conductivity decreased in the radial direction upon deformation. Muthaiah et al. [ 53 ] investigated the influence of the strain amplitude on the thermal conductivity coefficient of amorphous PE at different temperatures to understand the behavior of the value of κ upon the deformation up to 400% of the strain. Simavilla et al. [ 54 ] studied the strain dependence of thermal conductivity for PE and polystyrene (PS) entanglement melts. The strong anisotropy of κ values agreed well with the experimental values. Donovan et al. [ 55 ] investigated the influence of the additional off-axis strain on the thermal conductivity of polypropylene films using frequency-domain thermoreflectance and molecular dynamics simulations. A significant improvement in the value of κ along the deformation direction could be regulated by deformation in the orthogonal direction. Ito et al. [ 56 ] found an increase in the thermal conductivity properties of one polymer chain upon strain application. Generally, mechanical stretching is a useful technique for increasing the thermal conductivity of polymers for thermal interface materials, although the degree of improvement could be limited by the intrinsic thermal conductivity of the polymer and may be characterized by some anisotropy relative to the deformation [ 57 , 58 ]. Previously, computer simulations have made it possible to investigate how even small changes in the chemical structure of polymers can influence on the performance properties of thermoplastic PIs [ 59 , 60 ]. However, there is a lack of simulation studies where the influence of mechanical deformation on the thermal conductivity coefficient was performed for both amorphous and semicrystalline PIs, despite the fact that these polymers are very useful for the production of thermal interface materials. In this study, we investigated the effects of uniaxial stretching on semicrystalline and amorphous thermoplastic polymers. Two semicrystalline polyimides, BPDA-P3 and R-BAPB as well as the amorphous polyimide ULTEM TM were considered. To analyze the effect of uniaxial deformation on the thermal conductivity coefficient of semicrystalline and amorphous PIs, we determined the values of the thermal conductivity coefficient along and perpendicular to the deformation direction. The results were compared with the thermal conductivity properties of unoriented samples. Additionally, with respect to PI BPDA-P3, a comprehensive comparison of the influence of the type of structural ordering on the thermal conductivity was performed. The structural ordering of semicrystalline BPDA-P3, which appeared during uniaxial deformation, was compared with the properties of that PI self-ordered during 30-µs-long molecular dynamics simulations, which corresponded to an isothermal crystallization process. A comparative study of the thermal conductivity of BPDA-P3 polyimide ordered in different ways is extremely important from a computer simulation point of view, as it allows us to evaluate the difference between the thermal conductivity coefficients of two differently ordered samples. Thus, a short simulation of the mechanical stretching of a polymer sample can significantly reduce the simulation time required for the complete self-ordering of the polymer chains of a semicrystalline polyimide if the resulting thermal conductivity is the same as during a long simulation of self-ordering.",
"discussion": "3. Results and Discussion From the outset, it is crucial to determine which predetermined methods of partial charge calculation, namely AM1-BCC or HF/6-31G* (RESP), are best suited to parameterize the electrostatic interactions of the PI within the GAFF force field. The thermal conductivity coefficients of PIs were analyzed using the EMD and NEMD calculation methods. 3.1. Validation of GAFF Force Field for Computer Simulation of Polyimides Unfortunately, the literature does not provide any available data on the experimental determination of thermal conductivity coefficient κ for the PIs BPDA-P3 and R-BAPB. An analysis was conducted to compare the outcomes of a computer simulation with the empirical thermal conductivity coefficient of PI ULTEM TM . The κ value of the polyimide ULTEM TM provided by the Sabic Innovative Plastics website [ 64 ] was selected for comparison with the simulation results. The value of κ is 0.220 W/(m·K) at room temperature. To make a quantitative comparison between the experimental and simulation outcomes of κ values, the relative percentage deviation was computed as ( κ s i m − κ e x p ) / κ e x p × 100 % of calculated thermal conductivity values κ s i m from the experimental thermal conductivity value κ e x p . The results are presented in Table 1 . Based on a comparative analysis of the thermal conductivity coefficient of PI ULTEM TM at room temperature, the following conclusions can be drawn. All computational techniques produced a relative percentage deviation from the experimental value of κ , which exceeded 20%. Overall, the NEMD method, which was worse than EMD, predicted the value of κ exp . The EMD method yielded the most accurate thermal conductivity results when partial charges were assessed using the HF/6-31G* (RESP) method, with a relative percentage deviation of approximately 20.5% of the calculated κ s i m value from the experimental κ e x p value. The EMD method showed better agreement with the experimental thermal conductivity coefficient of PI ULTEM TM , which agrees well with the outcome of the study of the thermal conductivity properties of phase-change materials based on paraffin n-eicosane [ 72 ]. When using EMD, the AM1-BCC approach exhibited worse results by approximately 9% in terms of the relative percentage deviation compared with HF/6-31G* (RESP). The opposite trend was found for NEMD: the AM1-BCC partial charge calculation method showed an ~18% better reproduction of the experimental value of the ULTEM TM thermal conductivity coefficient than HF/6-31G*(RESP). However, the relative percentage deviation from the experiment for NEMD was much higher than that for EMD. Nevertheless, the disparity in the relative percentage deviation between AM1-BCC and HF/6-31G* (RESP) using EMD is rather low. The increase in the thermal conductivity coefficient in the simulation for both the calculation methods EMD and NEMD, and for the two partial charge calculation methods compared to the experimental value of κ , is qualitatively consistent with a small overestimation of the thermal conductivity coefficient, as previously shown for all-atom force fields [ 72 ]. This improvement in the thermal conductivity coefficient could be caused by the presence of additional vibrational degrees of freedom for all-atom-based models capable of enhancing phonon transport. As a result, we conducted further research to determine how these partial charge calculation methods can accurately replicate some of the main thermal–physical characteristics (coefficients of thermal expansion [ 59 ] and glass transition temperatures [ 103 ]) of the studied PIs. Consequently, as in our previous studies [ 104 , 105 ], (i) the ability of various partial charge parameterization methods to accurately replicate the coefficient of thermal expansion ( CTE ) in the glassy state was studied, and (ii) a comparison was made between the ratio of the glass transition temperatures and the corresponding experimental ratio ( Figures S4–S5 and Tables S1–S2 in the Supplementary Materials ). The results obtained revealed that the use of the HF/6-31G* (RESP) method reproduces the CTE value of PI ULTEM TM in the glassy state better than the AM1-BCC calculation, and the ab initio method HF/6-31G* (RESP) could qualitatively reproduce the experimental ratio T g U L T E M T M > T g B P D A − P 3 > T g R − B A P B between the T g of the considered PIs. Therefore, the HF/6-31G* (RESP) method is more convenient than the AM1-BCC one to study the influence of ordering on the thermal conductivity coefficient of PI. The influence of temperature on the thermal conductivity coefficients of PIs was also studied. It was found that an increase in temperature from 290 to 600 K led to an increase in the thermal conductivity of the undeformed samples, and there was a slight decrease in the thermal conductivity coefficient when the temperature rose above the T g values of the PIs. The results obtained were in good qualitative agreement with the experimental results [ 106 ], showing that the thermal conductivity coefficient of amorphous polymers increases with the increasing temperature to the polymer T g , and a decrease in the value of the thermal conductivity coefficient was observed at temperatures above T g . The ratio between the thermal conductivity coefficients κ B P D A − P 3 > κ R − B A P B > κ U L T E M T M of the considered PIs correlates with the ratio between the maximums of the vibrational density of state ( VDOS ) spectra ( Figures S6 and S7 in the Supplementary Materials ). It should be noted that for the unordered samples of the considered PIs, cooling from 800 K and 600 K resulted in nearly identical values of the thermal conductivity coefficient at T = 290 K. Therefore, for comparison with the results of deformed samples and self-ordering during 30 µs long molecular dynamics simulations, the thermal conductivity coefficients of the unordered samples will be similar at T = 290 K for systems cooled from T = 800 K and T = 600 K to T = 290 K. Thus, the thermal conductivity coefficient κ is calculated using the EMD approach and the HF/6-31G*(RESP) method for the partial charge calculation will be used. 3.2. Influence of Deformation on the Thermal Conductivity Coefficient To assess the impact of uniaxial deformation on the thermal conductivity of polyimide-based TIMs at various temperatures, the thermal conductivity coefficients were calculated at 290 K, 600 K, and 800 K ( Figure 3 ). The results showed an increase in the thermal conductivity coefficient κ of the deformed samples for semicrystalline BPDA-P3 and R-BAPB, as well as for amorphous ULTEM TM polyimide at T = 290 K. The strongest improvement in the heat transfer properties among all the PIs considered was observed for BPDA-P3 ( Figure 3 ). A similar enhancement in the thermal conductivity coefficient of the amorphous PI [ 34 ] and PE [ 54 ] samples was observed. To estimate the enhancement in the heat transfer properties for mechanically ordered samples, we computed the relative percentage deviation ( κ d e f − κ u n o r d ) κ u n o r d × 100 % of the thermal conductivity coefficient value κ def of the deformed samples from the value κ unord for the unordered samples, as shown in Figure 4 . Increases in κ values of 33% and 20.9% were observed for semicrystalline BPDA-P3 and R-BAPB PIs, respectively, during cooling from 800 K to room temperature. The weakest improvement in the thermal conductivity was observed for the amorphous ULTEM TM . This PI demonstrated a relative percentage deviation of 12.2% in κ value after deformation. A much better improvement in the κ value for glassy samples at room temperature was observed when cooling started at T = 600 K, at which semicrystalline polymers could be partially ordered. The relative percentage deviation almost doubled for semicrystalline R-BAPB and amorphous ULTEM TM PIs compared with the cooling that started at T = 800 K. At the high temperatures of 600 K and 800 K, the effect of mechanical orientation on the heat transfer properties deteriorated the thermal conductivity coefficient κ of the deformed samples compared to the values for the unordered PIs, as shown in Figure 4 . Note that the enhancement in thermal conductivity properties of thermally stable polyimides might improve by only 40% owing to the selected amplitude of mechanical elongation, whereas increasing the strain amplitude might sufficiently change the heat transport properties of amorphous and semicrystalline polymers [ 53 ]. The elongation of both semicrystalline and amorphous polymer samples can lead to anisotropy in κ value [ 41 ]. Furthermore, we estimated the parallel (||) and perpendicular ( ⊥ ) counterparts of κ value by using the EMD method, as shown in Figure 5 . To quantitatively estimate the anisotropy of the thermal conductivity coefficient of the deformed polyimide samples, we calculated the relative percentage deviation of the κ def value for the deformed samples from the κ unord value for unordered samples of considered PIs in different directions relative to the direction of elongation ( Figure 6 ). Concerning the deformed samples, the relative percentage deviation was evaluated in the directions parallel (||) and perpendicular ( ⊥ ) to the deformation direction. Both semicrystalline and amorphous PIs showed a significant increase in the κ value for all glassy PIs in the deformation direction compared to the κ value of unordered samples ( Figure 5 and Figure 6 ). However, for the semicrystalline PIs BPDA-P3 and R-BAPB, the enhancement in thermal conductivity is much more essential than that of the amorphous ULTEM TM . It is worth mentioning that if we compare the increase in the thermal conductivity coefficient in the deformation direction with the enhancement in the average thermal conductivity coefficient value, one can see a much higher increase along the deformation direction than along all three directions in average. The relative percentage deviation of PI BPDA-P3 along the deformation direction was 223.5% ( Figure 6 ), whereas the relative percentage deviation for the average thermal conductivity coefficient was lower and equals only to 39.2% ( Figure 4 ). In turn, in the perpendicular direction, the reduction in the size of the samples during deformation impaired the thermal conductivity. Similar results have proven that thermal conductivity anisotropy has been found for other polymers [ 42 , 45 , 46 , 53 , 55 ]. The improvement [ 41 , 45 ] of the thermal conductivity coefficient in the parallel direction and decrease in the κ value of PE [ 45 , 52 , 54 ], PS [ 54 ], and PI [ 44 ] in the lateral direction have been shown. An increase of more than 160% in the deformation direction was found for amorphous PE by the elongation of polymer systems by up to 400%. Figure 6 shows that the κ value in the perpendicular direction decreased from 40% to 90% for the semicrystalline polyimides BPDA-P3 and R-BAPB, as well as a slightly weaker decrease from 4% to 55% for the amorphous polyimide ULTEM TM . An analysis of the influence of uniaxial deformation on κ values of semicrystalline and amorphous PIs showed that the heat transfer properties of both types of PIs could be significantly modified by applying uniaxial deformation. Although the average enhancement in the κ value of the deformed samples for semicrystalline PIs at room temperature was close to 40%, the increase in the κ value in the direction parallel to the deformation was much higher than that in the perpendicular direction. In the direction parallel to the deformation, the enhancement in the thermal conductivity coefficient κ reached approximately 223.5% for semicrystalline PI and 93.3% for amorphous PI. 3.3. Comparison of the Influence of Polymer Ordering after Mechanical Elongation and Self-Ordering during Long Simulation on the Thermal Conductivity Coefficient of Semicrystalline Polyimide BPDA-P3 In this section, we discuss the effect of the difference of structural ordering on the thermal conductivity coefficient of semicrystalline polymer chains [ 107 , 108 , 109 ]. To investigate the effect of the difference of structural ordering on the thermal conductivity properties of semicrystalline PIs, we compared only the κ value of BPDA-P3 samples ordered as follows: (i) by application of uniaxial elongation to unordered BPDA-P3 samples and (ii) BPDA-P3 samples self-ordered during 30 μs long simulations [ 66 ]. Our previous study [ 66 ] examined the transport properties of BPDA-P3, a semicrystalline PI, during the unfolding and stretching of polymer chains. This ordering was initiated during tens of microseconds of simulations at temperatures slightly above the melting point of the PI BPDA-P3 ( T m = 577 K). After 30 µs of computer simulations, when the nematic order reached a value close to 0.97–0.98 [ 66 ], the initial configurations of the BPDA-P3 samples were compared with those of the mechanically stretched ones. In the present study, the systems that self-ordered during 30 μs long molecular dynamics were cooled down similarly from T = 600 K and T = 800 K to 290 K with a cooling rate γ c = 1.5 × 10 11 K/min. The temperature dependence of the density of the analyzed samples ordered during uniaxial elongation and self-ordered during a 30 μs long simulation is shown in Figure S8 in the Supplementary Materials . The ordering of the BPDA-P3 samples led to the densification of the systems. Both ordered samples that were cooled, starting from T = 800 K and T = 600 K, had very close temperature dependencies of the density. However, cooling from 600 K to room temperature led to much higher density of the PI samples. The κ values were calculated at 290, 600, and 800 K, as shown in Figure 7 . As expected, both samples ordered by mechanical deformation and self-ordered during a 30 µs long simulation demonstrated an increase in the κ value at room temperature ( T = 290 K). At T = 290 K, the samples cooled from T = 800 K showed a higher thermal conductivity coefficient in the case of mechanical deformation than the self-ordered samples during long simulations. Cooling from T = 600 K led to an increase in the κ value in the glassy state because of the much stronger preservation of the ordered structure of PI BPDA-P3. The preserved ordering of the polymer chains, which in the end greatly improves the heat transfer, compared to the situation when the structurally ordered samples cooled from T = 800 K correspond to a highly mobile melt. Although the samples were cooled quite rapidly compared to the experimental cooling rate, the instance configurations of the ordered samples apparently did not undergo substantial derangement of the polymer chains, and the κ value remained higher than that of the undeformed samples. At higher temperatures, when the temperature was 600 K or 800 K, the self-ordering during the 30 µs long simulation showed a greater increase in κ value compared to the mechanically elongated samples. To analyze how the thermal conductivity coefficient κ is enhanced in terms of quantity, we calculated the relative percentage deviation ( κ o r d − κ u n o r d ) κ u n o r d × 100 % of the thermal conductivity coefficient κ ord value for the ordered samples from that for the unordered ones’ κ unord value, as shown in Figure 8 . Analysis of the results in Figure 8 shows that the enhancement in κ values of BPDA-P3 for both ordering cases at room temperature when the samples were cooled from T = 600 K was rather close, and did not exceed 40%. However, when cooling was carried out from T = 800 K, the BPDA-P3 samples self-ordered during the long simulation showed a smaller improvement in the heat transfer properties, whereas the thermal conductivity properties of the mechanically ordered samples had a weaker deterioration. This might be because the deformation was carried out at a temperature close to the melting point of BPDA-P3. The deformation of the PI sample at a temperature close to the transition temperature may cause heat flux fluctuations that are combined with a decrease in the κ value. It is worth noting that at T = 600 K, the mechanically ordered samples demonstrated a decrease in κ value compared to the self-ordered samples over a 30 μs long simulation. When the temperature increased to 800 K, the heat transfer properties of both ordered systems increased by approximately 21.4% and 16.5%, respectively. The enhancement in the κ value might be related not only to the emergence of order in polymer chains, but also to an increase in the density of ordered systems [ 110 ]. Overall, both types of oriented systems exhibited a rather similar increase in κ value in the glassy state after fast cooling. The maximum enhancement in the thermal conductivity properties did not exceed 40% for either case when the cooling started at T = 600 K. However, when cooling started at T = 800 K, the mechanical deformation of the semicrystalline PI BPDA-P3 resulted in a more significant enhancement in the value of the thermal conductivity coefficient κ compared to κ value of self-ordered systems during a 30 μs long simulation due to the arbitrarily oriented polymer chains relative to different coordinate axes. With an increase in temperature ( T = 600 K and T = 800 K) greater than T g , the κ value of the ordered samples showed a slight decrease in enhancement compared to the κ value of the ordered sample in a glassy state."
} | 8,301 |
29847575 | PMC5976151 | pmc | 6,934 | {
"abstract": "Ocean acidification is expected to alter community composition on coral reefs, but its effects on reef community metabolism are poorly understood. Here we document how early successional benthic coral reef communities change in situ along gradients of carbon dioxide (CO 2 ), and the consequences of these changes on rates of community photosynthesis, respiration, and light and dark calcification. Ninety standardised benthic communities were grown on PVC tiles deployed at two shallow-water volcanic CO 2 seeps and two adjacent control sites in Papua New Guinea. Along the CO 2 gradient, both the upward facing phototrophic and the downward facing cryptic communities changed in their composition. Under ambient CO 2 , both communities were dominated by calcifying algae, but with increasing CO 2 they were gradually replaced by non-calcifying algae (predominantly green filamentous algae, cyanobacteria and macroalgae, which increased from ~30% to ~80% cover). Responses were weaker in the invertebrate communities, however ascidians and tube-forming polychaetes declined with increasing CO 2 . Differences in the carbonate chemistry explained a far greater amount of change in communities than differences between the two reefs and successional changes from five to 13 months, suggesting community successions are established early and are under strong chemical control. As pH declined from 8.0 to 7.8, rates of gross photosynthesis and dark respiration of the 13-month old reef communities (upper and cryptic surfaces combined) significantly increased by 10% and 20%, respectively, in response to altered community composition. As a consequence, net production remained constant. Light and dark calcification rates both gradually declined by 20%, and low or negative daily net calcification rates were observed at an aragonite saturation state of <2.3. The study demonstrates that ocean acidification as predicted for the end of this century will strongly alter reef communities, and will significantly change rates of community metabolism.",
"introduction": "Introduction The oceanic uptake of anthropogenic carbon dioxide (CO 2 ) emissions is causing ocean acidification (OA) [ 1 ]. OA not only lowers seawater pH, but also reduces the saturation state (Ω) of calcium carbonate (CaCO 3 ) minerals, and increases CO 2 and bicarbonate ion concentration. Predicting how marine communities will respond to OA is complicated, as many of these chemical alterations can act as drivers of change [ 1 , 2 ]. For example, the inhibition of calcification from declining pH and Ω [ 3 ], or the stimulus of photosynthesis from the increases in dissolved inorganic carbon (C T ) [ 2 ], may affect species performances. Such physiological responses may also cause disruptions of ecological interactions, further altering communities [ 4 , 5 ]. As OA is occurring progressively, the response of species and communities is likely to occur along a continuum as well. Individual species have displayed both linear responses [ 6 , 7 ], as well as non-linear thresholds or tipping points [ 8 ] along gradients of CO 2 , while the response of communities remains largely uninvestigated. To better predict how communities will be shaped under OA, there is thus a need for studies which investigate the response curves of communities to increasing CO 2 . Coral reefs are likely to be among the ecosystems most affected by OA [ 9 ]. Predictions are based on a multitude of single-species physiological studies [ 10 ], and several that have investigated changes at the community level. Community scale studies have centred around naturally occurring high-CO 2 analogues, such as volcanic CO 2 seep sites [ 11 – 13 ] or other oceanographic features affecting their carbonate chemistry [ 14 – 16 ], as well as larger-scale multi-species tank experiments [ 17 – 19 ]. While there is substantial variation in the responses between taxa, the general consensus predicts declines in biodiversity, a retraction of many calcifying species (e.g. scleractinian corals, coralline algae and foraminifera), an expansion of non-calcifying phototrophs (e.g. algae and seagrasses), and increased bioerosion. [ 20 ]. Coupled with the predicted changes in community composition under OA will likely be changes in community metabolism. However, scaling up OA effects on metabolic processes from individuals and species to the community level has proven difficult, and our current understanding is poor [ 21 ]. To date the best inferences have been based on naturally occurring seasonal carbonate chemistry changes [ 22 – 24 ], or the manipulation of seawater carbonate chemistry on coral reefs in situ [ 25 , 26 ] and in experimentation [ 18 , 27 ], as well as larger-scale mesocosm experiments [ 17 , 19 , 28 , 29 ]. These studies generally predict that rates of community photosynthesis and respiration will remain relatively unchanged from the reefs of today, while calcification and net CaCO 3 accumulation will decline. However, these investigations have mainly examined effects due to changes in seawater carbonate chemistry, without fully accounting for changes due to the longer-term shifts in benthic community composition that may occur under OA. For example, Ω declines may directly reduce calcification rates in numerous taxa [ 10 ], but if these taxa are then outcompeted by non-calcifers, community calcification rates may further decline. Similarly, OA can increase rates of community production by directly stimulating photosynthesis in some species [ 30 , 31 ], or indirectly by increasing the benthic cover of certain phototrophs [ 2 ]. To gain further insight into the community metabolic dynamics of coral reefs under OA, measurements must be made on communities that have developed in their entirety under altered seawater carbonate chemistries. The frequency and severity of disturbances affecting coral reefs is increasing [ 32 ], and scleractinian coral cover is now often well below 30% [ 33 ]. Scleractinian corals eventually re-enter communities, however it is early-successional non-scleractinian taxa (e.g. algae, sponges, and other sessile invertebrates) that increasingly dominate light exposed benthic reef communities [ 34 ]. Furthermore, shade exposed cryptic taxa within crevices of the reef matrix can account for the largest fraction of biomass in reef systems [ 35 ]. Both of these communities—the early successional benthic taxa on illuminated and shaded surfaces—are often overlooked in reef community metabolism studies, although their metabolism co-determines the carbonate chemistry conditions for newly settling corals within the benthic boundary layer. In this study we investigate how OA will shape the composition of early-successional benthic communities that live on the carbonate substrata of coral reefs, and the metabolic rates of the non-scleractinian components of reef communities that have developed under altered carbonate chemistries. To do so, non-carbonate settlement tiles were deployed in situ , under natural levels of light and shade, temperature and water flow, along CO 2 gradients at two volcanic CO 2 seep and two control sites in Papua New Guinea. Benthic communities developing on the upper light exposed tile sides, as well as the shaded crevice-dwelling taxa on the lower sides were investigated after five and 13 months, and their successional changes, and taxa-specific responses along the CO 2 gradients were explored. Response curves in community photosynthesis, respiration and light and dark calcification were then determined after 13 months.",
"discussion": "Discussion Ocean acidification is predicted to fundamentally alter benthic marine communities. Here we report a drastic shift in the composition and metabolism of early successional benthic coral reef communities along seawater carbonate chemistry gradients. The carbonate chemistry explained a far greater amount of change in communities than successional changes from five to 13 months, and differences between the two reefs. Shifts were more pronounced in the algae compared to the invertebrate communities, where a suite of non-calcifying algal groups largely replaced CCA on seep site tiles. Changes in CO 2 and community composition also affected community metabolism; rates of gross photosynthesis and respiration increased with increasing CO 2 , and with the cover of certain taxonomic groups, while 24-h net calcification decreased to low or even negative values. The present study adds to the mounting body of evidence predicting ecosystem-wide changes in benthic communities under OA. Here we found rapid increases in non-calcifying algal cover as pH declined along the carbonate chemistry gradients, and little evidence of threshold responses for these taxa. This pattern was largely consistent between light exposed and cryptic communities, with green filaments establishing dominance on the upper sides, and cyanobacteria and macroalgae on the lower sides. While results are not universal [ 14 ], patterns in benthic communities at CO 2 seeps in the temperate Mediterranean [ 5 , 49 – 51 ], as well as multiple tropical sites in the Indo-Pacific [ 11 , 12 ], concur with the present study. These studies similarly predict an increase in non-calcifying algae under OA, and there are suggestions that other non-calcifying phototrophs, such as seagrasses [ 11 ] and anemones [ 30 ], may also thrive. Interestingly, non-calcifying algae have increased abundances in the wider community at the seep sites in Milne Bay [ 11 ], however they do not dominate the benthos like on the settlement tiles of the present study, or at another tropical seep [ 12 ]. Grazers are diverse and abundant at the Milne Bay seeps [ 52 ], which may prevent the proliferation of algae on upper surfaces in the wider community. Similarly, longer-term competition with benthos that had not developed fully on the tiles (e.g. the scleractinian corals) may also constrain algal growth. Calcifying algae on the settlement tiles displayed dissimilar results between taxa along the CO 2 gradient. The cover of the lightly calcifying algae Peyssonnelia spp. increased at lower pH, albeit only in high-light environments. Some Peyssonnelia species have increased in abundance at other seep sites [ 49 , 51 ], suggesting that certain species of calcifying algae may be resilient to or even benefit from OA [ 42 , 53 ], perhaps due to the use of aragonite over high magnesium-calcite in their skeletons [ 51 ], and by using the additional C T for photosynthesis. The steep decline in CCA cover is consistent with data from multiple seep sites [ 8 , 12 ] and in experimentation [ 54 ] with potentially profound effects on coral reef communities [ 36 ]. The steeper decline in CCA on the lower tile surfaces, as well as the increase in Peyssonnelia spp. on the upper tile surfaces, indicates light intensity is playing a role in the response of these taxa to OA. Invertebrate responses to carbonate chemistry changes varied between taxa, and overall their cover and abundances did not respond as strongly as the algae did. It is important to note that the invertebrate communities in the present study were in relatively early successional stages, and consisted of shade-adapted communities without scleractinian corals (no invertebrates were found on the upper tile surfaces). The number of tube-dwelling polychaetes per tile declined with pH, as did the cover of ascidians. Similar declines were seen in tube-dwelling polychaete species at a Mediterranean seep [ 55 ], possibly due to reduced calcification in the juvenile stage [ 56 ]. Little is known about why ascidians appear to respond negatively to elevated CO 2 , however a previous study found the abundances of ascidians on natural reef substrata also declined with CO 2 exposure at the Milne Bay seeps [ 4 ]. Our study found no apparent effect of carbonate chemistry on the cover of the diverse groups of bivalves, bryozoans or sponges. Both bivalves and bryozoans are calcifying and considered sensitive to carbonate chemistry changes, but their CaCO 3 skeletons are somewhat protected from the surrounding seawater through external tissue layers [ 10 , 57 ]. Sponges have shown a mixed response to elevated CO 2 , with some species negatively responding, while species with phototrophic symbionts or siliceous spicules may respond positively [ 58 ]. Patterns in the 13 month old tile communities were largely established in the first five months, and successional changes between census periods were considerably weaker than the influence of the carbonate chemistry gradient for the majority of taxa. While the cover of some ephemeral (turf and green filamentous algae and cyanobacteria) and slower growing taxa ( Peyssonnelia spp ., other macroalgae and the invertebrate groups) changed between census periods, this did not significantly alter the patterns in the rest of the tile communities. For example, patterns along the CO 2 gradients in the cover of non-calcifying algae and CCA, which accounted for the majority of the tile communities, were largely consistent between census periods. This consistency between census periods contrasts a similar study at Mediterranean seeps, where Kroeker et al . [ 5 ] found commonalities between early settlement seep and control communities progressively diverged as competitive hierarchies were disrupted. Fabricius et al . [ 8 ], who closely examined patterns in CCA distributions on the tiles of the present study, concluded that it was recruitment limitation in the CCA at lower pH, rather than competition with other taxa, that established the patterns seen here. It is unknown to what extent the successional tile communities reflect the surrounding mature benthic communities. After 13 months, the tile communities blended in and greatly resembled the surrounding benthos (personal observation). Previous work at the Milne Bay seeps has similarly shown higher turf and macroalgae cover, and lower CCA cover, on natural substrate within the seep reefs compared to control reefs [ 11 ]. Regardless of any disparities between tile and mature benthic communities, early successional communities may become increasingly prevalent on coral reefs as the frequency and severity of disturbances increases [ 32 ], making larger contributions to overall reef composition and metabolic signals, with potentially important complications for the carbonate chemistry newly settling corals will experience within the benthic boundary layer. The present study presents the first investigation of metabolic changes for combined surface and cryptic subsurface reef communities that have developed entirely in situ under high CO 2 . Here we documented a 10% increase in gross photosynthesis and a 20% increase in respiration at pH T 7.8 compared to control sites with a pH T 8.0, but no change in net community production. Gross photosynthesis may increase under OA by directly stimulating photosynthesis [ 30 , 31 ], and/or by increasing the benthic cover of phototrophs [ 2 ]. Studies that have investigated metabolic changes under OA at the reef community scale are few and from quite different communities, however they have not observed the increases in gross photosynthesis reported here [ 17 , 20 , 27 , 28 ]. In the present study, models which included benthic community cover indicated that increases in gross photosynthesis were predominantly due to increases in the cover of non-calcareous algae, rather than the changing seawater carbonate chemistry per se . Respiration may increase as a consequence of increasing biomass or increased metabolism. Biomass estimates are unavailable for the tiles, however our models indicated that declining seawater pH, and increasing invertebrate cover, both significantly contributed to the observed increase in respiration at lower pH. OA is likely reducing reef calcification rates on coral reefs [ 29 , 26 ]. Light, dark and net calcification rates on the tiles all declined along the Ω Ar gradient, and our models indicated that it was these changes in Ω Ar , rather than shifts in tile communities, that were responsible. While results are not universal, OA is widely reported to reduce calcification in individual coral reef organisms [ 6 , 10 ] and at the community scale [ 17 – 20 , 29 ]. For example, Enochs et al . [ 20 ] found net daily calcification rates of light exposed coral reef communities on CaCO 3 substrata decreased linearly along CO 2 gradients, and became negative by pH T 7.8. The stronger response found by Enochs et al . [ 20 ] compared to the present study is thought to be because their study used CaCO 3 blocks as settlement substrata, and attracted many macro-boring organisms, yet did not include cryptofauna on the lower surfaces. When predicting OA effects on coral reef calcification, one must also take the permeable carbonate matrix and sediments into consideration. These are the largest sources of reef CaCO 3 [ 59 ], and they are more vulnerable to dissolution than calcifying organisms as they lack tissue layers to buffer them from the surrounding seawater [ 19 , 23 , 59 , 60 ]. For example, Comeau et al . [ 61 ] documented a 60% decline in the calcification of experimental coral reef communities at 1300 μatm pCO 2 , with half of this being attributed to sediment decalcification. Some estimates suggest that even if coral calcification rates are maintained under OA, the dissolution of carbonate sediments alone would result in reef loss [ 59 , 62 ]. Results from the calcification assays in the present study, lacking sediments and a CaCO 3 substrata, are thus likely to considerably under-estimate reef-wide dissolution rates expected under OA. Instead, they provide insight into how the calcification dynamics of a part of the biological components of coral reef communities may respond. There are several factors that preclude CO 2 seep sites from perfectly representing the future of the world’s oceans. Firstly, they are relatively small, and scaling up predictions to the world’s coral reefs will undoubtedly introduce some uncertainty. Secondly, the altered carbonate chemistry at the seeps is occurring in isolation from the warming that is also predicted for a high CO 2 world [ 46 ], and the combined effect of these two stressors can be greater than either in isolation [ 10 ]. Thirdly, the seep site A T is slightly elevated (by ≤ 6% of control values), which may increase calcification rates at the seep sites [ 26 ]. Increased dissolution of carbonate sediments under OA may locally increase A T , however sediment dissolution- and dilution-rates from the surrounding seawater are largely unknown [ 59 ]. And finally, the seep seawater carbonate chemistry is characteristically variable over short (i.e. hourly) time-scales [ 63 ] with uncertain consequences for coral reef communities. Scleractinian corals have been shown to be largely robust, or to even benefit from increased pH variability (when compared to static lowered pH), while other coral reef organisms (e.g. CCA) can be negatively affected [ 64 ]. On the other hand, community scale studies, which allow for interactions between species and their environment, are not easily conducted in laboratory settings. While seep sites studies are not definitive, they do provide unique opportunities to overcome some issues with laboratory-based OA studies (i.e. organism acclimation and species/environment interactions) and provide further contributions to scientific consensus about the severe effects ocean acidification is afflicting on marine communities. Results from the two seep sites investigated here, as well as other naturally occurring high CO 2 analogues [ 12 – 14 ], generally agree with a plethora of experimental work from the small- [ 6 , 10 , 65 ] to large-scale [ 17 – 19 , 25 , 29 , 66 ], in situ seasonal comparisons [ 23 , 44 ] and quantitative models [ 67 – 69 ]. All these generally predict considerable changes for coral reefs under ‘business as usual’ carbon emissions scenarios and that we will likely see shifts in community composition, with the proliferation of non-calcifying taxa and a retraction of many calcifiers. Increases in non-calcifying algae may lead to increases in community gross production, however gross gains may be balanced by increased respiration. Ecosystem-wide calcification and CaCO 3 accumulation rates will likely decline, owing to the carbonate chemistry changes and the dissolution of carbonate sediments and increased bio-erosion [ 14 , 20 ]. Unfortunately neither this study, meta-analyses [ 1 , 10 ], nor experimental comparisons of pre-industrial to present-day conditions [ 17 , 26 ] have shown signs that ecosystem acclimation will prevent the expected changes. This study further shows that many changes expected on coral reefs under increasing OA will occur along a continuum, indicating that the less CO 2 emitted into the atmosphere, the less deviation we will see from the reefs of today."
} | 5,250 |
34761294 | PMC8776678 | pmc | 6,936 | {
"abstract": "Acetobacteraceae is an economically important family of bacteria that is used for industrial fermentation in the food/feed sector and for the preparation of sorbose and bacterial cellulose. It comprises two major groups: acetous species (acetic acid bacteria) associated with flowers, fruits and insects, and acidophilic species, a phylogenetically basal and physiologically heterogeneous group inhabiting acid or hot springs, sludge, sewage and freshwater environments. Despite the biotechnological importance of the family Acetobacteraceae , the literature does not provide any information about its ability to produce specialized metabolites. We therefore constructed a phylogenomic tree based on concatenated protein sequences from 141 type strains of the family and predicted the presence of small-molecule biosynthetic gene clusters (BGCs) using the antiSMASH tool. This dual approach allowed us to associate certain biosynthetic pathways with particular taxonomic groups. We found that acidophilic and acetous species contain on average ~ 6.3 and ~ 3.4 BGCs per genome, respectively. All the Acetobacteraceae strains encoded proteins involved in hopanoid biosynthesis, with many also featuring genes encoding type-1 and type-3 polyketide and non-ribosomal peptide synthases, and enzymes for aryl polyene, lactone and ribosomal peptide biosynthesis . Our in silico analysis indicated that the family Acetobacteraceae is a potential source of many undiscovered bacterial metabolites and deserves more detailed experimental exploration. Supplementary Information The online version contains supplementary material available at 10.1007/s10482-021-01676-7.",
"conclusion": "Conclusion The family Acetobacteraceae belongs to the class Alphaproteobacteria , and members of this class are not generally considered prolific producers of specialized metabolites, despite some strains carrying more than forty BGCs (Mukherjee et al. 2017 ). A relatively small number of molecules have been characterized from this taxonomic class, however so far, no specialized metabolites (< 2 kDa) have been purified from strains of the family Acetobacteraceae . In this study we were able to predict that all members of the Acetobacteraceae are producers of hopanoids. These triterpenoids play a fundamental role in the integrity of the bacterial cell membrane, particularly under stressful conditions such as low pH, but given the presence of two distinct hopanoid BGCs in a number of acetous species it is possible that these metabolites have additional functions. The acidophilic group featured almost twice as many BGCs as the acetous group. Most of the strains in both groups carried at least one type-1 PKS, and most members of the acidophilic group showed at least one NRPS and one type-3 PKS. The acetous group was found to produce ribosomally synthetized peptides belonging to the linocin M18—encapsulin family. A smaller number of strains in both groups appear able to produce aryl polyenes, lactones and siderophores. Thus far, none of these specialized metabolites have been purified, and the translation of metabolic potential in silico to actual metabolic capability remains to be confirmed. Given the diverse ecological niches occupied by the Acetobacteraceae , including ponds, sludge, soil, sediments, fruits, flowers and insect guts, the specialized metabolites produced by these species are likely to be bioactive and may be suitable for biotechnological exploitation.",
"introduction": "Introduction Acetobacteraceae is an economically important family of bacteria, with several strains used for industrial biotechnology applications including the commercial production of vinegar and fermented foods, bacterial cellulose, and sorbose, a key precursor of vitamin C (Lynch et al. 2019 ; Murooka 2016 ; Pappenberger and Hohmann 2014 ). The family is divided into two groups: acetous and acidophilic species (Hördt et al. 2020 ; Komagata et al. 2014 ). Acetous species are also known as acetic acid bacteria (AAB) and most can transform ethanol into acetic acid, although there are some exceptions such as Asaia spp. (Malimas et al. 2017 ). AAB are typically found in flowers, fruits and other sugary organs of plants, and in traditional vinegars and other fermentation products (Yamada 2016 ), although some have recently been shown to consistently associate with insects (Guzman et al. 2021 ; Li et al. 2015 ; Roh et al. 2008 ). Acidophilic species appear to be phylogenetically more distant from the AAB (Hördt et al. 2020 ), and show diverse phenotypes and adaptations, including acidophilic, neutrophilic, thermophilic and phototrophic characteristics (Komagata et al. 2014 ). This group has been isolated from paddy soils, acid or hot springs, soil crust, sludge, sewage, freshwater ponds, air-conditioning systems, and certain Roseomonas strains have even been isolated from human patients (Dé et al. 2004 ; Sievers and Swings 2015 ). The Acetobacteraceae currently includes 44 genera and 177 valid species, split into 19 genera and 97 species of AAB, and 25 genera and 80 species of acidophilic bacteria (Parte et al. 2020 ). The family belongs to the order Rhodospirillales , class Alphaproteobacteria , and their closer siblings are the recently proposed families Stellaceae and Reyranellaceae , based on phylogenomic and phenotypic analysis (Hördt et al. 2020 ). Some species currently classified as acidophilic bacteria are likely to be assigned to new families in the future when more genomic data become available. No specialized metabolites (< 2 kDa) have been reported thus far from any member of the family Acetobacteraceae , but it remains an untapped potential source of natural products given that related taxa appear to carry tens of biosynthetic gene clusters (BGCs) based on wide genomic analysis (Mukherjee et al. 2017 ). The production of specialized metabolites has been intensively studied in streptomycetes and myxobacteria because they are known producers of antibiotics. The production of metabolites depends on the ecological context, in which the synthesized compounds confer competitive advantages to the producer, overcoming the energy costs of maintaining the BGCs (Hoskisson and Fernández-Martínez 2018 ; Jensen 2016 ). BGCs often encode not only enzymes but also other essential complementary proteins such as assembly scaffolds, metabolite resistance factors and regulatory effectors. Computational methods have been developed to identify the presence of BGCs in the exponentially growing resource of microbial genomic data (Medema et al. 2021 ; Medema and Fischbach 2015 ). The standard tool for this purpose is antiSMASH, which interrogates the protein sequences encoded in the genomes for sequence similarity to a library of hidden Markov models extracted from core biosynthetic proteins (Medema et al. 2011 ). The cluster boundaries are expanded to include other nearby core proteins, and accessory proteins in the vicinity are detected (Blin et al. 2017b ). The search is finalized by evaluating the similarity of the detected gene set to known BGCs. One of the main limitations of library-based genome mining is that it detects proteins similar to known biosynthetic proteins but excludes unknown proteins that might produce entirely new molecules (Blin et al. 2017a ). Here we took 127 published genomes of Acetobacteraceae type strains and used them for phylogenetic analysis and genome mining in order to find correlations between cladistics and the conservation of certain specialized biosynthetic traits. Our results will help to focus discovery efforts on bacterial producers of novel metabolites with potential applications in the pharmaceutical and agrochemical industries.",
"discussion": "Results and discussion GC content and genome size A graphic plot examination of the variation in GC-content and genome size values (Fig. 1 a) for the 139 type strains supported a rough separation of the family Acetobacteraceae into the acetous and acidophilic groups. The Acetobacteraceae genomes not classified as AAB (with the exception of Acidocella aminolytica 101 T ) showed a narrow GC content range (Δ ~ 11%mol) with values between 62.7%mol ( Roseomonas cervicalis ATCC 49957 T ) and 73.9%mol ( Crenalkalicoccus roseus YIM 78023 T ). However, this group showed a large variation in genome size (Δ ~ 4.8 Mbp), ranging from 3.03 Mbp ( Elioraea thermophila YIM 72297 T ) to 7.78 Mbp ( Dankookia rubra JCM 30602 T ). In contrast, the GC content of the AAB varied widely (Δ ~ 31%mol), with values between 36.8%mol ( Commensalibacter intestini A911 T ) and 67.7%mol ( Endobacter medicaginis LMG 26838 T ). However, this group showed less variation in genome size (Δ ~ 2.8 Mbp), ranging from 2.01 Mbp (“ Parasaccharibacter apium A29 T ”) to 4.83 Mbp ( Gluconacetobacter sacchari LMG 19747 T ). Interestingly, AAB genera isolated exclusively from the insect gut, consisting of the genera Bombella (= “ Parasaccharibacter ”), Commensalibacter and Entomobacter , clustered in a region of low genome size within the acetous group, suggesting an ongoing evolutionary reduction of genome size probably reflecting their symbiotic lifestyles. Fig. 1 GC content vs genome size plot and phylogenomic tree for Acetobacteraceae type strains a GC content and genome size plot grouping the type strains from each genus under the same symbol. The strains Azospirillum lipoferum 59b T and Skermanella aerolata KACC 11604 T are used as outgroups for the family Acetobacteraceae . The plot reveals three groups of bacteria with some degree of overlap, globally differentiated as mostly acidophilic, mostly acetous, and acetous species often associated with insects. b Phylogenomic tree inferred from 50 housekeeping protein sequences showing the two different groups of the family Acetobacteraceae . The topology of the tree is supported by both Bayesian and maximum likelihood inference methods. Distinct clades (based on monophyly and a shorter branch length distance) were proposed particularly for the acetous group. The species organization into clades is detailed in Supplementary Table 2 Phylogenomics Phylogenomic analysis based on core protein sequences confirmed that the acetous group originated from a lineage, probably already inhabiting low-pH environments, derived from the more basal acidophilic group (Fig. 1 b). Acetobacteraceae type strains were organized into suprageneric or infrageneric clades (Fig. 1 b and Supplementary Table 2) according to the position in the phylogenomic tree. Nine distinct clades were recognized within the acidophilic group: the early separating branch containing the genus Elioraea , followed by a number of recently proposed groups reorganizing the genus Roseomonas (Rai et al. 2021 ), the pool of strains representing Belnapia and the related genera Caldovatus , Crenalkalicoccus , Dankookia , Roseicella and Siccirubricoccus , Rhodovarius and the related genera Roseococcus and Rubritepida , Acidocella and Acidiphilium strains, and finally the lineages composed by strains of Acidibrevibacterium and Rhodovastum , which shared a late common ancestor with the acetous group. The polyphyletic origin of the genus Roseomonas sensu stricto observed in this study is in agreement with the recent reclassification (Rai et al. 2021 ). The topology of the acetous group confirmed the current accepted demarcation of most genera with a few exceptions. As previously suggested (Yamada et al. 2012 ), the type strain Gluconacetobacter entanii LTH 4560 T belongs to the genus Komagataeibacter . Given the high-support nodes indicating late common ancestors within certain members of a same genus, we proposed subgroups (clades named according to the most ancient described type species) for the genera Acetobacter , Gluconobacter and Komagataeibacter (Fig. 1 b). The topology of the phylogenomic tree obtained in this study is in full agreement with the latest accepted treatment (Hördt et al. 2020 ). Biosynthesis of specialized metabolites The BGCs identified in the family Acetobacteraceae using antiSMASH were organized in 10 groups according to the metabolite class or pathway: arylpolyene, ectoine, lactone, type-1 polyketide synthase (PKS), type-3 PKS, ribosomally synthesized and post-translationally modified peptide (RiPP), siderophore, terpenoid, non-ribosomal peptide synthetase (NRPS), NRPS/PKS hybrids, and miscellaneous BGCs. Because antiSMASH search is based on previously identified and studied clusters, it is possible that completely new biosynthetic pathways are missed, and thus the results here described do not necessarily exhibit the full biosynthetic potential of this group of understudied bacteria. All type strains of the family carried at least one BGC and the global average was 4.3 ± 2.3 BGCs/genome. Members of the acidophilic group carried on average 6.30 ± 2.63 BGCs/genome, which was twice as many as the acetous group, which carried on average 3.35 ± 1.48 BGCs/genome (Fig. 2 a). This difference was statistically significant ( p < 0.001) based on both the Kruskal–Wallis test and the Benjamini–Hochberg test. All the Acetobacteraceae type strains carried gene clusters involved in terpenoid biosynthesis. These BCGs were found on average at a frequency of ~ 2.4 BGCs/genome in the acidophilic group, and ~ 1.5 BGCs/genome in the acetous group (Fig. 2 b). In general, the acidophilic group carried a larger number of class-specific BGCs than the acetous group, and this was particularly evident for PKS and NRPS genes (Fig. 2 b). The genomes of some strains featured a high number of PKS-encoding genes, for example Roseomonas stagni DSM 19981 T , whereas in other genomes, for example Roseomonas aerophila NBRC 108923 T , NRPS-encoding genes were predominant. The one BGC that was present in higher numbers in the acetous group was involved in the production of RiPPs (Fig. 2 b). Fig. 2 Presence of biosynthetic gene clusters (BGCs) in the two groups of the family Acetobacteraceae . a The number of BGCs per genome was plotted for each type strain, organized according to the taxonomic classification into acetous and acidophilic species. b BGCs for the biosynthesis of different metabolite classes were plotted for each type strain and were organized according to the taxonomic classification into acetous and acidophilic species. The numbers inside the boxplots are the calculated mean values A direct correlation between taxonomy or phylogeny and the presence of certain types of BGCs was not evident, but certain trends were observed (Fig. 3 ). For example, the acidophilic group generally carried some miscellaneous gene clusters for phosphonates and indoles, whereas no acetous species carried these BGCs (Fig. 3 ab). Most of the species (~ 81%) in the genus Roseomonas carried at least one polyketide cluster with the exception of R. mucosa NCTC 13292 T , R. rosea DSM 14916 T , R. aerophila NBRC 108923 T and R. cervicalis ATCC 49957 T . All Gluconacetobacter species featured a type-1 PKS, whereas not a single strain of the related genus Komagataeibacter carried a polyketide cluster, and thus the product (still unknown) can be considered as a taxonomic marker. The genus Acetobacter tended to feature more BGCs among the acetous group (Fig. 3 ac), with the highest numbers identified in the strain A. senegalensis LMG 23690 T (seven BGCs). Acetobacter species were proficient in the biosynthesis of lactones and non-ribosomal peptides, whereas members of the Asaia-Bombella clade carried gene clusters involved in the biosynthesis of polyketides. Fig. 3 Phylogenomic analysis of the family Acetobacteraceae and their biosynthetic gene clusters (BGCs) as detected using antiSMASH. a Phylogenomic tree based on 50 housekeeping protein sequences. b Type and number of BCGs in the genomes of each type species. c Total number of BCGs with at least one core gene detected using antiSMASH. The subgroups were classified according to the class or pathway of the metabolite as follows: A = terpenoid, B = aryl polyene, C = ribosomally synthesized and post-translationally modified peptide, D = ectoine, E = lactone, F = siderophore, G = type-1 polyketide, H = type-3 polyketide, I = hybrid polyketide/non-ribosomal peptide, J = non-ribosomal peptide, K = other specialized metabolites Aryl polyenes Aryl polyenes (APEs) are bacterial pigments produced abundantly by the phylum Proteobacteria , and like carotenoids these unsaturated molecules play a role in the capture of free radicals to prevent oxidative stress (Schöner et al. 2016 ). The biosynthesis of APEs involves the loading of an aromatic precursor (usually 4-hydroxybenzoic acid) onto an acyl carrier protein (ACP), named ApeE, in a process catalyzed by ApeH (Grammbitter et al. 2020 ). The central enzyme β-ketoacyl-ACP synthase (ApeO/ApeC) elongates the chain in a decarboxylation Claisen condensation with malonate units, and the cetone product is reduced to alcohol by ApeQ and dehydrated to a double bond by ApI/ApeP in an iterative process. For some metabolites, the aryl polyene is linked to N -acetylglucosamine by the glycosyltransferase ApeJ. The presence of homologs of the core β-ketoacyl-ACP synthase among Acetobacteraceae allowed the identification of APE producers. The acidophilic group featured a higher proportion (19/43 = 44%) of APE gene clusters than the acetous group (28/96 = 29%), suggesting that bacteria readily exposed to sunlight, such as those inhabiting ponds, probably produce APEs for protection against UV radiation. The β-ketoacyl-ACP synthase encoded in the genomes of Gluconacetobacter sacchari LMG 19747 T and Swaminathania salitolerans NBRC 104436 T differed from the other homologs in the family (Supplementary Fig. 1a), and were probably transferred horizontally from other organisms (likely from Gammaproteobacteria, given that homologous proteins were identified in Yersinia, Serratia and Pseudomonas ). The unrooted tree of β-ketoacyl-ACP synthase (Supplementary Fig. 1a) was broadly congruent with the phylogenomic tree based on core genes (Fig. 1 b), with a distinct separation of the acetous and acidophilic groups. Some genes linked to the β-ketoacyl-ACP synthase gene were tentatively annotated as encoding an adenylate-forming protein, a dehydrogenase, a probable halogenase, and transport proteins, whereas others were hypothetical. The aryl polyene BCGs of the family Acetobacteraceae appear likely to produce as yet undescribed aryl polyene products. Ectoines Ectoines are bacterial natural products sharing a 4-carboxylic acid pyrimidine that promote survival in hyperosmotic environments (Czech et al. 2018 ). Ectoines are synthesized from l -aspartate-β-semialdehyde by the sequential action of three proteins: EctB, EctA and EctC (Czech et al. 2018 ). The final enzyme (EctC) is known as l -ectoine synthase, and catalyzes the transformation of N 4 -acetyl- l -2,4-diaminobutanoate to l -ectoine, acting as a marker for the identification of ectoine BGCs. Homologs of this EctC protein in Paenibacillus lautus NBRC 15380 (Czech et al. 2019 ) were detected in only five Acetobacteraceae type strains: Acidocella aminolytica 101 T , Acidiphilium cryptum JF-5 T , Acidiphilium multivorum AUI301 T , Acetobacter nitrogenificens DSM 23291 T and Gluconobacter wancherniae NBRC 103581 T (Supplementary Fig. 1b). The three members of the acidophilic group have a full ectoine cluster, including genes encoding other enzymes in the pathway such as EctA, EctB, EctD and the l -aspartate kinase (Ask) together with a transporter and a transcriptional regulator. Interestingly, the acetous clusters with an ectC gene carried no linked homologs of ectA , ectB, ectD or ask , and thus it is uncertain whether the encoded EctC protein is a functional l -ectoine synthase or has a different role. An EctA homolog was identified in A. nitrogenificens DSM 23291 T but not in G. wancherniae NBRC 103581 T . Homologs of the N -acetyltransferase EctB were detected in both strains but also in many other Acetobacter and Gluconobacter species, suggesting involvement in a more general pathway. Finally, no EctD homologs were detected in any species of the acetous group. These results suggest that the acetous group probably does not produce ectoines, and the functional role of EctC homologs in A. nitrogenificens DSM 23291 T and G. wancherniae NBRC 103581 T should be investigated in more detail. Hopanoids The most common protein sequence encoded in the Acetobacteraceae terpenoid clusters was used as a blastp query, resulting in the identification of an enzyme involved in hopene biosynthesis. Hopanoids are pentacyclic bacterial triterpenoids that confer fluidity and integrity to the cell membrane in a similar manner to sterols (such as cholesterol and sitosterol) in eukaryotes (Sáenz et al. 2015 ). All Acetobacteraceae type strains carried genes for hopanoid synthesis. The most common hopanoid BGC consisted of genes for squalene-hopene cyclase (SHC) and two squalene synthases. This type of cluster was often found in the acetous group, and hopanoids may therefore protect the cell membrane against injury caused by acetic acid (Belin et al. 2018 ; Welander et al. 2009 ). In the acidophilic group, these biosynthetic genes were not clustered together and typically one or more was missing. SHC is the central enzyme of hopanoid biosynthesis and is responsible for the cascade polycyclization of squalene leading to the pentacyclic hopene triterpenoid (Siedenburg and Jendrossek 2011 ). The unrooted tree based on SHC amino acid sequences showed two groups (Supplementary Fig. 1c), the major group bearing a perceptible phylogenetic signal. Certain species of Acetobacter , Gluconobacter and Komagataeibacter encoded two versions of the SHC protein (Supplementary Fig. 1c), and although most Acetobacteraceae carried the most common SHC, some carried only the second type. This second SHC shared a consensus sequence of ~ 30 amino acids near the C-terminus that is not present in the major SHC or in homologous proteins from Streptomyces but is found in some species of the genera Zymomonas , Bradyrhizobium and Rhodopseudomonas . Homologs of the two versions of SHC found in certain Acetobacteraceae also occur in Zymomonas mobilis and their activity has been verified experimentally (Seitz et al. 2012 ). The three-gene hopanoid core BGC of the acetous group also contained additional genes for accessory proteins, the most common of which were annotated as a glycosyltransferase, a FAD-dependent oxidoreductase, and a NAD-dependent epimerase/hydratase. In a number of species of Roseomonas (such as R. aerilata DSM19363 T \n R. nepalensis S9 T and R. oryzae KCTC42542 T ), the core hopene cyclase gene was linked to a phosphoenolpyruvate mutase gene, which is the marker for organophosphonic acid synthesis (Horsman and Zechel 2017 ), suggesting it is part of a hybrid cluster that generates a yet unknown compound. In Acetobacter malorum LMG 1746 T , the alternative hopene cyclase gene was linked to an auto-inducer synthase gene, indicating that further natural hopanoids with yet unknown functions may exist. Lactones Two types of lactone BGCs were found in the Acetobacteraceae genomes, encoding the enzymes needed for the production of acyl-homoserine lactones (AHLs) and β-lactones, respectively. AHLs are involved in quorum sensing (QS), an intercellular communication process that triggers coordinated gene expression (Waters and Bassler 2005 ). AHLs are QS auto-inducing factors because they bind to a transcription factor (LuxR in Aliivibrio fischerii ) which activates the expression of the gene encoding the AHL synthase (LuxI in A. fischerii ), resulting in the massive production of AHLs throughout the population. LuxR and LuxI homologs are widespread in Proteobacteria (Case et al. 2008 ; Schuster et al. 2013 ). AHLs are produced from S -adenosylmethionine by cleavage, cyclization and N -acylation with an ACP or acyl-coenzyme A (Schaefer et al. 2018 ). Genes for AHL biosynthesis were identified in only three species of the acidophilic group (7%): Acidocella aminolytica 101 T (two clusters), Acidibrevibacterium fodinaquatile G45-3 T and Roseomonas nepalensis S9-3B T . However, they were found more frequently in the acetous group (19/96 = 20%), being present in four species of Gluconacetobacter , three species of Komagataeibacter and twelve species of Acetobacter . Detailed analysis of protein alignments of the auto-inducer synthases (LuxI homologs) revealed three major groups, two of them specific for Acetobacter , and the third shared between Komagataeibacter and Gluconacetobacter (Supplementary Fig. 1d). Some Acetobacter species belonging to the orleanensis clades ( A. cerevisiae LMG 1625 T , A. malorum LMG 1746 T and A. orientalis 21F-2 T ) carried both types of auto-inducer synthases, suggesting the importance of QS in certain AABs used for the fermentation of must, fruit and cereal (Guillamón and Mas 2009 ; Iida et al. 2008 ; Valera et al. 2016 ). The protein sequence of the auto-inducer synthases from the acidophilic group were distantly related to those identified in the acetous group. Particularly those found in A. fodinaquatile G45-3 T and R. nepalensis S9-3B T showed to have significantly different sequences, evidenced by the long branches in the phylogenomic tree (Supplementary Fig. 1d), suggesting the existence of an alternative QS mechanism (or a different biochemical function) that should be investigated in future experiments. BGCs responsible for β-lactone biosynthesis were not identified in the acetous group but were found in thirteen acidophilic type strains, exclusive of the Roseomonas and Belnapia clades. The species Roseomonas wenyumeiae Z23 T and Siccirubricoccus deserti SYSU D8009 T carried two versions of the β-lactone BGC. Three β-lactone core enzymes were encoded by β-lactone clusters: a β-lactone AMP-binding protein supposedly catalyzing the coupling of a carboxylic acid (such as acetate) to coenzyme A, an HGML-like protein catalyzing the Claisen condensation of the acyl-CoA with a carboxylic acid to produce a β-ketoacid, and a dehydrogenase that reduces the intermediate to a β-hydroxyacid (Robinson et al. 2019 ). The final cyclization to the β-lactone is carried out by an ATP-dependent cyclase homologous to OleC (Robinson et al. 2019 ), but such a protein was not encoded by any of the BGCs. It is therefore unclear whether the product of these clusters is a β-lactone or a β-hydroxyacid. The β-hydroxyacid product may be a precursor in another specialized pathway, given that the β-lactone cluster in some Roseomonas strains (such as R. pecuniae DSM 25622 T and R. vastitatis CPCC 1121 T ) is fused with an NRPS cluster. Polyketides We identified type-1 and type-3 PKS genes in the Acetobacteraceae type strains. Type-1 PKS genes encode large proteins organized into modules that use ACPs to activate acyl-CoA substrates, whereas type-3 PKS genes encode products that act directly on acyl-CoA substrates and often produce cyclized aromatic polyketides (Jenke-Kodama et al. 2005 ; Shen 2003 ). We detected a type-1 PKS in 70% of the acidophilic species (30/43) and in around one third of the acetous species (33/96). The PKS genes were found in specific taxonomic groups such as the Roseomonas clade (Fig. 3 ab), both clades of Gluconacetobacter , as well as Asaia and Bombella-Saccharibacter and in certain species of Acetobacter and Gluconobacter . For a yet unknown reason, type-1 PKS genes were not found in the genus Komagataeibacter . The high degree of PKS conservation in the different AAB clades rules out horizontal transfer and suggests that the resulting metabolites conferred advantages on the common ancestor and remain beneficial to the extant species in their current ecological context. A basic motif found in most Acetobacteraceae type-1 PKS proteins consisted of the ordered domains KS-AT-DH-ER-KR-PP (ketosynthase-acyltransferase-dehydratase-enoylreductase-ketoreductase-phosphopantheteine acyl carrier). In each strain, this basic motif was accompanied by a variety of small domains including aminotransferases (AmT), AMP-binding domains (A), coenzyme A-binding domains (CAL), enoyl-CoA hydratase/isomerase domains (ACH), NAD-dependent epimerase/dehydratase domains (NAD), further KR or ER domains, and/or a combination of such domains. Intriguingly thioesterase domains could not be identified within the PKS protein or as stand-alone accessory proteins. In all cases, the KS domains clustered with the type-1 PKS gene, such as those associated with the synthesis of aureothin or certain aromatic polyketides (Chen and Du 2016 ). The presence of a single module suggests that the Acetobacteraceae PKS system is iterative and not modular. The PKS amino acid sequence is considered a good proxy to infer the number of metabolic products. The unrooted tree of Acetobacteraceae type-1 PKS proteins based on sequence alignment revealed four different clades, which we named α, β, γ and δ (Fig. 4 a). The α-group included all PKS proteins from the Roseomonas clade, except a second type-1 PKS identified in Roseomonas stagnii DSM 19981 T and Roseomonas algicola PeD5 T , both of which clustered in the δ-group together with Acidiphilium angustum ATCC 35903 T , Lichenicola cladoniae PAMC 26569 T and Rubritepida flocculans DSM 14296 T . The PKS of the α-group carried phylogenetic signal as the clades Pararoseomonas , Pseudoroseomonas , Belnapia , Neoroseomonas and Falsiroseomonas were clearly distinguished (Fig. 4 a). In all the α-group, a glycosyltransferase gene probably belonging to family GT4 (Breton et al. 2005 ), was found upstream of the PKS gene (Fig. 4 b). The α-group also included genes for a PLP-dependent aminotransferase, a formyltransferase and a capsular biosynthetic protein. The metabolite produced by these α-group type-1 PKS clusters is anticipated to have the same skeleton decorated with small variations given the different accessory proteins encoded within each cluster. The β-group included PKS proteins from the genera Acetobacter and Gluconobacter , and a branching group leading to the Asaia and Bombella clades (Fig. 4 a). The β-group type-1 PKS clusters differed from α, γ and δ clusters given the absence of a PLP-dependent aminotransferase gene, and instead the PKS gene was flanked by acyl ligase genes (Fig. 4 c). The PKS-encoding gene in Asaia spp. likely split into two genes, and is accompanied by proteins having hint-domain. O -heptosyltransferase gene was also consistently found in the β-group clusters, and was duplicated in the Bombella clade and the Gluconobacter spp. clusters. In addition, the PKS cluster from the Bombella clade was closely related to the cluster found in Swaminathania salitolerans NBRC 104436 T and both clusters shared the presence of an additional glycosyltransferase and a thioredoxin, suggesting they may produce sulfur-containing metabolites. The γ-group was restricted to members of the genus Gluconacetobacter , and intriguingly these PKS proteins were more closely related to those from the acidophilic group rather than the rest of the acetous group (Fig. 4 a) The γ-group type-1 PKS clusters (Fig. 4 d) were highly conserved in gene organization and protein sequence, and probably synthesize the same metabolite, perhaps with the exception of G. tumulisoli LMG 27802 T . These clusters encoded a PLP-dependent aminotransferase, two capsular biosynthetic proteins and two glycosyltransferases. A small number of type-1 PKS from acidophilic species clustered in the δ-group (Fig. 4 a), and the BGCs (Fig. 4 e) encoded a PLP-dependent aminotransferase and an oxidoreductase (and a sulfotransferase for the BGCs in Roseomonas spp.), located near the central type-1 PKS gene. Fig. 4 Type-1 polyketide synthase biosynthetic gene cluster in Acetobacteraceae . a Unrooted tree based on type-1 PKS showing the differentiation into four groups labelled α, β, γ and δ which correlate with certain taxonomic clades. Organization of the biosynthetic gene clusters for the type-1 PKS from the groups α b β c γ d and δ e showing the probable annotation of certain genes according to antiSMASH and blast analysis A type-3 PKS was identified in 21 of the 43 acidophilic type strains (49%) and from the acetous species, only in the lichenous strains Lichenicoccus roseus KEBCLARHB70R T and Lichenicola cladoniae PAMC 26569 T . This correlates with a specific evolutionary niche within plants but not lichens, where the metabolic product of the type-3 PKS cluster was unnecessary for phytosphere adaptation. Type-3 PKS was found in several strains of the genera Roseomonas (11/21 = 52%). The closest sequences beyond the Acetobacteraceae were identified in other Alphaproteobacteria , including Azospirillium , Methylopila , Microvirga and Paracoccus species. A similar type-3 PKS is ArsC (sequence identity ~ 29%) from Azotobacter vinelandii strain OP, which produces alkylresorcinols and alkylpyrones to protect its cysts against environmental injury (Funa et al. 2006 ). The type-3 PKS proteins from the acidophilic group are therefore likely to be involved in pyrone or resorcinol biosynthesis, and may also play a protective role because this group of bacteria thrives in sediments, soils, ponds and hot springs (Komagata et al. 2014 ) where solar radiation and desiccation can be detrimental. Type-3 PKS proteins from the family Acetobacteraceae could be assigned to three groups based on sequence alignment and phylogeny (Supplementary Fig. 2a). Specifically, we observed the divergence of Roseomonas frigidaquae JCM 15073 T and Roseomonas stagni DSM 19981 T (closer to Belnapia and Siccirubricoccus than to the main Roseomonas group). In addition to the central type-3 PKS, two other proteins were encoded in all the clusters: a methyltransferase and a FAD-dependent monooxygenase (Supplementary Fig. 2b). Interestingly, more closely related homologs of the methyltransferase were identified in other Rhodospirillales , such as Azospirillium , Indioceanicola and Skermanella species, but also in the myxobacterium Sorangium cellulosum , a recognized producer of specialized metabolites (Schneiker et al. 2007 ). The presence of methyltransferases and flavin-dependent monoxygenases is a common feature of certain type-3 PKS clusters particularly those found in fungi (Navarro-Muñoz and Collemare 2020 ) and in some myxobacteria (Hug et al. 2019 ), but the metabolite produced by Rhodospirillales is currently unknown. NRPS and hybrid NRPS/PKS clusters NRPS genes were present in 47% (20/43) of the acidophilic species, and some strains featured multiple NRPS or NRPS-like clusters such as R. aerophila NBRC 108923 T with four. The NRPS clusters were much less common among the acetous species, being present in only 15% (14/96). Like PKS genes, NPRS genes encode megasynthases organized into modules, including condensation (C), adenylation (A), thiolation (also known as peptidyl carrier protein, PCP), and thioesterase (TE) domains. Like the ACP in PKS, the PCP is activated by the transfer of a 4′-phosphopantetheine factor. Among the acidophilic species, seven of the 28 NRPS genes were trimodular, five were monomodular, six were bimodular, six were tetramodular, two were pentamodular and one hexamodular and one octamodular ( Roseomonas wenyumeiae Z23 T ). In contrast, ten of the 15 NRPS genes in the acetous group were monomodular, two were bimodular, one trimodular, one pentamodular and one hexamodular. The lower number of NRPS clusters among the acetous species may probably reflect genome reduction induced by plant speciation events. The lack of these specific NRPS clusters in both Acetobacteraceae clades exclusive to insects (Bonilla-Rosso et al. 2019 ) is consistent with this hypothesis, and suggests that such peptides are probably more important for bacteria living in soil, sediment or water environments, where there exists higher chances of encountering diverse microbes. Only a few species carried complete C-A-PCP-TE domains in a single protein (Supplementary Fig. 2c). In the acetous group, only two closely related Acetobacter species ( A. malorum LMG 1746 T and A. cibinongensis NBRC 16605 T ) featured these domains in a single monomodular NRPS, whereas A. aceti NBRC 14818 T featured NPRS genes with the C-A-PCP-TE domains split into adjacent modules and also contained further modules with AmT and CAL domains, which are more common in PKS genes. Two strains of Komagataeibacter carried an NRPS-like cluster ( K. diospyri MSKU9 T and K. swingsii LMG 22125 T ), but in both cases the NRPS gene contained A, PCP and TE domains, but no apparent C domain, suggesting either that C domains are provided by non-canonical hypothetical genes or that the cluster does not express a functional NRPS product and may be involved in other biosynthesis reactions, or maybe it is the result of translocation or recombination events. Similar NRPS-like genes encoding A, PCP and TE but not C domains were found in Roseomonas vastitatis CPCC 101021 T but its organization and the composition of accessory genes was different. Complete C-A-PCP-TE domains in a single module were also observed in trimodular clusters from Roseomonas frigidaquae JCM 15073 T and Rhodovastum atsumiense DSM 21279 T , in tetramodular clusters from Roseomonas rhizosphaerae YW11 T and Roseomonas rosea DSM 14916 T , in a pentamodular cluster from Roseomonas aestuarii JR169-1-13 T , and in a hexamodular cluster from Roseomonas mucosa NCTC 13291 T . The accompanying modules may provide adenylation or AMT, KR and ECH domains, which are most often found in PKS clusters, or a combination of these. Most of the NRPS clusters from the acidophilic group did not possess the complete minimal set of C-A-PCP-TE domains, and it is unclear if functional peptides are produced by these clusters. It is possible that non-canonical NRPS domains remain undetected by the current algorithm and are hidden in hypothetical accessory proteins. Multimodular NRPS clusters with repetitive C-A and KR domains, respectively, were identified in Acetobacter senegalensis LMG 23690 T and Komagataeibacter rhaeticus LMG 22126 T . Chimeric or hybrid NRPS-PKS clusters with contiguous PKS and NRPS modules were identified in Belnapia rosea CGMCC 1.10758 T , Lichenicoccus roseus KEBCLARHB70R T , Roseomonas algicola PeD5 T , Roseomonas stagni DSM 19981 T , Roseomonas tokyonensis K-20 T , Roseomonas wenyumeia Z23 T and Siccirubricoccus deserti SYSU D8009 T . The hybrid cluster of Roseomonas algicola PeD5 T showed a complex architecture, showcasing fourteen modules, containing two NRPS modules flanked by seven PKS modules and a number of accessory domains. The presence of two genes encoding for different efflux proteins within the BGC, suggest that the produced metabolite is biologically active, and merit exploration. Hybrid clusters present in Asaia bogorensis NBRC 16594 T and Asaia astilbis JCM 15831 T encoded two megasynthases (one PKS and one NRPS) in opposing reading directions. Gene expression in these clusters is probably regulated by a histidine kinase receptor. The NRPS amino acid sequence showed some similarity to vicibactin VbsS from Rhizobium spp. (Heemstra et al. 2009 ), and this megasynthase may similarly catalyze the trimerization of certain amino acid residues. These hybrid clusters are unique among the Acetobacteraceae type strains and they are likely to produce undiscovered bioactive metabolites, which deserve further detailed study. Ribosomally synthetized and post-translationally modified peptides RiPP gene clusters were identified in a handful of acidophilic strains including Acidocella facilis ATCC 35904 T , Rhodovastum atsumiense DSM 21279, Roseomonas algicola PeD5 T , Roseomonas aestuarii JR169-1-13 T and Roseomonas mucosa NCTC 13291 T . In contrast, such clusters were much more prevalent in the acetous group (Fig. 2 b), being present in all Komagataeibacter strains, all Gluconacetobacter strains except Gluconacetobacter johannae LMG 21312 T , in 75% (21/28) of the Acetobacter strains, and in 27% (4/15) of Gluconobacter strains (Fig. 3 ab). Interestingly, no RiPP clusters were found in Asaia or in Saccharibacter-Bombella clades. The only insect-associated AAB type strain carrying a RiPP cluster was Entomobacter blattae G55GP T , which is predicted to produce a yet unknown linear azol(in)e peptide. Roseomonas mucosa NCTC 13291 T was the only species to also carry a BGC encoding a YcaO cyclohydratase, which catalyzes ring formation in azol(in)e peptides. Roseomonas algicola PeD5 T was the only species of the family to be predicted to produce a lasso peptide. Finally, a cyanobactin peptidase gene involved in the final step of RiPP maturation was found in Roseomonas oryzae KCTC 42542 T , and it is likely that this strain produces a new cyanobactin-like peptide. The RiPP clusters found in the acidophilic group (except Roseomonas mucosa NCTC 13291 T ) encoded a DUF692-like protein homologous to MbnB from Methylosinus trichosporium OB3b, which binds iron and forms a complex with MnbC to catalyze the formation of an oxazolone-thioamide group on the core peptide sequence of methanobactin, a copper-chelating molecule (Kenney et al. 2018 ). In those species with a DUF692-like cluster, we did not identify a leader sequence or homologs of MnbC or the TonB receptor. However, this leader-core peptide sequence along with MnbC and TonB homologs were identified in the acetous group. Accordingly, Acetobacter oryzoeni B6 T , Gluconacetobacter asukensis LMG 27724 T , Komagataeibacter nataicola LMG 1536 T , K. rhaeticus LMG 22126 T and K. xylinus LMG 1515 T are likely to produce as yet uncharacterized molecules related to methanobactins. The core DUF692 protein encoded by A. oryzoeni B6 T , K. rhaeticus LMG 22126 T and K. xylinus LMG 1515 T had exactly the same sequence. An unrooted tree based on the DUF692 protein agreed well with the existing phylogeny, clearly distinguishing the acetous and acidophilic groups (Supplementary Fig. 2d), and intriguingly the protein from G. asukensis LMG 27724 T was located in the acidophilic cluster. The RiPP clusters found in acetous species can be assigned to two major groups: the DUF692 cluster also found in the acidophilic species and the linocin M18 cluster. The latter was exclusive to acetous species and was the most common cluster after the hopanoids, being present in 55/96 (~ 57%) of the acetous type strains. This cluster was present in all type strains of the genus Komagataeibacter , in 10/11 (91%) of the Gluconacetobacter and in (21/28) 75% of the Acetobacter species. Gluconacetobacter dulcium LMG 1728 T and Gluconacetobacter tumulisoli LMG 27802 T featured two linocin M18 clusters, and Gluconacetobacter aggeris LMG 27801 T and Gluconacetobacter tumulicola LMG 27725 T shared exactly the same core linocin M18 protein sequence. Intriguingly, the linocin M18 cluster was not found in any Asaia or Neokomagataea species, or in the Bombella-Saccharibacter and Ameyamaea-Tanticharoenia clades, suggesting this pathway is required for certain yet unknown ecological relationships with plants. The unrooted tree based on the linocin M18 protein (Supplementary Fig. 2e) was interesting because there was no clear genus demarcation between Acetobacter , Komagataeibacter and Gluconacetobacter . This suggests either evolution from a common ancestor with independence from constraints operating on core phylogenetic-signal carrying genes, or horizontal gene transfer. Notably, this cluster was not present in any of the insect-associated clades. Because none of the basal acidophilic strains can produce the linocin M18 biosynthetic protein, the ancestor protein in AAB was probably transferred from plant-dwelling members of the family Nitrobacteraceae such as Bradyrhizobium , given the presence of homologs in this genus. A similar linocin M18 cluster has been studied in Rhodococcus jostii RHA1 and was found to encode a DypB peroxidase and an encapsulin protein that together generate a biochemically active lignin degradation nano-compartment (Rahmanpour and Bugg 2013 ). The linocin cluster found in AAB also encoded an encapsulin and a Dyp-type peroxidase, suggesting this cluster is involved in lignin degradation. Siderophores Siderophores are iron-scavenging metabolites that allow producers to thrive in iron-depleted environments. They are particularly useful for microbial competition and are considered virulence factors in pathogenic organisms (Miethke and Marahiel 2007 ). Only NRPS-independent pathways for siderophore biosynthesis (Oves-Costales et al. 2009 ) were identified in the Acetobacteraceae , particularly in the basal phylogenetic clades of the acetous group (5/94) and only in one strain of the acidophilic group, Dankookia rubra JCM 30602 T (Fig. 3 a, b). Two types of siderophore BGCs were identified. One cluster, shared by D. rubra JCM 362 T and Granulibacter bethesdensis CGDNIH1 T , encoded two NRPS-independent siderophore synthases (IucA/IucC-like) (Supplementary Fig. 2f), homologous to proteins from strains of the order Hyphomycrobiales (class Alphaproteobacteria ) such as Methylobacterium , Pseudovibrio and Brucella spp. In addition, the cluster encoded an N -acetyltransferase and a flavin-dependent lysine N -monooxygenase, and the resulting metabolite is probably a yet undescribed siderophore. The second cluster was shared by two Gluconacetobacter species ( G. azotocaptans LMG 21311 T and G. tumulicola LMG 27725 T ) and encoded a single IucA/IucC-like synthase and for a number of proteins of unknown function (Supplementary Fig. 2 g). The siderophore cluster of E. blattae G55GP T is unique in the family Acetobacteraceae and the encoded proteins show homology to proteins from strains of the genera Azotobacter and Pseudomonas , and are distantly related to the clusters for vibrioferrin and xanthoferrin biosynthesis (Pandey et al. 2017 ; Tanabe et al. 2003 ). The siderophore cluster identified in the genome of Endobacter medicaginis LMG 26383 T is unique within the family and includes next to the iucA/iucC marker, a gene encoding for an anthranilate isomerase, a reaction typical of the phenazine biosynthetic pathway. Miscellaneous biosynthetic clusters The acidophilic type strains also encoded biosynthetic proteins for less common specialized metabolites such as phosphonates and indoles, but such clusters were not present in the acetous species. Thirteen strains (27%) in the acidophilic group encoded a homolog of phosphoenolpyruvate mutase and are likely to produce uncharacterized phosphonates. The presence of pyruvate decarboxylase and aminotransferase genes adjacent to the mutase indicated the formation of phosphonoacetaldehyde and finally 2-aminoethylphosphonate, which may be integrated into variety of end-products (Horsman and Zechel 2017 ). Two classes of phosphoryl mutase were identified in the clade, a shorter version present in Rhodovastum atsumiense DSM 21279 T and Roseomonas oryzicola KCTC 22478 T , and the larger and most frequent version in Belnapia rosea CGMCC 1.10758 T and five Roseomonas strains (Supplementary Fig. 2 h). Terpenoid biosynthesis genes were often closely linked to the mutase gene, suggesting that the product is an undiscovered terpene-phosphonate. N -acyl amino acids are synthesized from corresponding amino acid precursors by homologs of the N -acyl amino acid synthase NasY (Craig et al. 2011 ). Interestingly, NasY homologs were found exclusively in three type strains of the genus Acidiphilium , and this biosynthetic property is likely to be a marker of this genus. A putative homolog of PhzB, which catalyzes the synthesis of phenazine, was identified in the genome of R. vastitatis CPCC 101021 T , but no other genes related to phenazine biosynthesis were found in the vicinity. PhzB is a member of the nuclear transport factor 2 (NTF2) family, which may have other functions in bacterial cells (Eberhardt et al. 2013 ), so it is not yet clear whether this strain can produce phenazines."
} | 12,188 |
32127555 | PMC7054358 | pmc | 6,938 | {
"abstract": "Scleractinian “stony” corals are major habitat engineers, whose skeletons form the framework for the highly diverse, yet increasingly threatened, coral reef ecosystem. Fossil coral skeletons also present a rich record that enables paleontological analysis of coral origins, tracing them back to the Triassic (~241 Myr). While numerous invertebrate lineages were eradicated at the last major mass extinction boundary, the Cretaceous-Tertiary/K-T (66 Myr), a number of Scleractinian corals survived. We review this history and assess traits correlated with K-T mass extinction survival. Disaster-related “survival” traits that emerged from our analysis are: (1) deep water residing (>100 m); (2) cosmopolitan distributions, (3) non-symbiotic, (4) solitary or small colonies and (5) bleaching-resistant. We then compared these traits to the traits of modern Scleractinian corals, using to IUCN Red List data, and report that corals with these same survival traits have relatively stable populations, while those lacking them are presently decreasing in abundance and diversity. This shows corals exhibiting a similar dynamic survival response as seen at the last major extinction, the K-T. While these results could be seen as promising, that some corals may survive the Anthropocene extinction, they also highlight how our relatively-fragile Primate order does not possess analogous “survival” characteristics, nor have a record of mass extinction survival as some corals are capable.",
"introduction": "Introduction Scleractinian corals represent an ideal taxon to serve as a model for describing past and predicting future environmental trajectories of mass extinctions. Coral skeletons are widespread and well-preserved in the fossil record, and as reef-builders, Scleractinians support and promote biodiversity hotspots. While coral reefs represent only 0.2% of the oceans’ area, they harbor ~95,000 described species and represent about 5% of the world’s known species and ~35% of known marine species 1 . For these reasons they have been extensively studied and relatively well-monitored over the past few decades 2 , 3 . The International Union for Conservation of Nature (IUCN) Red List has recently reported an alarming trend, ca. 30% of the 96,500 species assessed on the IUCN Red List being threatened with extinction 4 , including one third of the world’s corals 5 . The earth has experienced five major mass extinction events (“The Big Five”) since the Cambrian period, each resulting in the loss of more than three-quarters of species biodiversity over a geologically short time interval. These events occurred at the 1) Late Ordovician (440 Mya), 2) Late Devonian (370–350 Mya), 3) End-Permian (251 Mya), 4) End-Triassic (201 Mya) and 5) End-Cretaceous, which is referred to as the ‘K-T’ event (66 Mya). It has been proposed that the sixth mass extinction is currently in progress 6 with increasing number of studies confirming that current extinction rates are more than 100 times higher than background extinction rates 7 . This rate of species loss is on par with previous extinction events 6 , 8 – 10 . This study tests the hypothesis of common evolutionary traits/dynamics that characterize both K-T and the current Anthropocene survivors. We use Scleractinian corals as the test subject due to their secretion of a calcium carbonate skeleton that provides a rich history for paleontological analysis that extends back to the Triassic period 11 , 12 . The End-Cretaceous (K-T) mass extinction The K–T mass extinction was one of the most destructive events in the Phanerozoic, resulting in global extinction of ~40% of total genera and 47% of marine invertebrate genera 13 . It is widely accepted that this extinction was triggered by a giant (10 km in diameter) meteorite impact at Chicxulub, Yucatan Peninsula, Mexico 66 million years ago 14 . The bolide event likely caused earthquakes, tsunamis and intense heat pulse that led to global wildfires, which had a strong impact on non-marine organisms 15 . The initial “fireball stage” was followed by a global “impact winter” caused by dust particles and other aerosols blocking sunlight, resulting in a 6 °C cooling of sea surface temperature. The impact winter lasted months to decades and influenced both marine and terrestrial productivity, leading to nutrient soup accumulations, and the formation of stable cold deep water 16 . In addition, sulfur and nitrogen volatile compounds were injected to the atmosphere, possibly leading to acid rain and ocean acidification 17 . The acid rain increased chemical weathering rates and resulted in a higher flux of bioavailable phosphorus into the ocean, resulting in additional nutrients to the nutrient soup 18 . The K-T boundary catastrophe reached deep-water dwellers, inferred by significant deterioration of deep-water benthic foraminifera communities 19 . After the relatively short-lived cool impact winter, rapid warming brought sea surface temperature (SST) back to pre-K-T levels followed by long-term warming, apparently related to greenhouse gases release 20 . Some other scenarios for the K-T event drivers include volcanism, multiple asteroid impacts, climatic changes and biotic stresses already affecting organisms previous to the K-T, as well as combinations of these factors 9 . Recent modeling experiments predict that the “impact winter” might have been balanced, or even outweighed, by global warming derived by water vapors, leading to a greenhouse effect 21 . Scleractinian corals are expected to be particularly sensitive to bespoke environmental changes such as: (1) prolonged darkness, (2) major temperature changes, (3) eutrophication and (4) ocean acidification. The recovery of coral reefs following the K-T event began with coralline algae, sometimes accompanied by photosymbiotic benthic foraminifera. While reef-building photosymbiotic corals suffered great losses at the End-Cretaceous, some genera survived and later became important reef builders in the Cenozoic 22 . Coral reefs were re-established ~2–5 Ma after the K-T, and became increasingly abundant in the Eocene (~10 Ma later). These reefs were composed of mostly novel community types, compared to previous Cretaceous reefs 22 . The Anthropocene extinction Homo sapiens first appear in the fossil record ~315,000 years ago 23 and since that time there have been several regional extinctions, becoming more frequent following the agricultural/neolithic revolution beginning ~10,000 years ago 9 . However, since the Industrial Revolution (1760–1840) and particularly after 1950, the Earth’s biosphere has experienced rapid changes in atmospheric conditions, warming of global temperatures and rising CO 2 levels. These are the highest levels over the past 800,000 years and temperatures are predicted to increase by ~1.8–3.4 °C and CO 2 levels by 170–420 ppm over the next century 24 , 25 . Increased atmospheric CO 2 concentrations are causing measurable ocean acidification 26 . This together with habitat fragmentation, pollution, overfishing and overhunting, the introduction of invasive species and pathogens, and expanding human biomass represents a combination of global wide extreme stressors combining in unprecedented ways for most species 6 , 27 . If these stressors are not mitigated, 75% of species are predicted to face extinction within the next few centuries to millennia 6 . Stony corals are particularly sensitive and are heavily impacted by direct human interferences (e.g. overfishing, pollution, coastal development and tourism damage), as well as warming and ocean acidification. They are predicted to suffer more than 30–50% extinction in the coming decades and century, respectively 5 , 24 . Current extinction risk assessment Realizing that extinction risk is realistic to increasing number of species in our era, the International Union for Conservation of Nature (IUCN) initiated the Red List of Threatened Species in 1964. It aimed to classify organisms at extinction risk in categories that may assist in managing conservation efforts efficiently. The Red List is recognized as the most comprehensive assessment of organisms’ extinction risk 28 with species assigned categories, from “extinct” to “of least concern”. This assignment is performed by taxonomic experts according to the IUCN Red List categories and criteria 29 that consider reports of population reduction, as well as threats and risk factors such as bleaching/disease susceptibility, severe fragmentation and destruction of habitat. In this study, we used data from the coral fossil record and the current IUCN Red List classification to compare coral survival between the K-T and the current extinction events.",
"discussion": "Discussion Deducing future trajectories, based upon past extinction events, has received increased scientific attention over the past decade 41 – 44 . Understanding the effects of biological traits on coral survival and future community structure is a matter of high priority 41 , 45 , both for what it predicts for the future of coral reefs and as a general model for the selective survival of various biological communities. In this study we searched the existing body of coral fossil data and coral traits through the K-T boundary event, together with their contemporary extinction status, to determine whether there are “survival traits” that are common both among survivors of the last major mass extinction (K-T). and at the present Anthropocene mass extinction. We found wide geographical distribution to be the most important determinant of coral survival both through the K-T and the Anthropocene (Table 2 , Fig. 2 ). This finding may reflect a geographic heterogeneity of these extinction events elevating survival chances for widespread genera. Wide geographical distribution was also noted as the only trait significantly characterizing bivalves survival during mass extinctions 46 , 47 . However, this result should be taken with caution regarding the Anthropocene extinction as it is based on analyzes of the IUCN Red List categories whose assessment might also take geographical distributions parameters into account. Our analysis shows that colonial, symbiotic, shallow Scleractinian genera were particularly impacted by the K-T extinction, a finding that is in agreement with 40 . We also show that these genera evolved in higher proportions following the extinction (Fig. 1 ) and are now the main contemporary tropical coral reef builders. However, this origination through the Tertiary is currently in jeopardy, with many colonial, symbiotic genera falling within the modern IUCN Red List threatened-extinction categories (Fig. 2 ) and community shifts are taking place over large reef areas that favor slow growing corals 34 . This transition may have previously occurred 48 , as it’s suggested that photosymbiotic colonial corals were disproportionately removed during the Triassic extinction 45 , the mid Cretaceous 40 and the Oligocene-Miocene extinction 49 . The sensitivity of the closely-related traits (colonial, symbiotic and shallow-dwelling) 39 to K-T and Anthropocene extinctions may be explained by the combined effect of eutrophication, warming and acidification which characterize both the modern situation and some parts of the K-T event 50 . These environmental changes in the modern ocean impact mostly euphotic surface waters and are particularly threatening to colonial symbiotic genera that reside in the upper, photic ocean, where they outcompete their algae competitors in low-nutrients environments. An alternative hypothesis is that Scleractinian corals retreated into deeper, off-reef niches where light was scarce 40 . In the deep reef environment, non-photosymbiotic solitary coral genera have an advantage, while symbiotic corals suffered higher extinction rates on shallow reef habitats during the K-T 40 . Decreasing average growth rates through the K-T event may also be related to the drop of photosymbiosis which would diminish growth rates due to lesser energy and photosynthates supply 51 . Decreasing growth rates may also reflect the coinciding high bleaching resistance found during the K-T (Fig. 1 ) since many fast-growing corals are thin-tissue branched colonies that are suggested to be more susceptible to bleaching 33 . Our analysis reports an evolutionary selection towards slower growing genera, rather than merely a decrease in the growth rates of a genus, as documented in recent ecological studies e.g 52 , 53 . Therefore, while fast-growing genera have higher extinction risk, it might also be due to other accompanying traits (coloniality, symbiosis, shallow habitat occupation and branching forms). In addition, coral genera with relatively greater disease or thermal stress resistance, were relatively dominant (~40%, ~30% respectively) in those that survived the K-T event and Paleocene and Eocene high temperatures (Fig. 1 ). Decreasing temperatures were later accompanied by the evolution of genera with higher proportion of vulnerable traits leading to a lower proportion (~20%) of resistant genera. This relative decrease of resistant genera during low SST periods may suggest that high SST may be a cause of increased mortality by disease and bleaching on a geologic time scale, such as following the K-T event. This finding is in accordance with 54 that shows higher disease effects coinciding with past thermal stress history. The high proportion of bleaching resistant genera after the K-T event reinforces the suggestion that bleaching resistance is a significant survival factor during mass extinctions 39 , 45 , 55 . This finding offers another perspective to the “adaptive bleaching hypothesis” 56 , which states that when environmental conditions change, one or more clades of photosymbionts is replaced by a new symbiotic consortium of photosymbionts that are better suited to the current conditions. Interestingly, for the Anthropocene event, bleaching susceptibility was found to be in the highest proportion (91%) within “Vulnerable”, rather than in the “Critical” (75%) Red List category. This could be interpreted as the bleaching threat being not yet fully established, or that other factors are stronger in determining probable extinction patterns. A recent model for coral endurance through global warming predicts that bleaching resistance will become a dominant factor for future coral survival 41 . There is potentially a difference in corals surviving moderate extinction (by possessing large colony size) and mass extinctions (by possessing small colony size). Ecological characteristics were previously found to be an important determinant of extinction risk in Caribbean reef-corals in the Neogene with colony size as the most important trait determining extinction rate. It was found that for the Neogene moderate extinctions, corals with small, massive colonies were most vulnerable 57 and large colonial corals survived. A similar trend, of higher extinction risk for smaller organisms during “background extinctions” as opposed to higher extinction risk for large bodied creatures in the Anthropocene, was found for marine vertebrates and mollusks 58 . These discrepancies imply different mechanisms allowing a coral to survive during moderate versus mass extinction. In moderate extinction conditions, large colony size (associated with a higher number of polyps per colony, fragmentation capabilities and high rates of recolonization after local extinctions) allows a coral to survive some disturbances. While in mass extinction conditions, such as the K-T (this study), small colony size allows corals in “sheltered” niches to survive. This tendency of smaller coral genera to survive through a mass extinction resembles the well-known “Lilliput Effect”, a trend towards smaller size of faunal elements associated with mass extinction events 59 . In this sense, the fate of large corals is similar to that of large terrestrial vertebrates, who were drastically impacted by the K-T mass extinction event 60 . While the conditions of the K-T mass extinction event were likely different from conditions of the current Anthropocene extinction, this study notes distinctive similarities between coral traits that survived the K-T mass extinction event and those that are least threatened in the current extinction event. This leads to the hypothesis that coral genera that evolved at or before the K-T event (“K-T corals” Fig. 5 ) may be better adapted to survive the modern extinction than genera that evolved afterwards (i.e. “Modern corals” Fig. 5 ). This holds true when analyzing the presence of “modern” and “K-T” genera along the IUCN Red List ( https://www.iucnredlist.org ) categories (Fig. 5 ). This analysis shows that “K-T corals” are less threatened by modern extinction, than corals that evolved after the Cretaceous period. Figure 5 Current “extinction status” of Pre and Post K-T coral genera (grey and black bars respectively). Pre K-T corals (N = 16 genera) are those found in the fossil record before 66 Mya, Post K-T corals (N = 59 genera) only emerged later than 66 Mya. Inset, the percentage of pre and post K-T threatened genera (Red List category higher than Near Threatened). Red List category for a genus was defined as the category of the least threatened species within the genus. We suggest that the similarity in traits allowing coral survival exists through two very different extinction events (Table 2 ) imply that there is a basic mechanism which enables a coral to become a “mass extinction survivor” and which may be maintained for tens of millions of years. Recognizing this, we might predict a long-term change in coral assemblages to more solitary, non-symbiotic communities, similar to early Paleocene assemblages. In this scenario, already evidenced in some of the world’s largest coral reef systems 34 , the dominance of corals in contemporary coral reefs may be replaced by algae or other invertebrates’ dominance, due to the loss of competitive advantage in low nutrient exploitation and structure building. A community shift in this direction will damage their reef-building capacity and structural complexity that supports the highest biodiversity among all marine ecosystems. This study provides alarming evidence that reef communities are currently in the process of transitioning into disaster communities, akin to previous extinction events. Recovery of coral reef ecosystems from the K-T mass extinction was slow (2–10 My), as judged by the Paleocene re-establishment of algal symbiosis 39 . While the slow recovery time of coral reefs following a mass extinction is distressing, we also call attention that Primates (the Order that also includes humans) are also increasingly becoming threatened with extinction 61 . And unlike the Order Scleractinia, Primates do not possess analogous “survival” traits that enable some species to transcend major extinction boundaries, nor does Homo sapiens or any other Primate species, have a track record of mass extinction survival."
} | 4,760 |
28526061 | PMC5438491 | pmc | 6,939 | {
"abstract": "Acaryochloris marina is an oxygenic cyanobacterium that utilizes far-red light for photosynthesis. It has an expanded genome, which helps in its adaptability to the environment, where it can survive on low energy photons. Its major light absorbing pigment is chlorophyll d and it has α-carotene as a major carotenoid. Light harvesting antenna includes the external phycobilin binding proteins, which are hexameric rods made of phycocyanin and allophycocyanins, while the small integral membrane bound chlorophyll binding proteins are also present. There is specific chlorophyll a molecule in both the reaction center of Photosystem I (PSI) and PSII, but majority of the reaction center consists of chlorophyll d . The composition of the PSII reaction center is debatable especially the role and position of chlorophyll a in it. Here we discuss the photosystems of this bacterium and its related biology.",
"conclusion": "Conclusion Photosynthesis in A. marina is unique as it utilizes photons from the far red light part of the spectrum which are low energetically but they are abundant in its habitat where no other photosynthetic organism absorbs such wavelength of light. This long wavelength absorption has been confirmed from UV–visible spectroscopy, EPR, ENDOR, FTIR and other instrumental techniques and growth tests. They required less amount of energy for excitation of their reaction center and so the amount of energy required during electron transfer from oxidation of water to reduce NADP + is smaller. The transfer of energy from extrinsic antenna to membrane bound intrinsic antenna containing chlorophyll d , is very efficient and three times faster than other oxygenic cyanobacteria. Although its genome is completely sequenced, but from its expanded genome, the role of different duplicated genes is still to be answered. The position and role of chlorophyll a in photosystem II needs to be discovered. In order to fully understand such unique type of photosynthesis there is a need for further spectroscopic and structural based studies of its photosynthetic membrane proteins."
} | 522 |
31091672 | PMC6631521 | pmc | 6,940 | {
"abstract": "An enormous number of bacteria live in almost every environment; from deep oceans to below the surface of the earth or in our gastrointestinal tract. Although biofabrication is growing and maturing very quickly, the involvement of bacteria in this process has not been developed at a similar pace. From the development of a new generation of biomaterials to green bioremediation for the removal of hazardous environmental pollutants or to develop innovative food products in a recent trend, researchers have used cutting-edge biofabrication techniques to reveal the great potential of 3D structured bacterial constructs. These 3D bacterial workhouses may fundamentally change our approach toward biomaterials."
} | 177 |
22895707 | null | s2 | 6,941 | {
"abstract": "Recent behavioral studies have given rise to two contrasting models for limited working memory capacity: a \"discrete-slot\" model in which memory items are stored in a limited number of slots, and a \"shared-resource\" model in which the neural representation of items is distributed across a limited pool of resources. To elucidate the underlying neural processes, we investigated a continuous network model for working memory of an analog feature. Our model network fundamentally operates with a shared resource mechanism, and stimuli in cue arrays are encoded by a distributed neural population. On the other hand, the network dynamics and performance are also consistent with the discrete-slot model, because multiple objects are maintained by distinct localized population persistent activity patterns (bump attractors). We identified two phenomena of recurrent circuit dynamics that give rise to limited working memory capacity. As the working memory load increases, a localized persistent activity bump may either fade out (so the memory of the corresponding item is lost) or merge with another nearby bump (hence the resolution of mnemonic representation for the merged items becomes blurred). We identified specific dependences of these two phenomena on the strength and tuning of recurrent synaptic excitation, as well as network normalization: the overall population activity is invariant to set size and delay duration; therefore, a constant neural resource is shared by and dynamically allocated to the memorized items. We demonstrate that the model reproduces salient observations predicted by both discrete-slot and shared-resource models, and propose testable predictions of the merging phenomenon."
} | 427 |
21749486 | null | s2 | 6,944 | {
"abstract": "Hydrogen (H₂) production by Thermococcus kodakarensis compares very favourably with the levels reported for the most productive algal, fungal and bacterial systems. T. kodakarensis can also consume H₂ and is predicted to use several alternative pathways to recycle reduced cofactors, some of which may compete with H₂ production for reductant disposal. To explore the reductant flux and possible competition for H₂ production in vivo, T. kodakarensis TS517 was mutated to precisely delete each of the alternative pathways of reductant disposal, H₂ production and consumption. The results obtained establish that H₂ is generated predominantly by the membrane-bound hydrogenase complex (Mbh), confirm the essential role of the SurR (TK1086p) regulator in vivo, delineate the roles of sulfur (S°) regulon proteins and demonstrate that preventing H₂ consumption results in a substantial net increase in H₂ production. Constitutive expression of TK1086 (surR) from a replicative plasmid restored the ability of T. kodakarensis TS1101 (ΔTK1086) to grow in the absence of S° and stimulated H₂ production, revealing a second mechanism to increase H₂ production. Transformation of T. kodakarensis TS1101 with plasmids that express SurR variants constructed to direct the constitutive synthesis of the Mbh complex and prevent expression of the S° regulon was only possible in the absence of S° and, under these conditions, the transformants exhibited wild-type growth and H₂ production. With S° present, they grew slower but synthesized more H₂ per unit biomass than T. kodakarensis TS517."
} | 394 |
21079766 | PMC2974637 | pmc | 6,946 | {
"abstract": "The gabbroic layer comprises the majority of ocean crust. Opportunities to sample this expansive crustal environment are rare because of the technological demands of deep ocean drilling; thus, gabbroic microbial communities have not yet been studied. During the Integrated Ocean Drilling Program Expeditions 304 and 305, igneous rock samples were collected from 0.45-1391.01 meters below seafloor at Hole 1309D, located on the Atlantis Massif (30 °N, 42 °W). Microbial diversity in the rocks was analyzed by denaturing gradient gel electrophoresis and sequencing (Expedition 304), and terminal restriction fragment length polymorphism, cloning and sequencing, and functional gene microarray analysis (Expedition 305). The gabbroic microbial community was relatively depauperate, consisting of a low diversity of proteobacterial lineages closely related to Bacteria from hydrocarbon-dominated environments and to known hydrocarbon degraders, and there was little evidence of Archaea. Functional gene diversity in the gabbroic samples was analyzed with a microarray for metabolic genes (“GeoChip”), producing further evidence of genomic potential for hydrocarbon degradation - genes for aerobic methane and toluene oxidation. Genes coding for anaerobic respirations, such as nitrate reduction, sulfate reduction, and metal reduction, as well as genes for carbon fixation, nitrogen fixation, and ammonium-oxidation, were also present. Our results suggest that the gabbroic layer hosts a microbial community that can degrade hydrocarbons and fix carbon and nitrogen, and has the potential to employ a diversity of non-oxygen electron acceptors. This rare glimpse of the gabbroic ecosystem provides further support for the recent finding of hydrocarbons in deep ocean gabbro from Hole 1309D. It has been hypothesized that these hydrocarbons might originate abiotically from serpentinization reactions that are occurring deep in the Earth's crust, raising the possibility that the lithic microbial community reported here might utilize carbon sources produced independently of the surface biosphere.",
"conclusion": "Conclusion Our results raise the intriguing possibility that hydrocarbons in very deep ocean rocks support microbial communities. Additionally, we show that the genetic potential for novel metabolic processes, such as carbon and nitrogen fixation, is present within an unexplored layer of ocean crust. Our findings, particularly regarding the presence of genes coding for methane cycling, have implications not only for Earth's subsurface, but also for other planets such as Mars. Methane on Mars is concentrated in some equatorial regions of the atmosphere, which suggests that it is derived from localized geological sources [54] . Although the exact mechanism by which methane forms on Mars is not known, serpentinization reactions in the Martian subsurface have recently been proposed [55] . Therefore, similar to the Atlantis Massif, the Martian subsurface may harbor methane-consuming prokaryotes. Future efforts should be directed towards quantifying the role endolithic prokaryotes play in methane cycling and in determining the sources of methane, and other hydrocarbons in marine crust. These findings will undoubtedly focus attention on obtaining more information on the geochemistry of formation fluids from deep ocean rocks, which are technically challenging to acquire requiring different sampling technologies than those used in the design of this exploratory study.",
"introduction": "Introduction Ocean crust covers nearly 70% of the earth's surface, with an estimated volume of 10 18 cubic meters. Microbial processes in this expansive subseafloor environment have the potential to significantly influence the biogeochemistry of the ocean and the atmosphere [1] . Recently, Delacour et al. [2] analyzed rock samples from the Atlantis Massif and reported that biomarkers were present in the gabbroic central dome (IODP Hole 1309D; this is the same drill hole we present an analysis of here) and in rocks from the Lost City Hydrothermal Field (LCHF). Delacour et al. [2] also determined that hydrocarbons were present in these basement rocks. These authors suggested that these hydrocarbons account for an important fraction of the carbon stored in the basement rocks of the Atlantis Massif. Hydrocarbons at the Atlantis Massif are the subject of a recent report by Proskurowski et al. [3] , who found that methane and other low-molecular-weight carbon compounds, which are abundant at the LCHF, appear to have formed abiotically from serpentinization reactions in olivine- and pyroxene-rich igneous rocks (peridotite). This water rock reaction evolves hydrogen [4] , [5] and higher alkanes [4] . Congruent with the geochemical conditions at the LCHF Schrenk et al. [6] found a low diversity of predominantly methanogenic and/or methanotrophic Archaea in LCHF chimneys. Brazelton et al. [7] reported that LCHF carbonates and fluids are dominated by methane- and sulfur-metabolizing communities. Together, these studies suggest that LCHF carbonates host a microflora that likely utilize the rock-seawater derived electron donors/carbon sources. Thus, the precedent for hydrocarbon utilizing microbes at the Atlantis Massif has been set. Beyond LCHF carbonates, much attention has been directed towards the basalt layer of ocean crust. Recent reports on the diversity of microbial life in marine basalts revealed that upwards of 13 or more clades of Bacteria [8] – [15] and two clades of Archaea [8] , [9] , [11] , [12] , [15] are present in this environment. Yet, little is known about the metabolic processes occurring in this environment, with only one report by Mason et al. [11] assaying for functional status of basalt microflora. Further, all but two of these studies [8] , [10] were conducted on surface basalts. Thus, even in the frequently studied basalt layer little is known about subsurface endolithic microorganisms. Our collective knowledge about endolithic microorganisms associated with igneous rocks in the marine environment stems from the aforementioned studies. To date, however, the microbiology of the intermediate layer between basalt and peridotite - the gabbro layer- has not been investigated, mostly due to the difficulty inherent in sampling the igneous portion of ocean crust, a topic that was recently reviewed by Schrenk et al. [1] . The Atlantis Massif, which is interpreted as an ocean core complex composed of deep crustal (gabbro) and upper mantle rocks (peridotite) that have been unroofed and exposed at the surface as a result of faulting [16 and references therein] , [ 17] , provided a rare opportunity to sample gabbros, which are generally beyond the reach of currently available drilling technologies. The goals of our study were to measure the cell density, phylogenetic diversity and metabolic diversity of endolithic microflora associated with the central dome of the Atlantis Massif. To accomplish our goals we used microscopy to determine in situ cell densities. Terminal restriction fragment polymorphism (T-RFLP), denaturing gradient gel electrophoresis (DGGE), cloning, and sequencing were used to assess the diversity and phylogeny of microorganisms associated with marine gabbros. Further, to provide insight into the potential metabolic diversity of the gabbroic crust microflora, we analyzed conserved regions of functional genes involved in nitrogen, carbon, sulfur, and phosphorus cycling with GeoChip, a functional gene microarray [18] .",
"discussion": "Discussion Cell density Cells counts from the rock samples recovered by drilling were below the limit of detection (<10 3 cells cm −3 rock). Cell densities much lower than those reported for basalts and LCHF carbonates and crustal fluids were detected in sediment sampled next to our sample site, providing support for our fining that cell densities within our subsurface rock samples were extremely low (below the level of detection). For example, the cell density in carbonate sediment sampled in neighboring Hole U1309A collected during Expedition 304 ( Table 1 ) was 1.15±0.95×10 4 cells cm −3 , far below values reported for basalts - up to 10 6 \n [34] , and 10 9 cells per g rock [14] . Carbonates from LCHF host up to 10 8 cells per g dry weight [6] and 10 9 cells per g rock [7] . Microbial diversity Recently, Orcutt et al. [35] used T-RFLP to evaluate two basalts that were identified by Santelli et al. [14] as low and high diversity samples from Loihi Seamount. Orcutt et al. [35] reported Shannon values (H') of 1.81 (low diversity sample) and 2.55 (high diversity sample). In comparison to these basalt samples Expedition 305 rocks appear to harbor a lower diversity of microorganisms, with an average Shannon value of 1.37, than the least diverse basalt sample. Interestingly, the depth variance in microbial diversity in Expedition 305 rock samples was related to percent rock alteration ( Figure S1 ). Samples with higher percent alteration had a greater diversity of microorganisms ( Figure S1 ). Alteration could result in, for example, changing rock permeability and oxidation state of the rocks, which may affect the microbial communities, by providing additional niches in gabbroic rocks. This potential niche expansion may be reflected by higher microbial diversity in the more altered rocks. Phylogeny Perhaps surprisingly, the gabbro microbial communities we observed do not appear to be endemic to ocean crust. Marine basalts and gabbros are nearly identical in chemical composition, thus we had hypothesized that similar communities of endolithic microorganisms specialized for growth in subsurface igneous rocks would be recovered from gabbros. Our analysis of gabbroic rocks revealed that there was no overlap in the microbial communities between these two rock types - none of the ocean crust clades that appear to be endemic to basalts that were delineated by Mason et al. [12] were found in gabbros. The absence of clades determined to be endemic to ocean crust within marine gabbros in conjunction with the widespread distribution in the environment (e.g. water, soil, and activated sludge) of cultured representatives that are closely related to our clones suggests that gabbroic microflora are not specifically adapted to life in ocean crust. Second, one would expect that microorganisms endemic to ocean crust (e.g. gabbro) would have 16S rDNA sequences that have diverged from related microbes that are not endemic to this habitat, inclusive of cultured and uncultured representatives from non-crustal environments. This was observed in Mason et al. [12] , in which several clades that appear to be endemic to basalt were not closely related to any cultured microorganisms, or to microbes found in seawater, or other non-crustal environments. The close relationship of gabbroic microorganisms to cultured hydrocarbon degrading Bacteria that are widely distributed in the environment suggests that gabbroic microflora are transient microorganisms that are not endemic to ocean crust. The high similarity of gabbroic microflora to both cultured and uncultured representatives previously reported in hydrocarbon dominated environments suggests instead that gabbroic microflora are not ocean crust specialists, such as those observed in basalts, but rather predominate in environments where hydrocarbons are present. The high similarity of 16S rRNA gene sequences of gabbroic endoliths to cultured representatives was surprising, yet not unprecedented in other hydrocarbon dominated environments, such as gas hydrates [19] and petroleum reservoirs [20] (sequences from Expedition 305 rocks were highly similar to clones from both of these studies). In fact, Lanoil et al. [19] noted the high similarity of the Bacteria (ca. 72%) in a gas hydrate sample that were related at the species level to cultured microorganisms as an unusual feature of the bacterial diversity in this hydrocarbon dominated environment. Similar to the results presented by Lanoil et al. [19] the Bacteria observed in high temperature petroleum reservoirs were 97–100% similar to previously cultured representatives [see Table 2 in 20]. More recently clones from crustal fluids [26] and from LCHF (∼5 km from our study site) carbonate chimneys and from vent fluids [7] were 100% similar to Ralstonia pickettii and to clones from rock sample 273 (1313.06 mbsf). Although the high similarity of microorganisms to previously cultured hydrocarbon degrading bacteria does not necessarily mean that the in situ gabbroic community shares the same genetic potential as their close relatives, the data does suggest that within disparate hydrocarbon dominated environments certain bacterial taxa are generalists, able to survive and to potentially degrade hydrocarbons in a myriad of environments, including deep subsurface igneous rocks, such as those analyzed in this study. Biogeochemical cycling In congruence with the close phylogenetic relationships of rock associated microorganisms to those from hydrocarbon-dominated environments and to known hydrocarbon degrading Bacteria, genes coding for hydrocarbon degradation were observed in both rock samples. For example, genes coding for aerobic methane-oxidation ( pmo and mmo ) were observed in rock samples 90 and 142 ( Table S1 ). Methylobacterium populi , closely related to clones in rock sample 142 (95% similar), was shown to grow on methane as a sole carbon and energy source [23] . Additionally, genes coding for aerobic toluene oxidation were present in rock sample 142 ( Table S1 ). Specifically, genes coding for toluene oxidation from Pseudomonas mendicina were present ( Table S1 ). Clones in rock sample 142 that were most closely related to Pseudomonas fluorescens , which is able to grow anaerobically on toluene [36] , were also highly similar to P. mendicina (96% similar). Taken together phylogenetic and functional gene analyses converge to suggest that hydrocarbon oxidation may be occurring in deep subsurface ocean crust. Beyond hydrocarbon oxidation, many other functional genes involved in carbon cycling were present ( Table S1 ). For example, genes coding for carbon fixation ( acsA , FTHFS, rbcL, rbcS) were present in both rock samples. Delacour et al. [2] reported that total organic carbon (TOC) ranged from 53–1015 ppm. TOC concentrations in rocks above and below sample 90 (448.90 mbsf) are lower than the TOC concentrations in rocks near sample 142 (701.05 mbsf) [see Table 1 in 2]. Interestingly, the relative abundance of genes coding for carbon fixation were slightly lower in rock sample 142 (2.4%) than in sample 90 (3.2%). This suggests that in local environments within ocean crust where organic carbon is low relative to other sections of the crust there is a genetic potential, in the form of genes coding for carbon fixation, to offset lower TOC concentrations by an increase in carbon fixation. \n Mcr genes coding for methane production were identified in both rock samples ( Table S1 ). Interestingly, the alpha subunit ( mcrA ) of the methyl conenzyme M reductase (MCR) from an anaerobic methane oxidizing archaea (ANME) was present in sample 142 [37] ( Table S1 ). Although mcr genes were present in our rock samples, no Archaea were observed in either rock or seawater samples despite numerous attempts to amplify archaeal 16S rRNA genes (see Methods ); therefore, it is unlikely that Archaea play a significant role in biogeochemical cycling in the marine gabbros analyzed here. Genes coding for denitrifying processes (e.g. nar G, nir K, nor B) were detected in both rock samples ( Table S1 ). Although the majority of characterized hydrocarbon-degrading microorganisms previously discussed are aerobic, both R. picketti \n [38] and P. fluorscens \n [36] have been shown to oxidize hydrocarbons by denitrification. This suggests that hydrocarbons be may oxidized anaerobically in the central dome of the Atlantis Massif. Analysis of metagabbros revealed that nitrogen concentrations are low (4.0 to 4.5 ppm) [39] . Thus nitrogen fixation in this environment would be paramount. Nitrogen-fixation in the marine subsurface was recently reported in hydrocarbon dominated seep sediments [40] . Further Mason et al. [11] reported that genes coding for nitrogen-fixation were present in a Juan de Fuca basalt. The presence of genes coding for nitrogen fixation in our gabbroic samples ( Table S1 ), in conjunction with the findings of Mason et al. [11] and Dekas et al. [40] , suggests that this process may be widespread in the marine subsurface. Previously unrecognized sites for nitrogen-fixation in the marine environment, such as subsurface rocks, may provide insight into the missing nitrogen sources in the ocean as presented by Deutsch et al. [41] . Other genes that were observed code for dissimilatory sulfate reduction ( dsrA and dsrB ) ( Table S1 ), largely from uncultured sulfate-reducing bacteria, which may suggest that hydrocarbons are degraded anaerobically by sulfate reducers [42] . Further, genes coding for cytochromes from, for example, Geobacter sulfurreducens, a metal- and sulfur-reducing Bacteria isolated from a hydrocarbon contaminated environment [43] were detected in both rock samples ( Table S1 ). The presence of genes coding for both aerobic and anaerobic respiration in the upper 700 meters of Hole 1309D are consistent with the redox conditions suggested by Delacour et al. [44] who reported that strontium and sulfur isotopes are elevated towards seawater values in the upper 800 m in 1309D. These isotopic values indicate that seawater has circulated in the upper portion of the central dome [45] and are correlated with a greater degree of serpentinization [2] . Seawater circulating within the top 800 m would provide a limited amount of oxygen that is required for aerobic processes, with a transition to anaerobic processes following oxygen depletion. Below 800 m these authors suggested that reducing conditions prevail and that seawater circulation is constrained to faults within the central dome. \n In situ hydrocarbons Our analysis of the isotopic signature recorded in carbonate in Hole 1309D rocks revealed δ 13 C carbonates averaged −5.0‰. Mantle carbon δ 13 C ranges from approximately −5.0 to −7.0‰ [2] , [46] , with the majority of our samples falling in this range ( Figure 2 ). Delacour et al. [2] reported that n -alkanes ranging from C 15 to C 40 (volatiles could not be measured) were present in rocks from the central dome. These alkanes were unbranched, with no carbon number predominance, and showed a decrease in abundance with increasing carbon number [2] . This profile is similar to carbon compounds synthesized abiotically by Fischer-Tropsch type (FTT) reactions [47] , such as at those at the LCHF [3] . Abiotic production of the unbranched alkanes (which would include methane) in Atlantis Massif samples is suggested. These hydrocarbons could provide carbon and energy to extant microbes in the interior of the Atlantis Massif. In 1309D rocks Delacour et al. [2] also identified the biomarkers squalane, hopane, sterane, pristane, and phytane. These alkanes were attributed to DOC input from seawater circulating throughout the Atlantis Massif [2] . The source of these alkanes may reflect input of marine DOC as these authors suggest, alternatively squalene (diagenetically transformed to squalane [2] ) has been identified in methanotrophs [48] and hopanes are found in a variety of prokaryotes [49] , including methanotrophs [50] . Steranes are ubiquitous in eukaryotes [51] , and although rare in prokaryotes, have been reported in a few microorganisms such as methanotrophs in the Methylococcales \n [48] , [52] . Pristane and phytane could originate from methanogenic Archaea [53] . The δ 13 C values of individual biomarkers were not determined in Atlantis Massif samples; therefore, their exact origin is not known but the results of our molecular analyses indicates that in situ microorganisms, and in particular methanotrophs, are the sources of biomarkers in 1309D. Conclusion Our results raise the intriguing possibility that hydrocarbons in very deep ocean rocks support microbial communities. Additionally, we show that the genetic potential for novel metabolic processes, such as carbon and nitrogen fixation, is present within an unexplored layer of ocean crust. Our findings, particularly regarding the presence of genes coding for methane cycling, have implications not only for Earth's subsurface, but also for other planets such as Mars. Methane on Mars is concentrated in some equatorial regions of the atmosphere, which suggests that it is derived from localized geological sources [54] . Although the exact mechanism by which methane forms on Mars is not known, serpentinization reactions in the Martian subsurface have recently been proposed [55] . Therefore, similar to the Atlantis Massif, the Martian subsurface may harbor methane-consuming prokaryotes. Future efforts should be directed towards quantifying the role endolithic prokaryotes play in methane cycling and in determining the sources of methane, and other hydrocarbons in marine crust. These findings will undoubtedly focus attention on obtaining more information on the geochemistry of formation fluids from deep ocean rocks, which are technically challenging to acquire requiring different sampling technologies than those used in the design of this exploratory study."
} | 5,386 |
26695630 | PMC4701829 | pmc | 6,948 | {
"abstract": "ABSTRACT Marine methane seeps are globally distributed geologic features in which reduced fluids, including methane, are advected upward from the subsurface. As a result of alkalinity generation during sulfate-coupled methane oxidation, authigenic carbonates form slabs, nodules, and extensive pavements. These carbonates shape the landscape within methane seeps, persist long after methane flux is diminished, and in some cases are incorporated into the geologic record. In this study, microbial assemblages from 134 native and experimental samples across 5,500 km, representing a range of habitat substrates (carbonate nodules and slabs, sediment, bottom water, and wood) and seepage conditions (active and low activity), were analyzed to address two fundamental questions of seep microbial ecology: (i) whether carbonates host distinct microbial assemblages and (ii) how sensitive microbial assemblages are to habitat substrate type and temporal shifts in methane seepage flux. Through massively parallel 16S rRNA gene sequencing and statistical analysis, native carbonates are shown to be reservoirs of distinct and highly diverse seep microbial assemblages. Unique coupled transplantation and colonization experiments on the seafloor demonstrated that carbonate-associated microbial assemblages are resilient to seep quiescence and reactive to seep activation over 13 months. Various rates of response to simulated seep quiescence and activation are observed among similar phylogenies (e.g., Chloroflexi operational taxonomic units) and similar metabolisms (e.g., putative S oxidizers), demonstrating the wide range of microbial sensitivity to changes in seepage flux. These results imply that carbonates do not passively record a time-integrated history of seep microorganisms but rather host distinct, diverse, and dynamic microbial assemblages.",
"introduction": "INTRODUCTION Marine methane seeps serve as islands of diverse and dense deep-sea life, with food webs extending from microorganisms to varied megafauna, including clams, mussels, and tube worms ( 1 – 3 ). Distinct habitats associated with methane seeps include sediments, bottom water, loosely consolidated carbonate protoliths (herein called “nodules”), fully lithified carbonate blocks and pavements (herein called “carbonates”), and, occasionally, wood. Marine methane seep microbial communities and corresponding geochemistry within sediments have been intensively investigated and have been found to frequently be dominated by microbial taxa performing anaerobic oxidation of methane (AOM), notably, anaerobic methane-oxidizing archaea (ANME) and deltaproteobacterial sulfate-reducing bacteria (SRB) ( 4 – 6 ). More broadly, seep sediments are biologically diverse locales that host microorganisms spanning many phyla and are often rich in Epsilonproteobacteria and Gammaproteobacteria in addition to the canonical AOM-associated taxa ( 6 – 9 ). A distinct “seep microbiome,” rich in Deltaproteobacteria , Methanomicrobia , and candidate divisions Hyd24-12 and JS1, is apparent when seep sediment- and nodule-associated microbial assemblages are compared to other marine environments ( 10 ). Authigenic carbonates, which are believed to form as a result of increased alkalinity associated with AOM metabolism, constitute the most pervasive solid habitat substrate at methane seeps but are historically less well sampled than sediments. Carbonates are known to host lipid ( 11 , 12 ) and ribosomal DNA ( 9 , 11 , 13 ) biomarkers, as well as record carbon isotopic compositions reflective of microbial AOM processes ( 14 , 15 ). Seep carbonates have recently been shown to host viable autoendolithic (organisms whose metabolism induces self-entombing mineral formation) Archaea and Bacteria capable of methane oxidation ( 16 , 17 ), as well as metazoan communities ( 18 ). Carbonates themselves occur in a variety of sizes, morphologies, and mineralogies. These include millimeter- to centimeter-scale poorly consolidated precipitates, termed “nodules” or “concretions,” occurring within seep sediments ( 19 – 21 ). Seep-associated carbonates are also frequently found exposed at the seafloor in isolated blocks with sizes from centimeters to tens of meters and continuous pavements ( 22 , 23 ), often extending both laterally and vertically from the site of contemporary methane seepage ( 24 , 25 ). Observations of carbonates at sites lacking contemporary seepage provide evidence that carbonates can outlive seepage processes on the seafloor, supported by the recovery of demonstrably seep-associated carbonates from geologic outcrops as old as 300 million years ( 26 ). Diversity relationships between microbial assemblages associated with seep sediments, nodules, and carbonates have just recently begun to be explored ( 9 , 19 ). Seepage flux can increase and decrease, as well as shift spatially, on a scale of days ( 27 ) to weeks ( 28 ) to centuries ( 27 , 29 ). Microbial assemblages presumably adapt to spatial and temporal changes in seepage flux, but the extent and rate of response in situ remain uncharacterized. Contemporary seepage activity is often defined categorically based on the presence or absence of diagnostic seafloor chemosynthetic communities within methane seeps. Active sites are defined, in this study and elsewhere ( 18 , 27 , 30 , 31 ), as hosting sulfur-oxidizing bacterial mats, clam beds, dense snail colonies, and/or methane ebullition, while low-activity areas lack those diagnostic indicators of contemporary seepage. Notably, low-activity sites are often within <10 2 m of active sites, frequently host carbonates, and can still exhibit microbial activity, including AOM, at reduced rates ( 16 ). Diversity surveys using conventional cloning and sequencing have shown that seep-associated archaeal assemblages, in which only a fraction of the taxa were ANME subgroups, differed based on local seepage activity. The same trend was not apparent in bacterial assemblage composition, which instead was more influenced by habitat substrate (sediment vs. nodule vs. carbonate) ( 9 ). Lipid biomarker profiles from seep sediment and microbial mat samples have been shown to be differentiated partially by sulfate reduction rate, which is likely in turn correlated with seep activity ( 32 ). Off-seep sites host microbial assemblages that are distinct from both active and low-activity sites, further indicating the existence of a seep microbiome ( 6 , 9 , 10 ). Here, a combined comparative and experimental in situ approach was applied to characterize the relationship between seep microbial assemblages, habitat substrata (carbonate vs. sediment vs. nodule vs. bottom water vs. wood), and varying seep activity (active vs. low-activity stations). By coupling a massive sampling effort of native, unperturbed seep carbonates to in situ transplantation and colonization experiments, we can leverage these compatible datasets to address two fundamental microbial ecology questions. (i) Do seep carbonates host distinct microbial assemblages? (ii) How sensitive are microbial assemblages to habitat substrate type and availability and temporal shifts in methane seepage flux?",
"discussion": "RESULTS AND DISCUSSION Carbonates host distinct and diverse seep microbial assemblages. Ordination of the sample set reveals the microbial assemblages to be most strongly differentiated by habitat substrate (i.e., carbonate, sediments and nodules, bottom water) ( Fig. 1A ) ( R = 0.49; P < 0.001; all analysis of similarity [ANOSIM] results are presented in Table S2 in the supplemental material). Habitat substrate is also the most significant factor associated with microbial assemblages as determined by distLM, accounting for 25% of the intersample variability. Furthermore, carbonates exhibit higher operational taxonomic unit (OTU) richness than the other substrates included in this study ( Fig. 2A ) (Chao1 estimates are given in the text; raw OTU rarefactions are given in Fig. S2 in the supplemental material). These trends are also observed in the macrofauna recovered from seep carbonates ( 18 ), confirming that carbonates host diverse benthic life across multiple trophic levels. Overall microbial assemblages of sediments and nodules are not statistically differentiable, as determined from ANOSIM tests, indicating that sediment-hosted nodules and exhumed seafloor carbonates behave as separate, distinct habitat substrates for microbial habitation ( Fig. 1A ; also, see Table S2 in the supplemental material). Sediments, nodules, and carbonates have recently been shown to host different bacterial, but not archaeal, assemblages in 16S clone library surveys ( 9 ), while recent examination of a subset of our iTag data demonstrated similar microbial communities inhabiting nodules and adjacent sediments, especially in active seep settings ( 19 ). FIG 1 Nonmetric multidimensional scaling ordination of microbial assemblages in this study. Each point represents the entire recovered microbiological assemblage from one sample; samples plotting closer to each other are more similar in microbial composition. Lower stress values indicate better representation of the intersample (dis)similarities in two dimensions. (A) Native, unperturbed samples of sediment, nodule, bottom water, and carbonate habitat substrates. Sample C2693 (orange arrow) represents a nodule-hosted microbial assemblage recently determined to be a biological outlier among sediment and nodules ( 19 ). (B) Ordination of only carbonate samples, representing the native, transplantation, and colonization treatments. (C) Ordination of only colonization samples, representing carbonate and wood substrates at active and low-activity stations. We cannot rule out the possibility that in panel A, bottom water microbial assemblages could be different from those of sediments, nodules, and carbonates because they were extracted by a different method (see “Genomic DNA extraction and 16S rRNA gene sequencing and processing”); the same could be true for the observed difference between carbonate- and wood-hosted assemblages in panel C. FIG 2 Collector’s curves of estimated Chao1 OTU 97 richness. Error bars show 1σ standard deviations. (A) Native microbial assemblages associated with carbonates, sediments, and nodules plus the two bottom water samples. Sediments and nodules were binned as one group because their associated microbial assemblages were indistinguishable according to ANOSIM tests (see Table S2 in the supplemental material). (B) Carbonate samples in this study, separated by treatment category. (C) Carbonate and wood colonization samples. Standard deviations are not given for bottom water and transplant-to-active sample groups, due to the low number of analyzed samples. Raw OTU rarefaction curves are given in Fig. S2A to C in the supplemental material. Examination the top thirty most abundant OTUs in our data set reveals a variety of Archaea and Bacteria comprising the samples ( Fig. 3A ), including taxonomies common to methane seep settings (e.g., ANME subgroups and Deltaproteobacteria ). The higher relative abundance of ANME-1 in sediments and nodules than carbonates is in agreement with previous clone library observations at Hydrate Ridge, while the recovery of epsilonproteobacterial sequences from sediments, nodules, and carbonates is in contrast to previous findings in which they were almost exclusively recovered from sediments ( Fig. 3A ) ( 9 ). Data from sequencing of mock communities suggest a slight bias for ANME-1 and a stronger bias against the recovery of ANME-2 sequences by the modified Earth Microbiome Project (EMP) protocol (David H. Case and Victoria J. Orphan, unpublished data). Thus, we note the relative abundance of these groups may in reality be slightly lower (ANME-1) or higher (ANME-2) than recovered in our iTag data set. However, the intersubstrate trends, which are similar for ANME-1 and ANME-2, should be unaffected. Abundance patterns of ANME and other taxa are discussed in detail in the sections below, in the context of results from our experimental manipulations. FIG 3 Boxplot of OTU relative abundances from the 82 native samples in this study. Sediments and nodules are binned as one group because ANOSIM tests revealed their associated microbial assemblages to be statistically indistinguishable. Boxplot centerline represents the median (50th percentile [ Q 50 ]). The top and bottom hinges represent Q 75 and Q 25 quartiles, respectively. The upper and lower whiskers correspond to the highest and lowest data points within 1.5 times the interquartile range ( Q 75 minus Q 25 ) from the median. Any data points outside that range are identified by gray dots. The same plotting format is applied to Fig. 5 . (A) Relative abundances of the top 30 most abundant OTUs in the data set, grouped by taxonomy. The full data set contains 1,057 OTUs, but the top 30 OTUs account for 1%, 67%, and 43% of the sequences recovered from bottom water, sediment/nodule, and carbonate substrates, respectively. (B) Relative abundances of OTUs revealed to be strongly associated with particular habitat substrates. Intersubstrate differences in microbial assemblages are a cumulative result of contributions from many OTUs; even OTUs strongly associated with a particular habitat substrate only contribute several percent to the total intersubstrate variability. Note that the JS1 OTU is the same in panels A and B—it is both highly abundant and strongly associated with sediments and nodules. The Marine Group 1 OTU in panels A and B is different—there is one Marine Group 1 OTU highly abundant in the data set in panel A and another Marine Group 1 OTU strongly associated with bottom water samples (panel B). This highlights the variable distribution of phylogenetically similar OTUs. The y axis of panel A also applies to panel B. Raw OTU data used to generate this plot are available in Table S3 in the supplemental material. Intersubstrate differences in microbial assemblage are the cumulative result of contributions from many OTUs, with no single OTU accounting for more than 2% of the total intersubstrate differences. Nonetheless, several OTUs can be identified which are strongly associated with one habitat type ( Fig. 3B ). Notably, taxa previously identified as diagnostic of the “seep microbiome” (i.e., JS1 archaea and Deltaproteobacteria [10]) are observed in our data set to be characteristic of sediments and nodules but not carbonate habitats (Fig. 3B). In determining the “seep microbiome,” Ruff et al. ( 10 ) examined methane seep sediments and nodules exclusively; our data thus corroborate their results but also further demonstrate that seep carbonates host distinct microbial assemblages. Carbonates, to the exclusion of other habitat substrates, are observed to host an OTU associated with the gammaproteobacterial JTB255 Marine Benthic group (Fig. 3B). The physiology of this group remains undetermined, though uncultured members have been recovered from a variety of marine sediments ( 33 , 34 ), including methane seeps ( 35 ). OTUs associated with the deltaproteobacterial SAR324 clade and thaumarchaeal Marine Group 1 are particularly abundant in the bottom water samples (Fig. 3B), although we note that a separate thaumarchaeal Marine Group 1 OTU is more abundant on carbonates than on other substrates (Fig. 3A). This exemplifies the potential for OTUs of similar phylogeny to be differentially distributed in the environment. The Shannon diversity index (H′), which measures evenness in addition to richness, is higher in the carbonates than either the sediments/nodules or the bottom waters (see Fig. S2D in the supplemental material). Carbonate-associated assemblages may exhibit distinct microbial molecular signatures due to either geochemical (i.e., preferential adsorption of metabolites to the carbonate matrix [ 36 ]), physical (i.e., a site for microbial biofilm attachment), or historic (i.e., formation within or above the sediment column [ 37 ]) factors. Examination of the OTU overlap among native habitat substrates (Fig. 4A) demonstrates that carbonates share more OTUs with sediments and nodules than bottom waters, supporting the hypothesis of formation within the sediments, followed by subsequent exhumation and exposure at the seafloor ( 14 , 37 ). However, bottom waters share more OTUs with carbonates than sediments or nodules, revealing that a subset of bottom water microorganisms do passively or actively inhabit carbonates exposed at the seafloor ( Fig. 4A ). Close overlap in assemblage composition is observed between some of the carbonates (~10 of 57, all from active seep stations) and sediments/nodules ( Fig. 1A ; also note the 10 carbonates highlighted in Fig. S3A in the supplemental material). It is possible that carbonate samples hosting microbial assemblages similar to sediments/nodules may have contained excess sediment entrained in the rock matrix upon recovery ( 37 ); alternatively, nodules in the overlapping region may have been sufficiently lithified to begin hosting carbonate-like microbial assemblages (e.g., nodule C2693 in Fig. 1A ), though this does not necessarily explain similarity of some sediment samples. The compositional overlap between ~10 active-station carbonate assemblages and sediment/nodule assemblages is not derived from geographic proximity, as the sediments/nodules from HR do not exclusively plot in close proximity to the carbonate samples, which are dominantly from HR (see Fig. S3A in the supplemental material), nor are the overlapping carbonates unified by seafloor station, mineralogy, or collection year. Demonstration of microbial variability within seep carbonates. The high total OTU richness of carbonates ( Fig. 2A ), combined with OTU overlap between carbonates and other substrates ( Fig. 4A ) could indicate that carbonates represent a passive repository of preserved and extant microorganisms. We tested this possibility by first examining in detail the native carbonate samples ( n = 57), which allows inference of the environmental indicators associated with differences between carbonate-hosted microbial assemblages. We then coupled these interpretations to the in situ transplantation ( n = 6) and colonization ( n carbonate = 20; n wood = 26) experiments, respectively (see below). FIG 4 Comparison of OTU 97 overlap among various samples and treatments. In order to ensure equal depth of sampling across each substrate type, two representative samples of each substrate were chosen randomly (sample numbers are given). The number in each region denotes the number of OTUs, and text size is proportional to OTU count (as is circle size for panels D to I). (A) OTU overlap between the four native seep habitat substrates examined in this study: sediments, nodules, carbonates, and bottom water. In order to minimize geographic bias in the analysis, samples were chosen from active stations at Hydrate Ridge south (the only exception was bottom water sample 5472, which was from an HR-South low-activity station). Note that carbonates host the richest OTU diversity (see the collector’s curve in Fig. 2A ), including a large number of OTUs which are distinct to carbonates. Carbonates share more OTUs with sediments and nodules than with bottom waters, possibly indicative of an origin within the sediment column and subsequent exhumation and exposure at the seafloor. Bottom waters contribute more OTUs to carbonates than to either sediments or nodules—consistent with the recovery of our carbonates from directly on the seafloor. (B and C) OTU overlap of active and low-activity control carbonates, and transplant-to-low-activity carbonates, for the HR-3/-4 and HR-7/-8 transplant experiments. Transplant-to-active carbonates were not included due to their low sample number ( n = 1 each for HR-3/-4 and HR-7/-8). (D to I) OTUs observed in native carbonate samples versus colonized carbonate samples as a function of Hydrate Ridge Station. Stations were included only if they received colonization carbonate deployments and we had recovered native carbonates from the same station (these criteria excluded HR-1, HR-2, HR-6, HR-11, and the Southeast Knoll). Left column (red [D to F]) are active stations, right column (blue [G to I]) are low-activity stations. In each case, the darker color represents the native carbonates and the lighter color represents the colonized carbonates. In most cases, the majority of recovered OTUs from colonization carbonates were also present in native carbonates. On their own, native carbonate-associated microbial assemblages demonstrate clear differentiation according to seep activity ( R = 0.45; P < 0.001) ( Fig. 1B ; also, see Table S2 in the supplemental material), mineralogy ( R = 0.44; P < 0.001) (see Fig. S4 and Table S2 in the supplemental material), and seafloor station ( R active stations = 0.31, P active stations = 0.002; R low-activity stations = 0.27, P low-activity stations = 0.037) (see Table S2 in the supplemental material). The similar parsing of native carbonate-hosted assemblages by seep activity and mineralogy is partially explained by our observation of a qualitative relationship between seep activity and carbonate mineralogy, with a higher proportion of aragonite-bearing carbonates recovered from low-activity stations (see Text S1 and Fig. S4 in the supplemental material). This suggests that seep activity and carbonate mineralogy are not independent environmental factors in our data set. The biogeographic differences between stations (~10 2 to 10 4 m) are in agreement with previous observations of within-seep microbial and geochemical heterogeneity ( 6 , 38 ) and recent findings that sediment-associated microorganisms in seeps exhibit “global dispersion and local diversification” ( 10 ). A distance-decay curve demonstrated that if a biogeographic effect on microbial similarity exists on a 10 5 - to 10 6 -m scale, it is masked by other environmental factors (see Fig. S5 in the supplemental material). We frame our further discussion in terms of seep activity because it is strongly associated with differences between carbonate-hosted assemblages, and importantly, our sample collection and in situ transplantation and colonization experiments were explicitly performed in order to test biological variability as a function of seep station activity. However, we emphasize that seep activity is a qualitative environmental indicator that may be correlated with other environmental factors, such as carbonate mineralogy. With regard to standard ecological metrics of OTU richness and evenness ( Fig. 2B ; also, see Fig. S2B in the supplemental material), seep activity does not differentiate the native carbonate-associated microbial assemblages. This indicates that while carbonates at low-activity stations host distinct assemblages, they are not less diverse than microbial assemblages from carbonates at active stations. Active-station carbonates are particularly rich in OTUs associated with putative sulfur-oxidizing organisms belonging to the epsilonproteobacterial and gammaproteobacterial families, Helicobacteraceae and Thiotrichaceae , respectively, compared to carbonates from low-activity stations ( Fig. 5 ; also, see Table S3 in the supplemental material). These organisms are likely supported by high sulfide concentrations produced by sulfate-coupled AOM at active seep stations. Among seep sediments in the Mediterranean Sea, epsilonproteobacterial Helicobacteraceae were found to be an indicator taxon for seepage ( 6 ), which our data corroborate. Data from hydrothermal vent systems also exhibit clear differences in abundance of putative sulfur-oxidizing Epsilonproteobacteria between active and low-activity (or inactive) sites, with increased abundance at active vent sites where delivery of reduced fluids is high ( 39 ). Furthermore, Epsilonproteobacteria have been observed in time-resolved experiments to rapidly respond to geochemical heterogeneity and experimental perturbations (i.e., colonization of fresh substrate) in hydrothermal vent systems ( 39 – 41 ). Physiologies of specific groups of the Gammaproteobacteria often include oxidation of either sulfur or methane ( 42 , 43 ), both of which are common at settings with increased delivery of reduced fluids. FIG 5 Boxplot of carbonate-associated relative abundance data of selected key OTUs identified by SIMPER, representing notable taxonomic groups. Note that data for some groups are combined from several OTUs (OTU data are reported individually in Table S3 in the supplemental material). Although in all cases a minority of the OTUs were identified for presentation (e.g., 3 of 8 for ANME-1), these generally represented the majority of the total sequences recovered from each taxonomic group (e.g., 95% of all sequences for ANME-1). When generated with all OTUs associated with each taxon, this plot does not change substantially (data not shown). ANME-1 archaea, which are the most abundantly recovered ANME in the entire iTag data set, exhibit wide ranges of relative abundance in both active and low-activity seep stations, with higher average relative abundance at low-activity stations, in agreement with previous clone library observations ( Fig. 5 ) ( 9 ). Similarly, the deltaproteobacterial family Desulfobacteraceae does not exhibit a clear difference in observed relative abundance as a function of seep activity ( Fig. 5 ). It thus appears that some ANME-1 and deltaproteobacterial OTUs may be relatively insensitive to seepage level. This was unexpected, as these are key taxa involved in the AOM process and therefore hypothesized to occur at higher relative abundance in methane-replete, presumably “active” seep stations. ANME-1 may perform methanotrophy even within carbonates at low-activity stations, consistent with recent reports of AOM associated with carbonates on the periphery of active seepage ( 16 ). Alternatively, relic DNA from AOM-associated organisms may be preserved within carbonate rocks, as the carbonate precipitation process causes self-entombment, potentially sealing off inhabited pores ( 9 , 11 , 13 , 17 ). Evidence for biomarker preservation within carbonates has been described for lipids, which are more recalcitrant to degradation than DNA ( 11 , 12 , 37 ). Demonstration of successional dynamics: transplantation experiments. The “snapshot” view of carbonate-associated microbial ecology is augmented by the seafloor transplantation experiments, which allow us to observe in situ microbial successional patterns by simulating seep quiescence and activation. In situ flux measurements at Hydrate Ridge have shown that seep activity can shift on week- to month-long timescales ( 27 , 28 ), indicating our 13-month transplantation experiments are relevant to contemporary processes at Hydrate Ridge and potentially in other methane seep regions. The OTU composition of the four active-to-low-activity transplanted microbial assemblages are statistically differentiable from both the native, active carbonate-associated microbial assemblages ( R = 0.32; P = 0.008) and the native, low-activity carbonate-associated assemblages ( R = 0.88; P < 0.001) ( Fig. 1B ; also, see Table S2 in the supplemental material). The four microbial assemblages transplanted from active to low-activity stations are more similar to the native, active carbonate assemblages (i.e., where they originated) than to the native low-activity assemblages (i.e., where they were transplanted) ( Fig. 2B and ANOSIM results). The four transplanted carbonates exhibit approximately 30% lower overall OTU richness than native carbonates ( Fig. 2B ), but in-depth analysis of OTU overlap between transplanted and native carbonates reveals a level of fine structure to the microbial turnover and succession ( Fig. 4B and C ). At the paired HR-3/-4 and HR-7/-8 stations, 68% and 52%, respectively, of the OTUs associated with native, active control carbonates were not recovered upon simulated seep quiescence after 13 months (see Table S3 in the supplemental material). The “lost” OTUs are supplanted by characteristic OTUs gained from the low-activity sites (28 and 37 OTUs, representing 18% and 17% of the recovered OTUs for HR-3/-4 and HR-7/-8 transplants, respectively) as well as OTUs unique to the transplants and not recovered from native carbonates (20 and 24 OTUs for HR-3/-4 and HR-7/-8, respectively) (see Table S3 in the supplemental material). Nearly half of the OTUs recovered among the HR-3/-4 and HR-7/-8 transplants were cosmopolitan OTUs that were also observed in both the native active and the native low-activity carbonates ( Fig. 4B and C ; also, see Table S3 in the supplemental material). Thus, a loss of over half the initial OTUs upon seep quiescence is masked by gain of new OTUs, both unique and shared with the low-activity controls. Combining the observation of overall similarity to native, active assemblages, diminished overall OTU richness, and specific turnover among the carbonates transplanted to low-activity stations, we can begin to paint a picture of microbial succession upon seep quiescence. Most major (i.e., highly abundant) constituent members of carbonate-associated microbial assemblages are resilient to 1 year of quiescence (or their DNA does not degrade), as evidenced by the fact that transplanted carbonates plot among the native, active controls in Fig. 1B . Indeed, of the four carbonates transplanted to low-activity sites, we observe that 49 to 90% of the recovered sequences are from resilient OTUs shared with the active-station controls (see Table S3 in the supplemental material). However, over the course of a year, low-abundance assemblage members are vulnerable to cessation of seep activity: the average relative abundance of lost OTUs in the native, active controls upon simulated quiescence was <0.5% (see Table S3 in the supplemental material). Examining specific taxa of interest, we find the gammaproteobacterial Thiotrichaceae OTUs remain at a relative abundance similar to that of the native, active carbonates, consistent with resilience to seep quiescence ( Fig. 5 ). In contrast, epsilonproteobacterial Helicobacteriaceae OTUs that are highly abundant in native, active carbonates had mostly disappeared after 13 months of simulated seep quiescence ( Fig. 5 ). Thus, two putative sulfur-oxidizing groups exhibit different 16S rRNA gene distribution, highlighting the potential for variable response to environmental change, even among taxa putatively belonging to the same guild. ANME-1 OTUs were recovered at high relative abundance in the carbonates transplanted to low-activity stations, consistent with the trend observed in native, low-activity carbonates and suggesting an ability to respond over a period of time that may represent, to ANME archaea, only a few generations ( 44 – 46 ). The two carbonates which experienced simulated seep activation (transplanted from low-activity to active stations) host microbial assemblages different from those in low-activity, native carbonates and somewhat similar to those in native, active assemblages ( Fig. 1B ), although this experimental set suffers from low sample number as a result of technical difficulties in recovering two of four originally transplanted carbonates. In juxtaposition to seep quiescence, which demonstrated resilience of the bulk microbial assemblages, our simulation of seep activation indicates that assemblages are relatively quick to respond to renewed seepage conditions. This is especially true among the epsilonproteobacterial Helicobacteraceae OTUs, which are recovered in high relative abundance in the carbonates transplanted to active stations, despite low relative abundance in the low-activity carbonates ( Fig. 5 ). Other OTUs (for example, gammaproteobacterial Thiotrichaceae ) clearly demonstrate a slower response to seep activation ( Fig. 5 ). Examination of two OTUs of putatively heterotrophic Chloroflexi , the Anaerolineaceae and Caldilineaceae , also reveals slow response to seep activation, despite their relatively high recovery among native, active seep carbonates ( Fig. 5 ). The Anaerolineaceae OTU also exhibits markedly higher tolerance to low-activity conditions than the Caldilineaceae OTU ( Fig. 5 ), highlighting the potential for different ecological expression among groups of similar phylogeny. The coupled transplant experiments provide strong evidence that many carbonate-associated seep microbial taxa are adapted to cycles of seep quiescence and activation. This may be ecologically advantageous in an environment where fluid flow has a tendency to fluctuate rapidly and frequently ( 27 , 28 ). Recalcitrance to seep quiescence is consistent with low but measurable AOM from carbonates at low-activity stations ( 16 ), and the physical buffering provided by carbonate habitats has been proposed as a factor for maintenance of microbial assemblage viability during periods of diminished seepage ( 17 ). Alternatively, we note that 3 of the 4 carbonates transplanted from active to low-activity stations were composed of a mix of calcite and dolomite—mineralogies more common at active stations than low-activity stations (see Fig. S4 in the supplemental material). If mineralogy significantly drives microbial composition, the observed recalcitrance to community shift may be explained by the fact that the transplanted carbonates bore mineralogies qualitatively associated with active-seep-type microbial assemblages. In contrast, the two samples transplanted from low-activity to active stations were aragonite/calcite mixes—a mineralogical composition regularly recovered from all seep stations regardless of activity (see Fig. S4 ). Thus, the observed shift to an active-seep-type community is more likely a function of the seep activity shift than of mineralogy. The rapid microbial rebound upon simulated seep activation may be analogous to previous observations of microbial community activation from deep terrestrial and marine subsurface environments ( 44 , 47 ). Species richness in carbonates transplanted to active stations is higher than the reciprocal transplants—though still lower than native carbonates—further indicating microbial assemblage responsiveness to simulated seep activation ( Fig. 2B ). Diminished OTU richness upon transplantation (in either direction) is also evidence against a “time-integrative” model of carbonate microbial assemblages: if carbonates were passive recorders of all historic seep microbial DNA, OTU richness would not be expected to decrease. Demonstration of successional dynamics: colonization experiments. Though our transplantation experiments best simulate the temporal variability of seepage for established microbial assemblages, they are limited in scope. To increase the interpretative power of our data set, we supplemented the transplant experiments with carbonate (calcite and dolomite) and wood (fir and pine) colonization experiments to address the successional patterns and responsiveness of seep microorganisms colonizing at the seabed under conditions of differing seep activity and colonization substrate type. Results from these experiments follow similar trends observed in the survey of native microbial assemblages where both habitat substrate ( R carbonate vs wood = 0.63, P < 0.001) and seep activity ( R = 0.38, P < 0.001) differentiate the recovered microbial diversity ( Fig. 1C ; also, see Table S2 in the supplemental material). In contrast to the survey of native carbonates, mineralogy did not contribute significantly to differences in total colonizing assemblage diversity ( P = 0.109) (see Table S2 in the supplemental material), further suggesting that the relationship between mineralogy and microbial diversity in the native carbonates may be due to a qualitative link between mineralogy and seep activity (see Fig. S4 in the supplemental material). Microbial assemblages colonizing carbonates exhibited higher OTU richness and evenness than those colonizing wood ( Fig. 2C ; also, see Fig. S2C and D in the supplemental material), substantiating the role of seep carbonates, specifically, as hosts of diverse microbial populations. While hosting comparable OTU richness to the native carbonates ( Fig. 2 ), the microbial assemblages colonizing the sterile carbonates at the seabed were, after 13 months, significantly different from native microbial assemblages collected in this study ( R = 0.65, P < 0.001) ( Fig. 1B ; also, see Table S2 in the supplemental material). This supports general trends in the transplant experiments, suggesting that more than 13 months is required to achieve a mature successional phase if it is assumed that given enough time the colonizing assemblages would eventually mimic the native assemblages. Alternatively, the colonization carbonates might never host microbial assemblages completely similar to the native carbonates, considering the different history of colonization carbonates (located at the seabed) and native carbonates (believed to have formed within the sediment column and later to have been exhumed). Notably, however, sterile carbonates incubated at the seafloor share most of their observed OTUs with the native carbonates ( Fig. 4D to I ). The discrepancy between colonization and native carbonates hosting quite different microbial assemblages ( Fig. 1B ) and yet sharing many OTUs ( Fig. 4D to I ) implies that assemblage differences are generally a function of differential OTU relative abundance, not of the presence/absence of different OTUs themselves. Indeed, an ANOSIM test on presence/absence-normalized data reveals a diminished, though still significant, strength of difference between native and colonized carbonate microbial assemblages ( R = 0.53, P < 0.001). In further support, among the six colonization/native pairings examined in detail ( Fig. 4D to I ), the majority (average 63%, range 46 to 84%; n colonization samples = 12) of the recovered colonization sequences were from OTUs shared between the colonization and native carbonates. In-depth analysis of OTU overlap at station HR-9, chosen because of the wide array of habitat types and experimental samples obtained there, reveals that of the various OTUs shared between native and colonized carbonate assemblages, many are also shared with sediment and nodule assemblages (see Fig. S6 in the supplemental material). This suggests some transference of sediment-hosted microbes onto the colonization carbonates. The mode of transfer is currently not known but may be associated with direct microbial motility ( 48 , 49 ), macrofaunal grazing/bioturbation ( 2 , 48 ), and/or advection from fluid flow or gas ebullition ( 50 ). At station HR-9, where 376 OTUs were reproducibly recovered from both colonization carbonates, 19% ( n = 71), 3% ( n = 11), and 4% ( n = 14) were exclusively sourced from carbonates, sediments/nodules, and bottom waters, respectively. The bottom water samples associated with this station contained 1% to 2% relative abundance of an OTU associated with the gammaproteobacterial Colwelliaceae , which were also recovered at moderate relative abundances from the colonization carbonates (<1% to 20%) ( Fig. 5 ; also, see Table S3 in the supplemental material) despite a lack of detection on either native or transplanted carbonates. This further indicates some transfer of bottom water microorganisms onto carbonates during early-phase succession and is consistent with common ecophysiology of Colwellia as generally marine, psychrophilic, motile, chemoorganotrophic microorganisms ( 42 ). Thus, OTU recovery from multiple nearby substrates, coupled to the observed difference between colonized carbonate and wood microbial assemblages after 13 months ( R = 0.63, P < 0.001) ( Fig. 1C ), could be explained by two hypotheses: either (i) OTUs are recruited from all surrounding habitats, followed by assemblage differentiation according to habitat substrate (i.e., carbonates diverge from woods), or (ii) carbonate colonization is a substrate-specific process from the very first microbial succession, and then over time occasional passive capture of OTUs from other habitat substrates occurs. In either case, the colonization data support the observation from native samples that carbonates host distinct microbial assemblages. Furthermore, carbonate distinctiveness is not simply a product of time-integrated passive capture of sediment-hosted microorganisms, nor does it depend on a history of burial in sediment. Microbial diversity within the carbonate colonization experiments is almost wholly explained by seep activity differences, in further support of observations from native carbonates ( Fig. 1C ) ( R = 0.81, P < 0.001). Indeed, OTUs associated with the epsilonproteobacterial Helicobacteraceae exhibit a wide range of relative abundances in the colonization carbonates at active stations but only a very minor amount of colonization at low-activity stations ( Fig. 5 ; also, see Table S3 in the supplemental material). The Thiotrichaceae OTUs also demonstrate colonization patterns reminiscent of distributions observed in the native carbonates, again indicating that putative sulfur-oxidizing OTUs are dynamic responders to carbonate substrate availability in regions of seep activity at the seabed. However, the specific Thiotrichaceae OTU observed to most strongly colonize experimental carbonates was different than the Thiotrichaceae OTU more frequently observed in the native carbonates (see Table S3 in the supplemental material)—demonstrating the potential for within-group variability in ecological expression. The recovery of Helicobacteraceae or Thiotrichaceae OTUs was not obviously tied to qualitative observations of bacterial mats upon recovery of colonized carbonates from the seafloor. Previous studies of microbial colonization in shallow marine sediments and near hydrothermal vents have observed a dominance of early-stage colonization by Epsilonproteobacteria ( 40 , 41 , 48 , 51 ), and similar ecological behavior appears to be occurring in methane seeps. The rapid colonization by Epsilonproteobacteria in various marine settings has been attributed to both a tolerance for rapidly changing physicochemical conditions and motility within many members of the class ( 40 , 48 ). We observe that our key Helicobacteraceae OTUs were recovered in high relative abundance in methane seep sediments and low relative abundance in bottom water samples (see Table S3 in the supplemental material); therefore, it appears likely the Helicobacteraceae recovered in the colonization experiments were inoculated from underlying sediments, in contrast to Colwelliaceae OTUs derived from overlying bottom waters. Colonization by the ANME-1-associated OTUs (the same OTUs as recovered from native carbonates) on the sterile carbonates was observed at low levels at both active and low-activity stations ( Fig. 5 ). Any level of colonization by ANME-1 is intriguing for two reasons. First, ANME-1 are believed to have doubling times on the order of several months, so the 13-month course of the colonization experiments could reasonably be expected not to have provided enough time for ANME-1 archaea to colonize and become established on the fresh carbonate substrates ( 44 – 47 ). Second, ANME-1 are obligate anaerobes typically associated with highly reducing conditions located deeper within the sediment column at seeps and near the sulfate-methane transition zone, not at the sediment/water interface, where the colonization experiments were located ( 52 ). An exception to this are the Black Sea “reefs,” composed partly of ANME-1; however, these grow into permanently stratified bottom water of the euxinic Black Sea ( 53 ). That ANME-1 OTUs are observed at significant levels in the sediment samples but at negligible levels in the aerobic bottom water samples ( Fig. 3 ) indicates that ANME-1 almost certainly colonize the carbonates seeded by the underlying sediments. This highlights the complexity of potential mechanisms driving regional and global between-seep dispersion of ANME-1 archaea and perhaps other ANME subclades, as previously observed ( 10 ) and perhaps accomplished through periodic sediment disturbance. In contrast to our observations, Archaea were not observed as early colonizers in hydrothermal vent colonization experiments, despite their presence within in situ vent communities ( 40 ). Our experiments suggest that ANME-1 archaea may exhibit phenotypes thus far undiscovered in seep settings or may be distributed by hydrological flow or macrofaunal movements (pumping, filtering, burrowing, defecation, etc.). Wood-colonizing microbial assemblages at methane seeps in the Mediterranean Sea have been observed to be different than surrounding, off-seep sediment-hosted microbial assemblages ( 54 ). Our data further demonstrate that even among active and low-activity seep stations, wood-colonizing microbial assemblages differ after 13 months ( Fig. 1C ). The stark difference between carbonate- and wood-colonizing assemblages in our data set highlights the importance of habitat substrate to deep-sea microbial assemblages. The mere presence of putative sulfur-oxidizing Epsilonproteobacteria and Gammaproteobacteria in the wood colonization experiments suggests that wood falls may act as ephemeral sulfide-rich reducing habitats, possibly representing stepping stones between seeps and vents for chemosynthetic communities, as has been hypothesized for metazoans and Bacteria ( 54 , 55 ). Our results are consistent with previous characterizations of native wood fall samples, as well as deep-sea benthic wood colonization experiments, which yielded observations of phylogenetically diverse microbial assemblages, including, but not limited to, the Bacteroidetes , Firmicutes , Spirochaetes , Epsilonproteobacteria , and Gammaproteobacteria ( 54 , 56 , 57 ), but very limited recovery of methanogenic and methanotrophic archaeal taxa ( 57 ). The lack of significant ANME colonization in the wood experiments (see Table S3 in the supplemental material) indicates that AOM-related archaeal taxa may have more difficulty spreading geographically via wood substrates than many Bacteria . That AOM-related archaeal taxa appear to be able to colonize carbonate substrates, even on relatively short timescales, indicates a possible mode of wide geographic dispersion. Other hypotheses have included transportation in the guts of deep-sea metazoans or distribution during ocean anoxic events ( 10 ), both of which may complement the apparent suitability of carbonate habitats for ANME. In summary, the deployment of in situ manipulation experiments, coupled to an extensive characterization of native microbial assemblages in association with various seep habitat substrates, has enabled unique insights into the ecology of seep microorganisms. Microbial assemblages associated with carbonates at methane seeps are distinct from, and more diverse than, other habitat substrates examined in this study: sediments, nodules, and bottom waters. Further, bulk carbonate-associated microbial assemblages are adapted to resist seep quiescence and poised to respond to seep activation over 13 months. OTUs associated with the epsilonproteobacterial Helicobacteraceae are particularly sensitive to seep activity. Colonization experiments corroborate the idea that carbonates host distinct and diverse microbial assemblages, and recovery of ANME-1 OTUs associated with the carbonates suggests more dynamic physiologies and/or distribution processes for these organisms than previously hypothesized. The difference in the microbial assemblages associated with native active and low-activity carbonates, coupled to the dynamics and decreased OTU richness observed in the transplant experiments, suggests that upon the final quiescence of a historic methane seep, the genomic microbial signatures recorded in carbonates could differ from those microbes which were present during active seepage. Investigation of our same research questions should be applied to lipid profiles, to investigate whether trends observed at the genomic level are likely to be preserved in the rock record and, in particular, whether microbial signatures in the rock record merely reflect the final, low-activity period of seep activity rather than the biological assemblage present during the most active phases of seepage and AOM."
} | 12,201 |
39136990 | PMC11348241 | pmc | 6,950 | {
"abstract": "Significance Generating dihydrogen (H 2 ) in an environmentally friendly way is an important challenge. A recently characterized O 2 -stable [FeFe] hydrogenase presents a unique opportunity for redirecting energy produced by Photosystem I (PS I) to H 2 production in a phototrophic biological system that only requires sunlight and earth-abundant elements. Our study presents a strategy to couple such an [FeFe] hydrogenase to PS I by fusing the former with a stomal subunit of PS I (PsaE). Not only does the chimeric nanoconstruct generate reasonably high rates of H 2 when illuminated, but it also functions in the presence of O 2 . By investigating catalytic properties and drawbacks of the nanoconstruct, this work sets the stage for engineering sustainable biofuel production in vivo.",
"discussion": "Discussion Fusion of PsaE to CbHydA1 Changes Its Catalytic Activity. Our experiments indicate a measurable impact of the fusion of PsaE on the activity of Cb HydA1, as evident from two- to threefold lower H 2 oxidation and proton reduction rates than that of a WT protein. In PFV experiments, we observed an increased propensity of the Cb HydA1 for inactivation in the PsaE-fusion variant compared to the WT enzyme. The shift of the reactivation CV wave to a lower potential in the Cb HydA1-PsaE fusion protein can be a consequence of either a lower mid-point reduction potential of the H inact state or faster kinetics of inactivation ( 39 , 40 ). Recently, using EPR, we demonstrated the existence of two structural isoforms of Cb HydA1. We proposed that the unique-to- Cb HydA1 isoform 1 [EPR signals H ox (1) and H ox -CO(1)] is the inactivation-preceding form that relates to the aberrant position of the Cys367 facilitated by a rearrangement of the respective flexible loop ( 39 , 40 ). The dominance of the isoform 1 in EPR measurements of CO-inhibited Cb HydA1-PsaE variant ( SI Appendix , Fig. S7 ) is in line with the observed downshift of the reactivation wave in PFV. Therefore, we consider it likely that adding PsaE to Cb HydA1 affects the mobility of the core protein structure around the H-cluster, resulting in an increased propensity for inactivation in the fusion protein and, thus, lower activity. Furthermore, isoform 1 may naturally be less active since it proposedly relates to an off-H + -pathway arrangement of Cys367 ( 40 ). We also cannot exclude the possibility that the fusion of PsaE to Cb HydA1 could affect the electronic structure of the two accessory [4Fe-4S] clusters (F-clusters) present in the enzyme. Indeed, there is a slight but noticeable shift in the ratio between H 2 evolution and H 2 uptake toward the former in the fusion protein. Also, as-prepared WT and fusion enzymes obtained under identical conditions show somewhat different ratios of H ox and H red H + states in IR ( SI Appendix , Fig. S5 ), possibly due to a somewhat different catalytic bias causing a shift of equilibrium between the two states. However, as both catalytic rates are lower in the fusion protein, further investigation into the redox potentials of the F-clusters is needed to understand this effect fully. Such experiments are underway in our laboratory. Electron Transfer in the Modified PS I Complexes. Our study of the P 700 ∙ + reduction kinetics after an actinic laser pulse provided invaluable insights into the electron transfer mechanisms with the modified PS I complexes. First, the experiments confirmed the early report by Yu et al. ( 55 ) that the absence of PsaE elongates the lifetime of P 700 ∙ + -[F A /F B ] − charge-separated state. Charge recombination rates are inextricably tied to the redox potential of the bound cofactors, the reorganization energies associated with the site, and the distance between the cofactors. Hence, our working hypothesis is that the lack of a PsaE subunit affects the mobility of the PsaC and PsaD subunits, ultimately resulting in a shift in the redox potential of either F A and F B or their distance from F x and thus causing a change in the lifetime of the charge-separated state. While understanding this effect is outside the scope of this work, it is an important line of research to address in the future, as the structural modifications imposed on PsaE by the fusion with Cb HydA1 may pose additional effects on the electron transfer pathways within the PS I complex. Incubation of PS I ΔPsaE with apo- and holo- Cb HydA1-PsaE results in a shift to PS I WT -like P 700 ∙ + -[F A /F B ] – recombination lifetimes, thus indicating a near-quantitative formation of the PS I ΔPsaE : Cb HydA1-PsaE complex. It is somewhat surprising that when we incubate PS I ΔPsaE with the apo-form of the fusion protein (F-clusters present, but no active site), there is very little of the long-lived phase, implying almost no electron transfer from PS I to the hydrogenase. However, it is important to note that the resulting data are averaged over 512 traces. Without an electron acceptor such as the active center (H-cluster), the F-clusters will likely remain reduced after the first couple of flashes for the duration of the experiment, assuming a high quantum yield of electron transfer. As a result, forward electron transfer from F A /F B to the hydrogenase domain will be effectively blocked. The time-resolved optical experiments on PS I ΔPsaE :holo- Cb HydA1-PsaE nanoconstruct in the presence of H 2 demonstrate that the electron transfer from [F A /F B ] – to the F-cluster is not entirely unidirectional. Were that the case, H 2 oxidation by the H-cluster would result in a reduction of the F-cluster but not the [F A /F B ] pair, and hence, we would expect kinetic traces to be similar to that of PS I ΔPsaE :apo- Cb HydA1-PsaE. To the contrary, our time-resolved experiments showed that under a 3% H 2 atmosphere, the charge recombination is substantially more complex. CONTIN analysis shows a complex multiphasic kinetic profile with the dominant decay phase being faster than that observed in the PS I ΔPsaE :apo- Cb HydA1-PsaE nanoconstruct. Therefore, the data indicate that the H 2 -uptake by the hydrogenase domain not only saturates the F-cluster, but also partially reduces the [F A /F B ] couple. This “backfire” effect could inhibit forward electron transfer within PS I, potentially accelerating charge recombination. In line with this suggestion, the removal of H 2 from the headspace resulted in a substantial elongation of the charge recombination lifetime. In this case, two primary decay phases are the PS I WT -like lifetime of ~100 ms and a long-lived charge-separated state >3 s. The long-lived phase, whereby P 700 ∙ + is reduced by ascorbate, increases in amplitude from ~4% of the total signal to ~30%. This increase provides compelling evidence for the forward electron transfer from PS I to hydrogenase with the consequent loss of the electron to proton reduction. The remaining 100 ms phase is likely wherein the fusion protein is bound, but forward electron transfer is not occurring in PS I. We note that in our spin-quantification experiments, only a third of Cb HydA1-PsaE fusion proteins contain a fully assembled H-cluster ( SI Appendix ). This alone could be sufficient to explain why forward electron transfer is only 31% efficient. However, whether there is a preference for binding holo- Cb HydA1-PsaE rather than apo- Cb HydA1-PsaE to PS I ΔPsaE is not known, as discussed below. Comparison with Other Nanoconstructs. The H 2 production rate we observe for our nanoconstruct (84.9 ± 3.1 µmol H 2 mg chl –1 h –1 ) is similar to other PS I-nanoconstructs tested under roughly comparable conditions ( SI Appendix , Table S1 ) ( 11 – 15 , 23 , 25 , 56 – 63 ). Due to subpar H-cluster incorporation rate of ~31% for the Cb HydA1-PsaE fusion protein, the maximum theoretical rate could be upward of ~250 µmol H 2 mg chl –1 h –1 or ~12 e – (PS I) –1 s –1 . However, we refrain from using such an extrapolation as many factors can contribute to the overall quantum yield. While data suggest a near-quantitative binding of a fusion protein to PS I ΔPsaE , it is possible that there is a binding preference of apo- vs holo-proteins, or steric limitations on the trimer affecting the number of Cb HydA1 domains able to interact with PsaC while bound. Also, the ratio between forward electron transfer and charge recombination rates may play a role, as well as the backfire effect discussed above. When discussing related PS I-based nanoconstructs, perhaps the most appropriate comparison would be that with the PsaC-fusion variants of [FeFe] hydrogenases 1 and 2 from C. reinhardtii ( 30 , 31 ). The overall electron throughput of these systems, as determined experimentally, appears to be 8 to 70 times higher than that observed in our case ( SI Appendix , Table S1 ). However, these measurements were carried out under in vivo conditions where reduction of P 700 ∙ + occurs via natural and efficient electron transport, while we utilized a soluble Cyt c 6 as an electron donor in vitro. Hence, the H 2 production rate is likely rate-limited in the latter by the recycling of P 700 rather than by the performance of the active chimeric nanoconstructs. Note that cross-linking Cyt c 6 to PS I resulted in a nearly sevenfold increase in H 2 production rate in the PS I-wire- Ca HydA1 nanoconstructs reported by the Golbeck group under comparable conditions, with a maximum 70-fold increase achieved by lowering pH (up to 2,200 μmol H 2 mg Chl –1 h –1 , SI Appendix , Table S1 ) ( 23 ). It would thus be interesting to perform similar Cyt c 6 cross-linking experiments for our nanoconstructs to estimate the maximum possible H 2 production rate for the Cb HydA1-PsaE:PS I ΔPsaE nanoconstruct. The Nanoconstruct Can Generate H 2 Under Aerobic Conditions. Demonstrating detectable H 2 production by the nanoconstruct under aerobic conditions is highly encouraging. It is important to note that there are several avenues by which O 2 can interrupt energy and electron transfer both within PS I and between PS I and the hydrogenase. Within PS I complex, O 2 can react with Chl triplet states during light harvesting ( 64 ). O 2 can also accept low-potential electrons from [F A /F B ] – cluster pair and even from the A 1 phylloquinones ( 65 – 67 ). A gradual reduction in O 2 concentration in the headspace ( SI Appendix , Fig. S8 B ) supports this notion. Therefore, electron transfer from [F A /F B ] – to Cb HydA1 is in competition not only with charge recombination but also with the reduction of molecular oxygen. For Cb HydA1, O 2 will also functionally inactivate the active site, requiring reducing equivalents from PS I for periodic reactivation. Given this combination of factors, it is not surprising that the rates of H 2 production by the Cb HydA1-PsaE:PS I nanoconstruct are lower than for anaerobically prepared samples. It may be possible to improve the efficiency of the forward electron transfer by varying the length and rigidity of the linker group to increase both protein activation, specific activity, and electron transfer from PS I. Further investigation into the interaction of O 2 with the parts of the nanoconstructs will be required to parse out the contributing factors and devise a mitigation strategy. Nonetheless, even with reduced rates in the presence of O 2 , the Cb HydA1-PsaE:PS I nanoconstruct shows promise for an in vivo system that functions in the presence of constantly changing levels of O 2 . In summary, we have successfully generated a unique chimeric protein nanoconstruct through the fusion of a 7.6 kDa stromal subunit of PS I (PsaE) from the cyanobacterium Synechococcus sp. PCC 7002 onto the C terminus of an oxygen-tolerant [FeFe] hydrogenase from C. beijerinckii ( Cb HydA1) via a flexible [GGS] 4 linker group. We show that Cb HydA1 can be synthetically activated in vitro and retain native-like bidirectional hydrogenase activity. Our transient absorption studies demonstrate that the PsaE substituent provides a mode for selective and nearly quantitative binding of the fusion protein to available PS I ΔPsaE -cores. We were able to detect light-induced H 2 evolution even in an aerobic environment. Our findings provide confidence in PsaE as a viable scaffold for the binding of exogenous proteins to PS I cores. The detailed investigation of the H 2 -producing chimeric nanoconstruct presented here establishes an important basis for future engineering in vivo H 2 -generating systems that can function within O 2 -evolving photosynthetic pathways."
} | 3,145 |
30036678 | null | s2 | 6,953 | {
"abstract": "The development of new heterologous hosts for polyketides production represents an excellent opportunity to expand the genomic, physiological, and biochemical backgrounds that better fit the sustainable production of these valuable molecules. Cyanobacteria are particularly attractive for the production of natural compounds because they have minimal nutritional demands and several strains have well established genetic tools. Using the model strain Synechococcus elongatus, a generic platform was developed for the heterologous production of polyketide synthase (PKS)-derived compounds. The versatility of this system is based on interchangeable modules harboring promiscuous enzymes for PKS activation and the production of PKS extender units, as well as inducible circuits for a regulated expression of the PKS biosynthetic gene cluster. To assess the capability of this platform, we expressed the mycobacterial PKS-based mycocerosic biosynthetic pathway to produce multimethyl-branched esters (MBE). This work is a foundational step forward for the production of high value polyketides in a photosynthetic microorganism."
} | 281 |
40063791 | PMC11929398 | pmc | 6,954 | {
"abstract": "Significance Studies of microbial community composition across time, space, or biological replicates often rely on summary statistics that analyze just one or two samples at a time. Although these statistics effectively summarize the diversity of one sample or the compositional dissimilarity between two samples, they are ill-suited for measuring variability across many samples at once. Measuring compositional variability among many samples is key to understanding the temporal stability of a community across multiple time points or the heterogeneity of microbiome composition across multiple experimental replicates or host individuals. Our proposed framework, F ST -based Assessment of Variability across vectors of relative Abundances (FAVA), meets the need for a statistic summarizing compositional variability across many microbiome samples all at once.",
"discussion": "Discussion We have introduced an index to quantify variability across samples of microbiome composition. We defined the measure through an analogy with the population-genetic statistic F ST , considering microbiome samples in place of populations and microbial taxa in place of alleles. FAVA equals 0 if and only if all microbiome samples are identical, and 1 if and only if each sample contains only a single taxon and more than one taxon is present across all samples ( Fig. 1 ). FAVA can be used as a measure of compositional variability across time points, spatial sampling locations, host individuals, or replicates, quantifying the temporal variability, spatial heterogeneity, or replicability of microbial communities. Because FAVA takes values between 0 and 1 irrespective of the number of sampled taxa, we can compare FAVA values between very different datasets, such as data on abundances of different taxonomic categories. To demonstrate the FAVA framework’s performance as a measure of microbiome variability across many samples, we analyzed two microbiome datasets: an investigation of ruminant microbiome composition along the gastrointestinal tract ( 38 ), and a longitudinal study of human gut microbiome composition before and after an antibiotic perturbation ( 43 ). In the ruminant data, we found that compositional variability across individuals—either within a host species or across host species—was consistently lower at the end of the gastrointestinal tract than in the middle, supporting the view that substantial interindividual heterogeneity is missed when microbiomes are monitored by fecal sampling alone ( Fig. 2 B and D ) ( 18 , 36 ). We found that, in all gastrointestinal regions, taxonomic abundances were much more variable across individuals than were functional abundances, a result that corroborates observations of microbial functional redundancy in the gastrointestinal tract ( Fig. 2 D ) ( 39 ). In the human microbiome data, we found that antibiotic perturbations destabilize microbial communities, resulting in elevated temporal variability following an antibiotic ( Fig. 4 E ). Computing weighted FAVA in sliding windows across temporal samples for each subject increased the granularity of this analysis. Although elevated variability lasted for only one to two weeks postantibiotic on average, few subjects returned to preantibiotic variability levels during the study duration ( Fig. 4 C and D ). We also highlighted the FAVA framework’s ability to quantify temporal variability separate from compositional state by focusing on subjects XDA and XMA, who returned to their preantibiotic variability levels ( Fig. 4 C ) even though only XMA returned to the original composition ( SI Appendix , Fig. S4 ). We introduced two extensions of FAVA: weighted FAVA (Eq. 11 ), which can incorporate both similarity among taxa and distance between samples into the computation, and normalized FAVA, which accounts for the abundance of the most abundant taxon, allowing for more meaningful measurement of variability across small numbers of samples. In our analysis of human gut microbiome data over time ( 43 ), the use of weighted FAVA helped to account for both the combination of weekly and daily samples and the broad range of species appearing in the data. FAVA values can be influenced by the choice of weights. For example, SI Appendix , Fig. S5 presents two hypothetical OTU tables with a large difference in FAVA when weighted by taxonomic similarity, despite having identical unweighted FAVA values. Nevertheless, in our analysis of human microbiome data, although individual FAVA values shift with the incorporation of weights, FAVA values computed across postantibiotic samples are consistently higher than those computed across preantibiotic samples, irrespective of weighting by sampling times, taxonomic similarity, or both ( SI Appendix , Fig. S6 ). Analyzing a higher taxonomic level can be viewed as a special case of weighting by taxonomic similarity. For example, to analyze family abundances in place of species abundances, we would define each entry s k , ℓ of the species similarity matrix to equal 1 if species k and ℓ belong to the same family, and 0 otherwise. The taxonomic similarity matrix considered in SI Appendix , Fig. S5 , for example, is equivalent to supposing that taxa I and K are each in separate families, whereas taxa J and L are in the same family. The result of this figure can consequently be interpreted to mean that matrix 1 has higher FAVA when computed using species (unweighted) rather than family abundances (weighted), while matrix 2 has lower FAVA when computed using species (unweighted) rather than family abundances (weighted). We observe a similar composition-dependent relationship between taxonomic level and FAVA results in the data from Xue et al. ( 43 ) ( SI Appendix , Fig. S7 A ). We computed FAVA across all samples from each antibiotic-taking subject from Xue et al. ( 43 ) using relative abundances of either bacterial families or species. Considering all subjects together, we do not observe a significant difference in FAVA values between the two levels of analysis (Wilcoxon signed rank test, P = 0.17). However, many individuals exhibit sizeable changes in FAVA values depending on the taxonomic level analyzed. SI Appendix , Fig. S7 B highlights the compositions of the three subjects with the largest difference (XAA), smallest difference (XDA), and nearest difference to zero (XGA), comparing FAVA values computed using species and family abundances. Subject XAA’s higher species-level than family-level FAVA value is driven by large shifts in species composition within a single family whose abundance remains relatively constant, similar to matrix 1 in SI Appendix , Fig. S5 . Subject XDA’s higher family-level than species-level FAVA value is a result of a large shift in abundances of families containing many component species, each with only small shifts in abundance—similar to matrix 2 in SI Appendix , Fig. S5 . Finally, the species and family abundances in subject XGA follow very similar trajectories, producing similar species and family-level FAVA values. We emphasize that comparisons of FAVA values between datasets with different numbers of categories, such as between species and family abundances ( SI Appendix , Fig. S7 ), or between taxonomic and functional abundances ( Fig. 2 D ), are enabled by the mathematical design of the FAVA measure. Under a Dirichlet model describing abundances in a set of categories, FAVA depends on the Dirichlet variance but does not otherwise depend on the abundances themselves; simulation of OTU tables in two scenarios, with 3 and 99 taxa, illustrates an identical, linear relationship with Dirichlet variances used for the simulations, irrespective of the number of taxa ( SI Appendix , Fig. S1 A ). As an alternative to FAVA, the variability among a set of samples can also be measured with the mean of a pairwise statistic across all pairs of samples; in the same simulations of SI Appendix , Fig. S1 A , computing one such statistic, the mean Bray–Curtis dissimilarity across pairs of samples, we observe in SI Appendix , Fig. S1 B a strong dependence of the statistic on the number of taxa in the OTU table, so that it cannot be straightforwardly used to compare variability between tables with different numbers of categories. We note that in the human microbiome analysis, we might have expected FAVA values to depend on data quality, as measured by the number of sequence reads used to estimate the relative abundances of bacterial taxa in microbiome samples. Variation in sequencing depth across samples could lead to varying accuracy in the estimation of abundances of bacterial taxa across samples, potentially shaping results of the FAVA framework. However, when subsampling reads from each microbiome sample and recomputing FAVA on the subsampled datasets, we find that FAVA values are largely unchanged, so that the sequencing depth is likely sufficient for their accurate estimation ( SI Appendix , Fig. S8 ). Our framework, which we have implemented in an R package, contributes to a large body of methods for the analysis of microbiome relative abundance data ( 30 , 31 ). We emphasize, however, that the FAVA framework is a multisample compositional variability measure, setting it apart from the many existing measures of pairwise compositional similarity, such as Unifrac, Bray–Curtis dissimilarity, and the Jensen–Shannon divergence ( Fig. 1 A ) ( 26 , 27 , 45 ). For example, two separate collections of microbiome samples can have identical values of FAVA, but wildly different mean compositions (e.g., Fig. 4 B and C ). Similar results in the FAVA framework therefore reflect similarities in the spatial or temporal dynamics shaping variability, not compositional similarity. The FAVA framework complements diversity statistics such as the Gini-Simpson index, which summarize the diversity of taxa present in each sample rather than the variability of taxa across samples. For example, in the ruminant analysis, the Gini-Simpson diversity generally increases from the beginning to the end of the gastrointestinal tract, whereas FAVA peaks in the small intestine ( SI Appendix , Fig. S9 ). The FAVA framework builds on a rich literature of frameworks for hierarchical partitioning of genetic, taxonomic, and phylogenetic diversity across individuals and communities ( 46 – 50 ); indeed, F ST has sometimes been used as a measure of compositional variability in ecological contexts ( 51 ). Future applications of the FAVA framework can span the range of questions that researchers pose about compositional variability, from understanding temporal variability in infant microbiomes ( 52 , 53 ) to quantifying the repeatability of community assembly across experimental replicates to identifying the timing of compositional stability in serial passaging experiments ( 9 , 12 ). Because the FAVA framework measures a fundamentally different phenomenon relative to existing methods for microbiome analysis, it can facilitate studies of previously challenging research questions relating to temporal stability, individual heterogeneity, spatial variability, and replicability."
} | 2,790 |
30538275 | PMC6461840 | pmc | 6,955 | {
"abstract": "Many trees form ectomycorrhizal symbiosis with fungi. During symbiosis, the tree roots supply sugar to the fungi in exchange for nitrogen, and this process is critical for the nitrogen and carbon cycles in forest ecosystems. However, the extents to which ectomycorrhizal fungi can liberate nitrogen and modify the soil organic matter and the mechanisms by which they do so remain unclear since they have lost many enzymes for litter decomposition that were present in their free-living, saprotrophic ancestors. Using time-series spectroscopy and transcriptomics, we examined the ability of two ectomycorrhizal fungi from two independently evolved ectomycorrhizal lineages to mobilize soil organic nitrogen. Both species oxidized the organic matter and accessed the organic nitrogen. The expression of those events was controlled by the availability of glucose and inorganic nitrogen. Despite those similarities, the decomposition mechanisms, including the type of genes involved as well as the patterns of their expression, differed markedly between the two species. Our results suggest that in agreement with their diverse evolutionary origins, ectomycorrhizal fungi use different decomposition mechanisms to access organic nitrogen entrapped in soil organic matter. The timing and magnitude of the expression of the decomposition activity can be controlled by the below-ground nitrogen quality and the above-ground carbon supply.",
"introduction": "Introduction A large portion of nitrogen (N) in forest soils is found in organic form, primarily as amides and amines, but also as heterocyclic-N molecules [ 1 ]. These N molecules are associated with polyphenols, polysaccharides, lignin residues, lipids, and other degradation products of plant and microbial origin that are present in the soil organic matter (SOM) [ 2 ]. The capacity of forest trees to assimilate organic N is limited [ 3 ]. Access to organic N sources, such as proteins or chitin, requires decomposition to make the organic N molecules accessible and, subsequently, to liberate N from those molecules. Plants are generally thought to depend on microbial decomposition to access the soil N [ 4 ]. A long-standing hypothesis proposes that the ectomycorrhizal (ECM) fungal symbionts have a key role in this process [ 5 , 6 ]; however, the extent of the involvement of ECM fungi in SOM decomposition and mobilization of N compounds is debated [ 7 , 8 ]. ECM fungi evolved several times from saprotrophic ancestors [ 9 , 10 ]. These ancestors probably utilized diverse decomposition strategies resembling those seen in white-rot (WR) wood decayers, which use an enzymatic system for the decomposition of lignocellulose; brown-rot (BR) wood decayers, which utilize a two-step mechanism involving hydroxyl radicals (∙OH) generated by Fenton chemistry and hydrolytic enzymes [ 11 ]; and litter decomposers, which presumably use enzymatic decomposition systems similar to those of WR fungi [ 12 ]. During the transition from saprotrophic to symbiotic lifestyle, ECM fungi lost a large number of plant cell wall-degrading enzymes (PCWDEs) [ 9 , 10 ]. The convergent gene losses in relation to PCWDEs seen in ECM lineages have been used as an argument against a major role of ECM fungi in SOM decomposition [ 8 ]. However, ECM lineages have lost many, but not all, genes coding for PCWDEs, with diverse types and numbers of genes related to decomposition retained across lineages [ 10 ]. The high variability of the retained PCWDE-coding genes and the diverse evolutionary backgrounds of ECM lineages suggest that ECM fungi could have also retained and adapted some features of the decomposition mechanisms to the symbiotic lifestyle [ 13 ]. To what extent ECM fungi make use of the remaining decomposition systems is not well understood. At least some ECM fungi oxidize organic matter in a SOM extract in the presence of an energy source (i.e., glucose) [ 13 , 14 ]. Furthermore, ECM Cortinarius species encode ligninolytic class-II peroxidases, whose gene transcription correlates with the peroxidase activity in the boreal forest soil, supporting the hypothesis that these species may play an important role in SOM decomposition [ 15 ]. In addition, decomposition activities of ECM fungi have been inferred by ecological studies that relied on enzymatic assays detecting the activity of various hydrolytic and oxidative PCWDEs [ 16 , 17 ]. However, one limitation of such studies is that the assays are performed with ECM root tips and not the mycelium colonizing the soil substrate. Several of those assays, in particular the ones probing for oxidative enzyme activity, are unspecific and do not properly capture the decomposition activity [ 18 ]. Alternative hypotheses for the role of the detected enzymes include the decomposition of dead root tips [ 18 ] and remodeling of the root cell wall during host colonization [ 9 , 19 , 20 ]. Additionally, the environmental cues that regulate SOM decomposition in ECM fungi are not known. Laboratory experiments revealed that the oxidative decomposition system in the ECM fungus Paxillus involutus is expressed only in the presence of an energy source (i.e., glucose) [ 21 ]. In contrast, field studies based on enzyme assays suggest that ECM fungi can produce PCWDEs and metabolize SOM when the amount of carbon (C) supplied by the host plant is low [ 16 ]. Moreover, it is not known if the expression of the decomposition system of ECM fungi and the liberation of organic N compounds are concurrent. If so, the two processes might be regulated in conjunction and by similar nutritional signals, including the availability of inorganic and organic N sources. To address these questions, we used time-series spectroscopy and transcriptomics to analyze two species of ECM fungi with independent evolutionary histories and contrasting growth characteristics. P. involutus is characterized by a rapidly growing mycelium, and a so-called long-distance exploration type [ 22 ]. The species is nested within a paraphyletic assemblage of BR wood decayers in the Boletales [ 10 ], and oxidizes SOM using a nonenzymatic Fenton-based system [ 14 , 23 ]. By contrast, Laccaria bicolor develops a slow-growing, medium-distance smooth exploration subtype mycelium [ 22 ]. It belongs in the Agaricales and probably evolved from litter-decomposing saprotrophs [ 10 ]. The L. bicolor set of enzymes presumably involved in the degradation of PCW derived polymers is larger than that of P. involutus ; several of these enzymes are expressed during growth on a SOM extract [ 10 , 13 ].",
"discussion": "Discussion As demonstrated by recent studies, SOM extracted from the humic soil layer of a Norway spruce stand using hot water contains all major classes of biomolecules found in bulk SOM [ 13 ]. P. involutus and L. bicolor , when supplemented with glucose, have the capacity to decompose SOM by relying on oxidative mechanisms [ 13 ]. Here, using time-series spectroscopy and the same type of SOM extract, we show that SOM oxidation is linked to the liberation of organic N from SOM (Figs. 1 and 2 ). In both fungi, SOM oxidation was initiated when the readily available N source, i.e., ammonium, had been depleted. Following oxidation, organic N sources were utilized. In agreement with the findings of a previous study [ 21 ], the increased level of oxidation products declined when the fungi experienced C limitation. Therefore, despite their different evolutionary histories and foraging strategies, we observed functional convergence between the two species at the level of both SOM decomposition processes (SOM oxidation and liberation of organic N) and the regulation of SOM decomposition by similar nutritional signals, including limitation of inorganic N sources, presence of organic N source(s), and access to an energy source, e.g., glucose. Sequential assimilation of N sources is common in fungi, with the ammonium utilized before other, less preferred, N sources [ 56 , 57 ] and this was observed in our study as well. To examine whether P. involutus and L. bicolor sensed ammonium depletion, as has been observed in other fungi, we analyzed the regulation of genes homologs to ones that are upregulated during ammonium limitation in S. cerevisiae . Although most of those genes are present in the genomes of P. involutus and L. bicolor , and similar number of genes related to N-assimilation and N-metabolism were upregulated in both species, the types of upregulated genes differed markedly (Fig. 3 ). One gene that was upregulated in L. bicolor encoded an ammonium permease; this gene was located in a clade of fungal permeases, including Gap1 (Fig. S7 ). In S. cerevisiae , Gap1 is up-regulated during ammonium limitation and acts as a transceptor, i.e., both as a transporter and a receptor sensing the presence of amino acid substrates [ 58 ]. P. involutus and other related species from the Boletales examined here lack sequences in this clade. The genes upregulated during ammonium limitation in P. involutus included ones encoding inorganic N transporters and a Dur3 homolog ( PiDur3 ) [ 59 ], which encodes a plasma membrane transporter of urea and polyamines in S. cerevisiae . The data presented in the current study indicated that SOM oxidation by P. involutus and L. bicolor proceeded in conjunction with hydrolysis. However, the components of the decomposition mechanisms and their regulation were distinct and different in the two fungi. The data suggested that SOM decomposition by P. involutus is a two-step mechanism of oxidation and hydrolysis, controlled by N limitation and C limitation, respectively. By contrast, SOM decomposition by L. bicolor is a one-step mechanism that involves a combined activity of oxidative and hydrolytic enzymes triggered by N limitation and sustained during C limitation. Reduced iron (Fe 2+ ) is required for the generation of hydroxyl radicals (∙OH) in the Fenton reaction [ 11 ]. Such iron was detected in the organic matter extract after incubation with P. involutus at the onset of SOM-oxidation ( t 2 ) using X-ray absorption spectroscopy. This coincided with the expression of genes encoding enzymes involved in the biosynthesis of the Fe 3+ -reducing metabolite involutin, InvA5 and InvD [ 55 ]. At this stage of SOM decomposition, only a small number of genes encoding extracellular enzymes were upregulated in P. involutus . However, at later time points and during CD, P. involutus expressed a larger number of extracellular hydrolytic enzymes, including proteases, chitinases, oxidases, and glycoside hydrolases (Fig. 4C ). These genes were upregulated together with several genes associated with autophagocytosis, suggesting that P. involutus was undergoing C starvation response involving mycelial autolysis [ 43 ]. At the same time, the fungal biomass declined (Fig. 1A ). Although some of the upregulated genes coded for extracellular enzymes that were probably involved in the degradation and assimilation of released cellular material, others (GH10, GH15, GH31, GH5_5, and LPMOs) most likely encoded enzymes involved in the decomposition of the PCW-derived material present in the SOM extract. The above observations suggest that ∙OH generation is to some extent temporally separated from the synthesis of extracellular (both oxidative and hydrolytic) enzymes. Evidence of such a temporal separation of the production of ∙OH and proteolytic enzymes was recently presented in a study examining the protein decomposition by P. involutus [ 60 ]. A temporally separated two-stage oxidation-hydrolytic mechanism was recently shown to be utilized during wood decay by the BR fungus Postia placenta [ 61 ]. In contrast with P. involutus , increased Fe 2+ levels were not observed in the SOM extract of L. bicolor at any stage of the incubation. This suggested that the mechanism of SOM decomposition utilized by L. bicolor probably does not involve a nonenzymatic ∙OH oxidation. Furthermore, the L. bicolor biomass did not decrease and autophagy-related genes were not upregulated during C depletion, suggesting that the two fungal species respond differently to C starvation cues. Unlike in P. involutus , ND induced the expression of a number of L. bicolor genes associated with the decomposition of PCW derived polymers by saprotrophic fungi, including ones encoding LPMOs and CRO1, and genes associated with the decomposition of microbial products. Homologs of these genes were either lacking or not up-regulated in P. involutus (Fig. 3 ). Moreover, in L. bicolor , during C starvation, the expression of SOM decomposition genes upregulated during ND increased further, and a small set of additional genes, potentially also involved in SOM decomposition, was upregulated (Fig. 4C ). The genes that responded to both ND and CD, or only to CD, included ones that are typically involved in the decomposition of PCW polymers by saprotrophic fungi, including pectin (GH28 and CE8), cutin (CE5), cellobiose (GH3), starch (GH13_1), small polysaccharides (AA7) [ 11 ], and cellulose (AA9), as well as the fungal cell wall component chitin (CE4). These genes also included ones coding for oxidative enzymes, such as DyP and HalPrx, which have been suggested to act on lignin-like compounds and participate in detoxification processes [ 62 ]. The observation that a gene of a hexose import transporter LbMST1.3 [ 54 ] was upregulated together with several PCWDE genes in L. bicolor (Fig. 3 ) suggests that this fungus indeed possesses some capacity to assimilate C released during SOM decomposition. Collectively, these data indicated that immediately after the onset of ND, a number of genes related to the enzymatic decomposition of PCW and microbial polymers were upregulated in L. bicolor , and continued to be upregulated during C depletion. Studies have suggested that some of the remaining PCWDE genes seen in ECM fungi have been recruited for the modification of the PCW of the host during mycorrhizal formation [ 19 , 20 ]. Here, we provide evidence that at least some PCWDE genes are upregulated in the absence of a host during N and/or glucose limitation as part of the SOM decomposition mechanisms of ectomycorrhizal fungi. This suggests that while during evolution of the ectomycorrhizal lineages the PCWD machinery got reduced [ 10 ], the remaining genes could have been either incorporated into the SOM decomposition mechanisms of ECM fungi or recruited as PCW modifying genes during mycorrhizal formation. Such a diversification of functions could happen even within a gene family. This hypothesis is supported by the fact that the endoglucanase GH5-5 gene from L. bicolor that was shown to participate in the remodeling of the PCW during mycorrhization [ 20 ] was expressed at low levels and not significantly regulated in our experiments. The decomposition mechanisms of P. involutus and L. bicolor are distinctly different, in agreement with their diverse evolutionary origins, i.e., a BR wood decayer vs. litter-decomposing fungus [ 10 ]. In spite of these differences, the action of both mechanisms is controlled by N and C availability. This further suggests that the availability of photosynthetic products along with the type of available N in soil could act as a dual control over the decomposition activities of ECM fungi. That the host plant might actively control the decomposing activities of ECM fungi by controlling the amount of photosynthetic carbon provided to the fungus is suggested by soil microcosm experiments with Pinus sylvestris seedlings and ECM fungi including P. involutus . 14 C pulse labeling of the seedlings showed that the amount of plant C allocated to the fungal mycelium was high at the early phase of colonization of litter patches, but the C flow dropped when these areas were fully colonized and the available nutrients were assimilated [ 63 ]. Alternatively, seasonality might control the decomposition activities of ECM fungi as suggested in field studies using enzyme assays [ 16 , 64 ]. Further studies are needed to examine how the decomposition activities of ECM fungi are regulated when the fungi are growing in the field and in association with their plant host. Data generated from experiments in pure culture system as used in this study will provide the tools including the molecular biomarkers that could accurately predict the decomposition activities of ECM fungi in situ. Such markers must capture the action of both enzymatic and nonenzymatic (i.e., Fenton-based) reactions. The impact of ECM fungi on soil carbon cycling remains controversial. Genomic comparisons have suggested that the decomposition potential of ECM fungi is much smaller than that of their saprotrophic ancestors, based on the gene losses for PCWDEs seen in ECM lineages [ 8 , 10 ]. However, our results suggest that under nitrogen limitation ECM fungi oxidize SOM, while under nitrogen and/or carbon limitation some of the remaining PCWD related genes are upregulated. Whether these genes are used to only modify the SOM in order to further access entrapped nitrogen sources [ 65 ] or to release metabolic C is not clear, but the upregulated sugar transporter in L. bicolor suggests that some of this C can be assimilated by the mycelium. The impact of those PCWD related genes on C cycling processes might be considerable, particularly in deeper soil horizons where ECM species dominate [ 23 ]. Furthermore, ECM species might have an indirect impact on soil C cycling by affecting the availability of N. The upregulation of genes involved in the decomposition of proteins and other microbial-origin N compounds suggest that ECM fungi are able to access, decompose and assimilate organic N entrapped in SOM compounds. By doing so, ECM fungi may induce or potentiate N limitation of free-living, saprotrophic microbial decomposers, which may impede the soil C cycling and increase soil C storage [ 66 , 67 ]."
} | 4,495 |
30962472 | PMC6453922 | pmc | 6,957 | {
"abstract": "Marsh edge retreat by wave erosion, an ubiquitous process along estuaries, could affect vegetation dynamics in ways that differ from well-established elevation-driven interactions. Along the marshes of Delaware Bay (USA) we show that species composition from marsh edge to interior is driven by gradients in wave stress, bed elevation, and sediment deposition. At the marsh edge, large wave stress allows only short-statured species. Approximately 17m landward, decreasing wave stress and increasing deposition cause the formation of a ridge. There, high marsh fugitive and shrub species prevails. Both the marsh edge and the ridge retreat synchronously by several meters per year causing wave energy and deposition to change rapidly. Yet, the whole ecogeomorphologic profile translates landward in a dynamic equilibrium, where the low marsh replaces the high marsh ridge community and the high marsh ridge community replaces the mid-marsh grasses on the marsh plain. A plant competition model shows that the disturbances associated with sediment deposition are necessary for the high marsh species to outcompete the mid-marsh grasses during rapid transgression. Marsh retreat creates a moving framework of physical gradients and disturbances that promote the co-existence of over ten different species adjacent to the marsh edge in an otherwise species-poor landscape.",
"conclusion": "Conclusions Waves are agents of recurrent catastrophic (vegetation removal) and non-catastrophic (vegetation stress) disturbances. In Delaware Bay, energetic waves and sand availability on the intertidal flat create sharp gradients in physical-sedimentary disturbances within 20 m of the marsh edge. These gradients structure the vegetation community into distinct zones that differ from the gradual elevation-driven zonation exhibited in the marsh interior. An abrupt shift in species composition occurs where bed stresses reduce wave energy and sediments fall out of the water column. There, stress-tolerant species shift to competitive high marsh species. The reduction in the plant adaption time associated with the deposition disturbances allow the high-marsh plant community to colonize the ridge during the relatively short time period between formation and destruction of the ridge. Sea-level rise and marsh edge erosion have the potential to cause net marsh loss as well as replacement of high marsh habitat with low marsh habitat, especially where the marsh is subject to coastal squeeze and is impeded to transgress inland. This loss of high marsh habitat reduces the overall diversity and complexity of the marsh vegetation community, potentially leading to negative effects on wildlife species that depend on high marsh habitat (e.g.) 48 . In Delaware Bay, edge erosion is occurring more rapidly than landward transgression of the marsh into upland habitat 49 , indicating that high marsh species at the upland border may be lost. Here, we show that the presence of ridge near the marsh edge creates a novel high marsh habitat in an otherwise low marsh landscape. The disturbance associated with wave deposition allows rapid colonization of fugitive and high marsh species, allowing this community to transgress landward in dynamic equilibrium with the rapidly moving landscape. Thus, while the process of edge erosion is causing a net marsh loss, a high marsh habitat is created and maintained, providing a diverse and accessible habitat for wetland biota.",
"introduction": "Introduction Coastal marshes are a transitional ecosystem between the land and the sea, providing a number of important ecological services including habitat for ecologically and economically important species, sequestration of carbon and other pollutants, and coastal protection 1 . In a regime of rising sea-level and anthropogenic modifications, however, marshes are increasingly experiencing deterioration and loss. Even in the absence of sea-level rise, mature salt marshes, which tend to slope steeply to the intertidal flat, are being lost due to inherent lateral retreat by wave-induced edge erosion 2 , 3 . As marshes retreat laterally, low marsh plant species are predicted to migrate landward at the expense of mid- and high marsh species, particularly in areas with barriers to inland migration 4 , 5 . Loss of high marsh species at the upland boundary can significantly lower overall vegetation diversity and complexity, as low elevations tend to be dominated by one or a few stress-tolerant species. Coastal wetlands are ideal systems to test hypotheses of drivers of plant community patterns due to strong gradients in abiotic conditions, relatively low species richness, and striking plant zonation 6 – 8 . Plant community dynamics in salt marshes have been determined primarily for the relatively stable marsh interior, where discrete disturbances are infrequent. In the marsh interior, competitive interactions tend to increase as tidal flooding and salinity decline, generally regulated by elevation 8 – 10 . Less competitive, stress-tolerant species, such as Spartina alterniflora , are relegated to frequently-flooded low marsh elevations by competitive exclusion by mid-marsh species such as Spartina patens 8 , 10 . Similarly, high marsh shrubs are limited by the physical stresses at lower elevations and outcompete marsh grasses in areas of reduced flooding 11 , 12 . Therefore, sea-level rise and increased inundation generally causes the replacement of mid- and high marsh species with low marsh grasses 13 , 14 . An exception to the paradigm of elevation-driven competition might be present at the marsh-estuary boundary, where physical processes due to waves create frequent disturbances that may affect plant community structure and species interactions 15 . Waves directly shear the vegetation and rework the soil surface, an effect that is greatest at the marsh edge and declines exponentially inland 16 . Waves also have an indirect effect of depositing sediments and plant debris (i.e., wrack), which can smother existing plants 17 , 18 and provide opportunities for rapid colonization by fugitive and high marsh species sensu 8 , 19 . Along these high-energy marsh shorelines, frequent wave-induced disturbances along with gradients in elevation may control vegetation patterns, whereby community structure and species diversity depend on the frequency and intensity of disturbances, species-specific growth rates, and niche preferences 20 . These interactions, however, have yet to be described or modeled for high-energy marsh edge environments. At the estuary-marsh transition, physical disturbances are also compounded by a retreating marsh edge boundary, which forces the vegetation to rapidly adapt to changing physical conditions. This setting contrasts with the channel-interior marsh boundary, which remains relatively stable over time and only varies gradually with sea-level rise. Given that rates of marsh edge retreat can be up to 10 meters per year 21 , 22 present study (Table S1 ), it is unclear whether the vegetation near the marsh edge is simply lost in succession or adapts. With the inherent retreat of marsh boundaries when sediment input is less than what is eroded 2 , 3 , characterizing linkages between vegetation patterns and underlying morphodynamics is essential for a mechanistic understanding of succession and species movement patterns in rapidly changing estuary-marsh landscapes. With this study, we aimed to examine how intense physical drivers interact with biotic interactions to influence vegetation structure (i.e., species composition, richness, height, and density) along a rapidly retreating the marsh-estuary boundary. A simple plant competition model was developed to explain field observations that included both physical disturbances and interactions among species. Previous models have simulated competition between multiple salt marsh plant species 23 , 24 , but have not accounted for physical disturbances other than those associated with inundation. In addition, previous plant competition models 23 – 25 have assumed that the parameters driving the stochastic vegetation dynamics adapt instantaneously to the relative fitness, an assumption that might not hold where ecotones are migrating at extremely fast rates (e.g., >10 m yr −1 ).",
"discussion": "Model Results and Discussion The model recreates a bulldozer-like effect, where the whole edge-platform-ridge profile migrates landward in a dynamic equilibrium, similar to a previous model for eolian dune migration 30 . The rapid transgression of marsh topography requires quick adaptation of the vegetation to preserve the floral zones as a cohesive unit e.g., 15 . In both the low marsh and the marsh plain, the vegetation matches the local fitness: the low marsh is dominated by S. alterniflora as waves remove stress-intolerant plants, whereas the marsh plain is dominated by S. patens , which has a competitive advantage over stress-tolerant species (Fig. 4D ). Noticeably, the vegetation on the ridge deviates from the local fitness, i.e., S. patens dominates over the shrub species despite the latter having higher fitness at that elevation. This arises from the time limitation for shrubs to colonize the ridge when in competition with S. patens prior to experiencing the effect of edge retreat and thus succumbing to wave stresses (Fig. 4E ). In order for the shrub community to dominate on the ridge, the adaption time needs to be shorter, e.g., to 0.1 yr, thus allowing rapid colonization and establishment (Fig. 4E ). We suggest that this shorter adaption time is provided by the wave disturbance. Specifically, we suggest that sediment deposition eliminates or reduces interspecific competition with dense perennial grasses, thus allowing new high marsh species to colonize the ridge more readily. Indeed, newly deposited sandy sediment provides a well-drained substrate which can be readily colonized by high marsh plants, while wrack smothers existing plants and traps additional sediments before colonization. Multiple processes including vegetative colonization, stochastic dispersal, seedling establishment, and competition must occur on time-scales shorter than ridge migration. In general, disturbances tend to favor colonization by seed 31 and create increased resource availability (e.g., higher light levels) favorable to seed germination and seedling emergence for many species 32 , 33 . Species such as P. australis , an invasive grass, often found at the high marsh – upland transition, readily exploits these disturbances due to its prolific seed production 34 , widespread seed dispersal 35 , and persistence in the seed bank 36 . Similarly, the spread of P. australis both along the shoreline 37 and in the high marsh 19 has been attributed to sediment and/or wrack deposition that elevate the substrate above mean high water 38 . These model results agree with previous studies that have shown that sediment deposition events can stimulate low marsh grass productivity only up to an approximate threshold depth of 30 cm; above this threshold mortality occurs thus allowing the colonization of high marsh species via seed dispersal and seedling recruitment 39 , 40 . Noticeably, the effect of wrack deposition on the ridge differed from what is commonly observed in the marsh interior. Wrack deposited in the marsh interior smothers extant vegetation and can trigger a loss of elevation and soil carbon 41 . Following eventual wrack decay, vegetation succession may begin with the initial colonization of salt tolerant species 8 or, alternatively, ponds may form 42 , 43 . On the contrary, at the marsh edge, wrack was generally admixed with sand deposits resulting in an alternative ecological trajectory where the ridge substrate provided relatively porous, aerated conditions that promoted the rapid decay of wrack and colonization of high marsh species. Thus, the combined effect of sand and wrack deposition on the ridge was to accelerate colonization by fugitive and high marsh plants. Finally, even in the presence of disturbances (i.e., short adaption time), very fast rates of marsh retreat (20 m yr −1 ) resulted in lower shrub biomass on the ridge and lower S. alterniflora biomass in front of the ridge (Fig. 5 ). This result agrees with the field observations of low vegetation structure in areas with the fastest edge retreat rates (Fig. S6 ). This finding also supports the idea that in the presence of fast changing physical conditions, plant biomass (or the expected value of the biomass) is not necessarily in equilibrium with the local fitness as assumed in previous models 23 . For example, the model predicts that for extremely fast rates of marsh edge retreat (30 m yr −1 ), high marsh species would not be able to colonize the ridge. Figure 5 Comparison of the plant distribution for different marsh edge retreat rates (with the same adaption time equal to 0.1 years). ( A ) Retreat rate of 1 m/yr; and ( B ) Retreat rate of 20 m/yr. Note that in the former case the shrub vegetation on the ridge and the S. alterniflora in front of the ridge are higher than in the latter case. Parameterizing the effect of disturbances by modifying the adaption time is a simple and efficient approach to simulate complex ecological interactions. This approach might complement the “windows of opportunity” approach, which recently has been successfully applied to simulate the encroachment of Spartina on the bare mudflat facing the marsh edge 44 . The adaption time approach can be useful in other ecotones with different types of competing plant species (e.g., marshes vs mangroves, or mangroves vs terrestrial vegetation) and different physical disturbances (e.g., episodic flooding, salinity, droughts, fires, and diseases). Importantly, our observations and model results support existing theories explaining patterns of species richness and niche separation. Low species richness or outright mortality is predicted to correspond to a high frequency and intensity of disturbances, particularly where environmental stress is also high 45 , similar to conditions at the marsh edge. Wave disturbance weakens and abiotic stress declines with increasing distance from the marsh edge, increasing the potential for species interactions and increasing the number of possible species. In the absence of sediment deposition and ridge formation, which is predicted by the model in the case of β equal to zero (i.e., in the absence of sand re-deposition), low marsh species would directly transition to mid-marsh grasses as the result of competitive exclusion. Model results contribute to mounting support that disturbances increase the rate of competitive exclusion e.g., 46 , in this case, by allowing rapid colonization and competitive exclusion of ridge species over mid-marsh grasses. The maintenance of these vegetation zones, albeit in non-equilibrium with the local fitness as the marsh edge and ridge migrate landward, is a function of metacommunity dynamics where localized disturbance, dispersal ability, and rapid growth rates facilitate the near-continuous re-colonization of the retreating marsh edge and ridge. The ridge community is comprised of fugitive species, which are poor competitors but efficient colonizers, as well as high marsh species, which are competitively-dominant. Overall, the combination of frequent disturbances and a strong abiotic gradient (i.e., elevation) provides temporal niche opportunities for spatial differentiation among species creating a relatively diverse co-existence of species in an otherwise species-poor environment 47 ."
} | 3,900 |
29109975 | PMC5665595 | pmc | 6,958 | {
"abstract": "Population genomic simulations predict coral adaptation only under mitigated climate change scenarios.",
"introduction": "INTRODUCTION Predicting future species response to climate change requires detailed knowledge of the link between organismal success and environment. Under most predicted scenarios, local temperatures will surpass current thermal tolerance limits of many species ( 1 ). However, many traditional climate envelope models for predicting species responses to climate change assume that thermal limits are static over time ( 2 ). Although more complex models have attempted to integrate variation in tolerance in the form of plasticity or evolutionary adaptation into estimates of vulnerability to climate change ( 3 , 4 ), these models are often hindered by a dearth of information about the genetic underpinnings of climate-related traits, which often define the rates and limits of adaptation. Increasing observations suggest that genetic variation associated with thermal tolerance could provide the raw material necessary for adaptation to increasing temperatures ( 5 , 6 ). Population genomic studies from a wide array of taxa have consistently found allelic variation associated with climate traits ( 7 – 9 ). These genetic variants tend to be numerous and distributed across a broad array of gene functions, suggesting that thermal tolerance may be a cumulative product of many small-effect loci ( 10 , 11 ). Because thermal tolerance is often not determined by one or a few loci of large effect and thus is unlikely to be fixed between populations, most populations likely contain some standing variation in genetic thermal tolerance. As climate change proceeds, natural selection might increase the frequency of these thermal tolerance genotypes through differential survival and propagation of preadapted genomes. The main questions that revolve around this process are whether the rates and limits of adaptation are high enough to (i) assure a population’s survival, (ii) delay its environmental demise, or (iii) neither. In the ocean, reef-building corals are among the organisms most vulnerable to rising temperatures. Although corals inhabit a wide variety of environments, they are thought to live at just 1° to 2°C below their upper thermal limits ( 12 ). Despite this, a number of studies have provided evidence for adaptation of coral populations to local temperatures ( 13 , 14 ) and for heritable variation in thermal tolerance in corals ( 11 , 15 ). We capitalize on a previous study ( 11 ) on the coral Acropora hyacinthus in which we identified alleles at many loci associated with thermal tolerance across a temperature gradient in American Samoa. Here, we focused on a high-latitude population of the same species in Rarotonga, Cook Islands, asking whether this population harbored the same warm-adapted alleles found in our previous study, thus providing the potential for adaptation of this population to warmer temperatures. Next, we used a coupled environmental, evolutionary, and demographic simulation to predict survival or extirpation under future climate change scenarios. Finally, we tested whether migration of heat-tolerant genotypes, for example, through outplanting, could increase the likelihood of population persistence. Our results are preliminary because our knowledge of genotype-phenotype effects and population dynamics in this system is imperfect. Nevertheless, we develop a general framework by which questions of future evolutionary rate can be answered and show that evolution in this population can substantially increase future survival.",
"discussion": "DISCUSSION Our simulations of one high-latitude coral population show that the rate and limits of adaptation to temperature may be sufficient to prevent the extinction of some populations under mild future emissions scenarios. On the basis of the parameters in our models, increased temperature leads to selection for higher frequencies of warm-adapted alleles. When increases in temperature are slow enough, the demographic deficit caused by selection is made up by increased population fitness and higher population growth rates. By contrast, in our model, more severe emissions outpace the evolutionary response, and mortality from selection is high enough to drive population extinction. The tipping point for the switch from adaptation to extinction as temperature change quickens is likely to be system-specific, but the basic scenario we see in our model is likely to be more general. Our finding that faster warming in the future will likely decrease the probability of persistence is also seen in experimental studies of response to environmental stress in model systems such as Escherichia coli ( 20 ) and yeast ( 21 ). In these cases, rapid environmental change leads to adaptive collapse: Marked demographic decline as mortality increases, followed by increased loss of adaptive loci due to drift, and extinction before adaptation can restore higher fitness. In addition, models show that if the rate of environmental change outpaces adaptation, then populations become maladapted and unable to restore positive growth rates ( 20 ). A striking feature of our results is that the cool-water population in Rarotonga harbors alleles identified as warm-adapted in populations located 800 km away. Although determining whether this is a more general phenomenon will require more research on more species, there are several reasons to think that there is wide occurrence of warm-adapted alleles in marine taxa. Many marine species have long-distance dispersing larvae, connecting populations over large distances ( 22 , 23 ). These dispersal patterns tend to distribute alleles broadly, even as they allow buildup of frequency differences between locations. In other studies of adaptive divergence of long-distance dispersing marine populations, alleles are seldom private, and polymorphisms tend to be seen across large ranges ( 9 , 24 ). As a result, clines across environmental gradients tend to be shallower in marine species, reflecting the balance between dispersal and selection ( 22 , 23 ). Major assumptions in the type of simulation we conduct here involve robust estimations of the allele frequencies of adaptive loci, the genetic architecture of thermal tolerance, and the map between environmental and genetic traits. A highly polygenic basis of thermal tolerance has been found in a number of systems ( 7 , 8 ), including corals ( 15 ), suggesting that polygenic models are the most appropriate. However, some corals have specific loci with major effects on survival in the face of poor water quality ( 25 ) or sustained heat ( 26 ). For a precise estimation of genetic architecture, there is a need for further development of analytical tools, because current methods for detection of loci under selection can be riddled with false positives ( 27 , 28 ). In addition, because genetic architecture estimated from natural populations can reflect environmental parameters unrelated to the key question, in our case temperature, further studies across replicate populations are needed for a more precise estimation of architecture. Careful consideration of experimental design and parameter choice is therefore necessary. We show that a rough estimation of genetic architecture might suffice for these types of predictions, because there is little difference in model outcome when more than ~100 loci encode the thermal tolerance trait. However, discovery that future climate tolerance was enabled by a few genes of major effect would substantially change the model. Another area needing more information is the accurate estimation of phenotype, in our case, thermal tolerance, produced by different multilocus genotypes. We assume that there are many loci involved in heat tolerance and that all loci contribute equally without interactive effects, initial assumptions that are common in quantitative trait evolution. For example, in plant systems, numerous genetic variants predicted from genome scans across broad-scale contemporary environments can, together, predict performance under laboratory stress conditions ( 7 , 29 ). These high-throughput measurements of tolerance and fitness are more difficult in nonmodel systems. Future efforts should therefore focus on precise and quantitative phenotypic measures to refine the map between genotype, phenotype, and future environment. Future work should also focus on developing more realistic ecological scenarios within the simulation framework. Because we focused on a single isolated population, we used an extremely simple population genetic simulation with equal reproductive output from all surviving individuals, although fecundity can vary markedly and is influenced by factors such as depth and size ( 30 , 31 ). Other extensions could include intermediate-latitude populations, integration of dispersal and connectivity, and even ecological factors such as resource availability and competition [reviewed by Hoban ( 32 )] to create broader ecological frameworks that may be applicable to more species embedded in metapopulations. Although our simulation provides a simple framework in which to begin using genomic data to predict climate change response, there are many, more complex interactions that would be beneficial to integrate into future models [reviewed by Bay et al . ( 33 )]. The potential for plasticity to buffer short-term exposure to increased temperatures is well known for corals ( 14 , 34 , 35 ), but evidence in other systems shows that this plasticity can either promote or hinder adaptation ( 36 ). More careful work on dissecting the interaction between plasticity and adaptation of thermal tolerance limits in corals is needed to parameterize this model. Another potential component in coral thermal tolerance is the role of the algal symbiont (genus Symbiodinium ) ( 37 , 38 ). Future studies could pair our model for host adaptation with local acquisition of adapted symbionts and the potential for symbiont shuffling in response to increased temperatures ( 39 ). Finally, alleles associated with increased thermal tolerance could be maladaptive in other conditions. For example, Howells et al . ( 13 ) found that warm-adapted individuals from the northern Great Barrier Reef experienced bleaching and mortality during winter low temperatures when transplanted onto the southern Great Barrier Reef. Future incorporation of these and other considerations will improve our ability to accurately predict population-level responses to ongoing climate change in other systems. Despite these caveats, simple adaptive models of coral populations in different future scenarios have the benefit of showing the potential value of genetic diversity in current populations and of the impact of relaxing assumptions that tolerance traits are unchangeable in climate envelope predictions. Given the importance of adaptive landscapes in future climate responses in corals and other species, we used our integrated model to ask whether outplanting efforts could realistically increase the adaptive capacity of this high-latitude population. Adding a few corals with highly adaptive gene combinations could sometimes rescue the populations in our simulations. However, it is important to put the scale of outplanting in context. The level of assisted gene flow we simulated (10 individuals) is equivalent to replanting 1% of the population with adult, reproductive colonies every year. Accounting for expected levels of transplant mortality ( 17 ) and death of juveniles, this would require transplanting nearly 2 to 5% of the population annually until the year 2100. A bigger population than the small 1000 individual size we assume here would require proportionately more input. This would be a substantial annual effort with ongoing financial support for nearly a century. Although these simulations suggest that this effort could have important adaptive consequences, it is important to note that incoming individuals must be genetically preadapted to future scenarios: Adding large numbers of individuals from populations with few heat tolerance alleles did not affect the outcome. This suggests that extreme care must be taken when selecting individuals for outplanting. Logistic and precautionary considerations (such as introduction of invasive species or disease) strongly suggest that only local outplanting should be conducted among local reefs with different thermal adaptations. Overall, there are many scientific, legal, and ethical concerns that should be carefully considered for any relocation project ( 40 , 41 ). A major outstanding question in climate change biology is whether the rate of adaptation will keep up with the rate of environmental change. We show that the current adaptive inventory of one coral population in Rarotonga is sufficient for moderate but not strong climate change. A growing number of studies show standing genetic variation in climate-related traits in many species, often involving large numbers of loci ( 7 , 10 , 24 , 42 ). Other studies have become much better at predicting patterns of future environmental change on a local level. It may now be broadly possible to combine genomics and environmental modeling into a predictive framework that asks when and where evolution will be a major force in climate response."
} | 3,342 |
21907300 | null | s2 | 6,959 | {
"abstract": "Understanding in vivo regulation of photoautotrophic metabolism is important for identifying strategies to improve photosynthetic efficiency or re-route carbon fluxes to desirable end products. We have developed an approach to reconstruct comprehensive flux maps of photoautotrophic metabolism by computational analysis of dynamic isotope labeling measurements and have applied it to determine metabolic pathway fluxes in the cyanobacterium Synechocystis sp. PCC6803. Comparison to a theoretically predicted flux map revealed inefficiencies in photosynthesis due to oxidative pentose phosphate pathway and malic enzyme activity, despite negligible photorespiration. This approach has potential to fill important gaps in our understanding of how carbon and energy flows are systemically regulated in cyanobacteria, plants, and algae."
} | 208 |
39659428 | PMC11628275 | pmc | 6,960 | {
"abstract": "Event-based cameras are suitable for human action recognition (HAR) by providing movement perception with highly dynamic range, high temporal resolution, high power efficiency and low latency. Spike Neural Networks (SNNs) are naturally suited to deal with the asynchronous and sparse data from the event cameras due to their spike-based event-driven paradigm, with less power consumption compared to artificial neural networks. In this paper, we propose two end-to-end SNNs, namely Spike-HAR and Spike-HAR++, to introduce spiking transformer into event-based HAR. Spike-HAR includes two novel blocks: a spike attention branch, which enables model to focus on regions with high spike rates, reducing the impact of noise to improve the accuracy, and a parallel spike transformer block with simplified spiking self-attention mechanism, increasing computational efficiency. To better extract crucial information from high-level features, we modify the architecture of the spike attention branch and extend it in Spike-HAR to a higher dimension, proposing Spike-HAR++ to further enhance classification performance. Comprehensive experiments were conducted on four HAR datasets: SL-Animals-DVS, N-LSA64, DVS128 Gesture and DailyAction-DVS, to demonstrate the superior performance of our proposed model. Additionally, the proposed Spike-HAR and Spike-HAR++ require only 0.03 and 0.06 mJ, respectively, to process a sequence of event frames, with model sizes of only 0.7 and 1.8 M. This efficiency positions it as a promising new SNN baseline for the HAR community. Code is available at Spike-HAR++ .",
"conclusion": "5 Conclusion In this paper, we proprse an energy-efficient and lightweight Spike-HAR family for event-based human action recognition, to adaptively emphasize on local spatial features as well as temporal features. Spike-HAR and Spike-HAR++ surpass existing methods in accuracy on the SL-Animals-DVS, N-LSA64, DVS128 Gesture, and DailyAction-DVS datasets. Furthermore, Spike-HAR and Spike-HAR++ require only 0.03 and 0.06 mJ to recognize a single action event stream, reducing the power consumption of 99.27 and 98.55% compared to the Evt, respectively. It demonstrates the applicability of spiking transformers for human action recognition and their potential application in human-machine interaction and edge HAR devices. In the future, it is promising to develop a more complex large-scale event-based HAR benchmark to further evaluate the performance of the Spike-HAR family in practical applications.",
"introduction": "1 Introduction Human action recognition (HAR) involves identifying and understanding human movements and has numerous applications in the real world (Sun et al., 2022 ). For instance, HAR can be employed in visual surveillance systems to detect hazardous activities and monitor human behavior, thereby ensuring safe operations (Lin et al., 2008 ). Additionally, HAR can facilitate sign language recognition (SLR). According to the latest data from the World Federation of the Deaf, there are 70 million deaf individuals worldwide using over 200 sign languages (Murray, 2018 ). However, learning sign language can be challenging and time-consuming, creating communication barriers for the deaf community (Hu L. et al., 2023 ). To address this issue, HAR for sign language recognition has been extensively researched. Most of the works focused on using RGB or gray-scale videos as input for HAR (Wang et al., 2017 ; Kındıroglu et al., 2022 ; Vázquez-Enríquez et al., 2021 ; Mercanoglu Sincan and Keles, 2022 ; Shen et al., 2024 ; Wang F. et al., 2023 ), due to their popularity and easy access. However, the recognition results of RGB-based HAR methods are inevitably influenced by the motion blur inherent to RGB cameras and static background noise (Wang et al., 2019 ; Wang Y. et al., 2022 ). As an emerging neuromorphic sensor, the event camera detects changes in brightness for each pixel independently, generating an event stream asynchronously and sparsely. The difference between RGB video frames [from LSA64 (Ronchetti et al., 2023 )] and DVS event frames (from N-LSA64) is shown in Figure 1 . The event camera features high temporal resolution, low latency, low power consumption, and a wide dynamic range (Su et al., 2022 ), which can effectively address issues related to motion blur and static background noise. That is, event cameras hold significant advantages in the field of HAR. The current state-of-the-art (SOTA) approaches for event-based HAR involve firstly designing event aggregation strategies converting the asynchronous output of the event camera into synfirst chronous visual frames, followed by processing using Artificial Neural Networks (ANNs) (Ghosh et al., 2019 ; Amir et al., 2017 ; Baldwin et al., 2022 ; Cannici et al., 2020 ; Innocenti et al., 2021 ; Sabater et al., 2022 ), which require considerable computational power, posing challenges for deployment on edge devices. Figure 1 (a) Comparison of RGB video frames and DVS data frames for sign language Opaque (one-handed sign). (b) Comparison of RGB video frames and DVS data frames for sign language breakfast (two-handed sign). As third-generation neural networks, Spike Neural Networks (SNNs) are designed with biological plausibility, mimicking the dynamics of brain neurons to encode and transmit information in the form of spikes (Maass, 1997 ). Compared to ANNs, the event-driven nature of SNNs significantly reduces energy consumption when running on neuromorphic chips (Zhang et al., 2023 , 2021 ). However, current SNN-based HAR tasks still face challenges of lack of datasets and low recognition accuracy (Shi et al., 2023 ). In this paper, we propose two models, Spike-HAR and Spike-HAR++, to simultaneously reduce power consumption and enhance recognition accuracy in event-based HAR. Spike-HAR integrates a patch embedding (PE) block, parallel transformer blocks, a spike attention branch, and a classification head. To further improve performance, we modify the architecture and position of spike attention branch in Spike-HAR according to the Hu et al. ( 2024 ) and extend it to a higher dimension, proposing Spike-HAR++, which enables better extraction of crucial information from high-level features. As illustrated in Figure 2 , experiments on the SL-Animals-DVS dataset (Vasudevan et al., 2022 ) demonstrate that both models significantly outperform other event-based HAR systems while maintaining lower levels of power consumption. Figure 2 Accuracy vs. inference energy of different neural methods implemented in Intel Stratix 10 TX (Corporation, 2023 ) (for ANNs) or ROLLS (Qiao et al., 2015 ) (for SNNs). The size of the markers denotes the number of parameters. This paper is an extended version of our prior work (Lin et al., 2024 ) accepted by BMVC 2024. The main differences with the conference version are as follows: (1) besides the Spike-HAR based on the Parallel Spiking Transformer (referred to as Spike-SLR in the BMVC version), we newly propose Spike-HAR++, which is better suited for recognizing long-duration actions; (2) the application scope of the models are extended from sign language recognition to human action recognition, with comprehensive testing conducted on two additional datasets: DVS128 Gesture (Amir et al., 2017 ) and DailyAction-DVS (Liu et al., 2021 ), achieving SOTA performance; (3) a detailed overview about traditional ANN-based and SNN-based HAR methods, as well as the development of spiking transformers are discussed in the related work. To sum up, the main contributions of this paper are listed: (1) We propose the Spike-HAR family, i.e., Spike-HAR and Spike-HAR++, which mainly consists of a powerful parallel spike transformer block. To the best of our knowledge, it is the first spiking transformer specifically designed for event-based HAR. To enhance the model's spatio-temporal attention to fine-grained action features while maintaining energy efficiency and a lightweight design, we employ a parallel spiking transformer. In this architecture, multi-layer perceptrons (MLPs) and simplified attention sub-modules (CB-S3A) operate in parallel to improve overall efficiency. (2) We first introduce attention mask mechanisms into SNNs and incorporate a spike attention branch in our model to extract key regions from the input event streams. Additionally, we improve the attention operation for Spike-HAR++, utilizing high-dimensional features extracted through a patch embedding (PE) block to accommodate the recognition of long-duration actions. Experiments demonstrate that, although the parameter count and power consumption of Spike-HAR++ increase slightly, the accuracy of HAR improves significantly. (3) Experimental results on the public datasets SL-Animals-DVS (Vasudevan et al., 2022 ), N-LSA64 (Ronchetti et al., 2023 ) [converted using the v2e (Hu et al., 2021 ) method], DVS128 Gesture (Amir et al., 2017 ), and DailyAction-DVS (Liu et al., 2021 ) show that the proposed Spike-HAR family effectively balances model size and recognition accuracy. Specifically, the proposed Spike-HAR and Spike-HAR++ require only 0.03 and 0.06 mJ, respectively, to process a sequence of event frames, with model size of just 0.7 and 1.8 M. In the rest of the paper, Section 2 presents the related work on event-based HAR and spiking transformers. Section 3 begins with an overview of the overall architecture of Spike-HAR and Spike-HAR++, followed by a detailed description of each model component. Section 4 introduces four HAR benchmark datasets and evaluation metrics, along with rigorous ablation studies, visualizations, and performance evaluations of the proposed models. Finally, Section 5 concludes the paper."
} | 2,426 |
33121198 | PMC7693107 | pmc | 6,962 | {
"abstract": "Super-hydrophilicity is a desired but rarely reported surface finish of polymer materials, so the methods for achieving such a property represent a great scientific and technological challenge. The methods reported by various authors are reviewed and discussed in this paper. The super-hydrophilic surface finish has been reported for polymers functionalized with oxygen-rich surface functional groups and of rich morphology on the sub-micrometer scale. The oxygen concentration as probed by X-ray photoelectron spectroscopy should be above 30 atomic % and the roughness as determined by atomic force microscopy over a few nm, although most authors reported the roughness was close to 100 nm. A simple one-step oxygen plasma treatment assures for super-hydrophilicity of few polymers only, but the technology enables such a surface finish of almost any fluorine-free polymer providing a capacitively coupled oxygen plasma that enables deposition of minute quantities of inorganic material is applied. More complex methods include deposition of at least one coating, followed by surface activation with oxygen plasma. Fluorinated polymers require treatment with plasma rich in hydrogen to achieve the super-hydrophilic surface finish. The stability upon aging depends largely on the technique used for super-hydrophilization.",
"conclusion": "4. Conclusions Plasma methods for obtaining super-hydrophilic surface finish of polymer materials were reviewed and discussed. The straightforward technique is a treatment with oxygen plasma. The combination of oxygen-rich surface functional groups and roughness on the sub-micrometer scale enables super-hydrophilicity of any polymer providing the morphology and functionalization are adequate. So far, this technique was elaborated only for Poly(ethylene terephthalate) samples. A rather broad range of roughness as deduced from AFM imaging was reported to assure the super-hydrophilic surface finish. The effect, however, is of short duration since the hydrophobic recovery was reported by all authors. Typically, the superior effects are lost already several minutes after the plasma treatment. This technique is therefore useful only in cases when such a surface finish is needed for immediate deposition of a third coating. For example, before printing, gluing, or painting. The adhesion of any coating on a polymer treated using this method should be particularly good—any liquid (paint, glue, etc.) will enter the pores and interact chemically with the surface functional groups. A more sophisticated technique is a simultaneous etching of a polymer material and depositing an inorganic material. A feasible configuration is a plasma reactor that enables the deposition of small quantities of a reactive metal (such as aluminum) upon the treatment of a polymer sample with oxygen plasma. The technique was elaborated for few polymers but should apply to any fluorine-free polymer. Namely, the super-hydrophilicity using this method is due to the presence of clusters of metal oxides on the surface of extremely rough polymers. The adhesion of any coating should be equally good as in the case of metal-free surface finish, but a possible drawback of this technique is long treatment time. An alternative to such one-step methods is a deposition of various coatings of rich morphology followed by surface activation with oxygen plasma. Such treatments apply to any polymers, but the adhesion may be problematic since the coatings consist of nanoparticles that may not stick well either to the polymer substrate or any coating deposited after obtaining the super-hydrophilic effect.",
"introduction": "1. Introduction Products made from polymers or polymer composites are usually manufactured using extrusion or injection molding. The products will assume a rather smooth surface and the chemical nature of the polymer materials will enable moderate or poor wettability. Such a surface finish is often inadequate, in particular when the product should be coated with a third material, for example at gluing, printing, or deposition of a coating with specific functionality. The surface properties are modified by various techniques including treatment with aggressive chemicals, irradiation with beams of photons or charged particles, and mechanical treatments such as brushing or sandblasting. All these techniques enable modification of either the topology or the surface chemistry or both. Increased hydrophilicity is achieved by introducing polar functional groups on the polymer surface, and roughness is increased either by etching of the original material or deposition of a coating with rich morphology. The combination of the rich morphology on the sub-micrometer scale and polar surface functional groups will lead to a super-hydrophilic surface finish. Such a surface finish is probed by one of the simplest experimental techniques—deposition of a small droplet and measuring the contact angle between the substrate and the liquid. The super-hydrophilic surface finish will cause the spreading of a liquid of even high surface energy on a surface area much larger than the droplet diameter. The most commonly used liquid with high surface energy is distilled water. Furthermore, rough surfaces of super-hydrophilic materials favorite the capillary effect, so the water is distributed within the pores or channels over a large surface area. In any case, the water contact angle (WCA) is immeasurable low for polymer products of super-hydrophilic surface finish. The super-hydrophilicity is not as permanent, but the WCA slowly increases with storage time. The loss of hydrophilicity is usually called “hydrophobic recovery”. A popular technique for modification of surface properties of various polymers is a brief treatment with non-equilibrium gaseous plasma. Different aspects of plasma–surface interaction have been elaborated on by various authors and summarized in review papers such as [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ]. A polymer product is exposed to plasma which is a source of radiation (of particular importance is radiation in the ultraviolet (UV) and vacuum ultraviolet (VUV) range of photon energies), positively charged ions of different kinetic energies, and neutral reactive particles such as molecular fragments including neutral atoms. The surface finish depends on the treatment parameters. Different authors have used different plasmas, but only a few of them reported on the super-hydrophilic surface finish. This paper aimed to summarize recent achievements in this scientific niche and discuss the advantages and drawbacks of techniques that enable such a rarely-observed property of polymer materials.",
"discussion": "3. Summary and Discussions The literature survey indicates a couple of methods for plasma-stimulated super-hydrophilicity of polymer materials: one-step exposure to reactive plasma, typically oxygen; two- or several-steps involving at least one deposition of a coating with high roughness followed by surface activation with oxygen plasma. 3.1. One-Step Exposure to Oxygen Plasma The treatment of polymers with oxygen-containing plasma is a natural choice for surface hydrophilization since the oxygen-rich functional groups assure for high wettability of a polymer material. The super-hydrophilic surface finish, however, is achieved only by providing the plasma treatment also causes an appropriate roughness. Schematic of the one-step procedure for achieving super-hydrophilic surface finish is presented in Figure 1 . A polymer sample is subjected to radiation arising from transitions between O-atoms, which appear predominantly in the red and VUV part of the spectrum. The sample is also exposed to neutral atoms in the ground and metastable excited states, positively charged molecular and atomic ions, and metastable molecules ( Figure 1 a). The VUV radiation causes bond scission in the surface film [ 18 ], while the atomic and molecular species may interact chemically, causing both functionalization and etching. Exposure to reactive oxygen plasma species rarely leads to the super-hydrophilic surface finish of polymers. The reason is an inappropriate combination of roughness and functionalization. Although not often mentioned in scientific literature, the exposure to non-equilibrium oxygen plasma affects not only the very surface but also a thicker surface film. For example, Bruce et al. [ 32 ] exposed polystyrene to energetic ions and explained the evolution of the surface roughness by buckling instability. The surface layer became stiff upon ion treatment, and a large mismatch in the stiffness of the affected layer and bulk polymer wrinkles the surface layer to minimize the elastic energy as shown schematically in Figure 1 b. The explanation provided by Bruce et al. [ 32 ] takes into account a rather large kinetic energy of positively charged ions. In many cases, however, the polymer samples assume the floating potential, so the kinetic energy of ions is too low to trigger such an effect. The stiffness of the surface film may also be changed by other effects, for example, synergetic effects of VUV radiation and exothermic surface reactions, as explained by Lehocky et al. [ 33 ]. Another explanation for rich surface morphology upon exposure of polymer samples to oxygen plasma is the preferential etching of the amorphous phase as elaborated by Junkar et al. [ 34 ]. In any case, only a handful of polymers become super-hydrophilic upon exposure to oxygen plasma as in Figure 1 . Furthermore, the super-hydrophilic effect using the technique of Figure 1 is far from being permanent as a rapid hydrophobic recovery was reported by all authors using this technique. Fluorinated polymers will not become super-hydrophilic using this technique since the reactive oxygen species from plasma will not form polar oxygen-rich surface functional groups but will instead cause etching only. Such polymers are better treated according to the technique described by Nguyen and Yajima [ 30 ]. The hydrophobic recovery was suppressed using a capacitively coupled RF discharge. The experimental setup is shown schematically in Figure 2 . The powered electrode is much smaller than the grounded housing, so a strong DC self-biasing occurs [ 35 ]. A sheath forms spontaneously next to the powered electrode at such experimental conditions. The ions entering the sheath from bulk plasma are accelerated in the high electric field within the sheath and gain a rather large kinetic energy. They bombard the powered electrode and cause weak sputtering of the electrode material. The atoms sputtered from the powered electrode condensate on the polymer substrate, as shown in Figure 2 b. A thin film of the electrode material (often aluminum) is deposited on the polymer sample. If the polymer itself is biased (for example, by placing a piece of polymer on the powered electrode), the sample will be subjected to energetic oxygen ions causing etching of the organic compound. The electrode material will form clusters that will prevent chemical etching on the top so that etching will occur only within gaps between the inorganic clusters. The final effect will be nanostructuring, as shown in Figure 2 c. The clusters of oxidized electrode material will persist on the top of the polymer sample and assure a long-term super-hydrophilic surface finish, as elaborated by Gogolides’ group [ 8 ]. The surface finish depends predominantly on the properties of inorganic clusters rather than the polymer substrate, so this technique applies to practically any polymer. The adhesion between the polymer substrate and the inorganic clusters may not always be optimal. Another drawback may be excessive heating of the polymer sample upon bombardment with the positive ions. Fluorinated polymers may become super-hydrophilic using this technique, but the adhesion of the inorganic clusters is often inadequate. 3.2. Deposition of a Coating with High Roughness Followed by Surface Activation The drawbacks of the one-step technique are irrelevant when a two-step procedure is employed to obtain the super-hydrophilic surface finish of various polymers. The polymer material is first activated by a brief treatment in plasma of virtually any gases. The activation will assure for a certain concentration of polar functional groups that will, in turn, assure for good adhesion of a uniform coating. According to the state-of-the-art, a PDMSO-like coating is preferred. A widely used method for deposition of such coatings is an application of weakly ionized plasma sustained in HMDSO. The gaseous precursor partially dissociates upon plasma conditions, and the radicals stick to the surface of any material facing plasma and form a thin film containing carbon, hydrogen, silicon, and oxygen. The concentrations of these elements in the deposited film depend enormously on the experimental conditions [ 36 ]. The sample is then exposed to oxygen plasma to etch the organic component from the surface of the PDMSO-like coating leaving dense nanoparticles containing silicon oxides. A schematic of the procedure is shown in Figure 3 . Prior to the deposition, the surface of the polymer substrate should be free from impurities and preferably activated by brief plasma treatment. Such treatment is used routinely on an industrial scale [ 37 ]. Plasma could be sustained at a very low power density in different gases, including the residual atmosphere in the plasma reactor [ 38 ]. Namely, even weak functionalization of a fluorine-free polymer with polar functional groups will assure the desired adhesion of a thin film synthesized by plasma polymerization. The polymer samples are exposed to a variety of radicals presented in weakly ionized HMDSO plasma, as shown in Figure 3 a. Prolonged treatment (typically a few minutes) will enable the deposition of a PDMSO-like coating without affecting the properties of the substrate. Once the film of an appropriate thickness (often around 100 nm) is deposited, the samples are treated by oxygen plasma ( Figure 3 b). The reactive oxygen species will etch away the organic component of the PDMSO-like coating, thus forming densely distributed nanoparticles consisting predominantly of silicon oxides. The as-synthesized nanoparticles assure for the super-hydrophilic surface finish. This technique was used by Ruben et al. [ 26 ] and Wei et al. [ 27 ]. Both authors reported super-hydrophilic surface finish and the low concentration of carbon as deduced from XPS spectra. The nanoparticles were probably almost free from carbon since the C signal probably arose from photoelectrons emitted from the less-affected segment of the coating between the neighboring nanoparticles. Even a very mild plasma treatment is sufficient for obtaining the surface finish as in Figure 3 c since both authors used a discharge power of solely 15 and 10 W, respectively. The technique is especially useful for specific applications (capillary flow and proliferation of biological cells). However, the adhesion of any coating on the substrate of the surface finish as in Figure 3 c may be questionable. A similar, but more complex technique as in Figure 3 was also adopted by Lin et al. [ 25 ] and Airoudj et al. [ 29 ]. The method presented schematically in Figure 3 could be simplified by merging the deposition and oxygen plasma treatments, for example by deposition of SiO x nanoparticles from gaseous plasma sustained in oxygen with an admixture of HMDSO. The nanoparticles prepared this way, however, may not adhere to the polymer surface. This drawback is suppressed by depositing nanoparticles of a third material onto a plasma-activated polymer substrate. The technique was elaborated by Kuzminova et al. [ 23 ] and Kylian et al. [ 24 ]. The schematic of such a procedure is shown in Figure 4 . A polymer substrate is first activated similarly as in Figure 3 . Then, and without breaking vacuum conditions, the substrate is exposed to nanoparticles of the typical diameter of 100 nm, as shown in Figure 4 a. The nanoparticles form a thin layer on the polymer substrate ( Figure 4 b). The layer of nanoparticles is then coated with a thin film of carbon-depleted PDMSO-like coating ( Figure 4 c). The coating is rich in oxygen and thus provides binding sites for any further coating. The technique is scalable and assures for long-terming super-hydrophilicity. The adhesion between the nanoparticles and the polymer substrate may limit the application of this method."
} | 4,116 |
29054870 | PMC5734045 | pmc | 6,963 | {
"abstract": "ABSTRACT There is a growing interest in the use of microbial fermentation for the generation of high-demand, high-purity chemicals using cheap feedstocks in an environmentally friendly manner. One example explored here is the production of isoprene (C 5 H 8 ), a hemiterpene, which is primarily polymerized to polyisoprene in synthetic rubber in tires but which can also be converted to C 10 and C 15 biofuels. The strictly anaerobic, acetogenic bacterium Clostridium ljungdahlii , used in all of the work described here, is capable of glycolysis using the Embden-Meyerhof-Parnas pathway and of carbon fixation using the Wood-Ljungdahl pathway. Clostridium - Escherichia coli shuttle plasmids, each bearing either 2 or 3 different heterologous genes of the eukaryotic mevalonic acid (MVA) pathway or eukaryotic isopentenyl pyrophosphate isomerase (Idi) and isoprene synthase (IspS), were constructed and electroporated into C. ljungdahlii . These plasmids, one or two of which were introduced into the host cells, enabled the synthesis of mevalonate and of isoprene from fructose and from syngas (H 2 , CO 2 , and CO) and the conversion of mevalonate to isoprene. All of the heterologous enzymes of the MVA pathway, as well as Idi and IspS, were shown to be synthesized at high levels in C. ljungdahlii , as demonstrated by Western blotting, and were enzymatically active, as demonstrated by in vivo product synthesis. The quantities of mevalonate and isoprene produced here are far below what would be required of a commercial production strain. However, proposals are made that could enable a substantial increase in the mass yield of product formation. IMPORTANCE This study demonstrates the ability to synthesize a heterologous metabolic pathway in C. ljungdahlii , an organism capable of metabolizing either simple sugars or syngas or both together (mixotrophy). Syngas, an inexpensive source of carbon and reducing equivalents, is produced as a major component of some industrial waste gas, and it can be generated by gasification of cellulosic biowaste and of municipal solid waste. Its conversion to useful products therefore offers potential cost and environmental benefits. The ability of C. ljungdahlii to grow mixotrophically also enables the recapture, should there be sufficient reducing equivalents available, of the CO 2 released upon glycolysis, potentially increasing the mass yield of product formation. Isoprene is the simplest of the terpenoids, and so the demonstration of its production is a first step toward the synthesis of higher-value products of the terpenoid pathway.",
"introduction": "INTRODUCTION Isoprene is a hemiterpene (C 5 H 8 ), used primarily for polymerization to cis -1,4-polyisoprene, the main constituent of natural rubber (as in vehicle tires and surgical gloves), and for styrene-isoprene-styrene block copolymer (as in thermoplastic rubber and adhesives). Isoprene can also be converted by oligomerization and hydrogenation to produce blendable, saturated C 10 and C 15 biofuels ( 1 ). The annual global production of petrochemically derived isoprene is close to one million tons ( 2 ). With the increase in the volatility of oil prices, clean energy policies likely to result in a decline in U.S. petroleum consumption and thus in the availability of C 5 fractions from naphtha cracking, and the significant threat of fungal disease to rubber plantations in Asia and Africa ( 3 ), there is an increased demand for a more sustainable and reliable generation of isoprene, particularly for the production of synthetic cis -1,4-polyisoprene for tires. Isoprene is also the simplest product of the terpenoid pathway. The ability to engineer the production of isoprene is therefore a first step toward the engineering of a host organism to produce this and higher-value terpenoids. Two evolutionarily distinct pathways exist for the generation of isopentenyl-pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP), the starting point for all terpenoid biosynthesis, including isoprene. These are (i) the 2- C -methyl- d -erythritol-4-phosphate/1-deoxy- d -xylulose-5-phosphate (DOXP/MEP) pathway, found in most prokaryotes and which begins with the condensation of pyruvate and glyceraldehyde-3-phosphate, and (ii) the mevalonic acid (MVA) group of pathways found in most eukaryotes, archaea, and some bacteria, which generally begins with the condensation of two acetyl coenzyme A (acetyl-CoA) molecules. Efforts at aerobic microbial biosynthesis of isoprene from glucose, using heterologous protein synthesis of the MVA pathway, have shown higher rates, yields, and titers of isoprene synthesis in Escherichia coli than those from the native DOXP/MEP pathway ( 2 , 4 ). The exact reason for the higher yields and titers with the MVA pathway is still unclear but may have to do with reduced regulatory control associated with the expression of heterologous MVA pathway genes (e.g., from Saccharomyces cerevisiae and Enterococcus faecalis ) or more severe flux bottlenecks in the DOXP/MEP pathway ( 4 , 5 ). In the approach demonstrated here, we use fructose and syngas individually as sources of carbon and reducing equivalents ( 6 ) to be metabolized to mevalonate and to isoprene in the acetogenic bacterium ( 7 ) Clostridium ljungdahlii . This organism is Gram positive and a strict anaerobe, capable of autotrophic growth on H 2 plus CO 2 or CO (syngas) or all three together using the Wood-Ljungdahl ([WL] reductive acetyl-CoA) pathway ( 8 , 9 ), the final product of which is acetyl-CoA. The conversion of syngas to isoprene via acetyl-CoA requires less energy using the MVA pathway than using the DOXP/MEP pathway (which would require gluconeogenesis) and so is the pathway of choice where ATP is limiting. C. ljungdahlii can also grow heterotrophically on organic substrates (e.g., hexoses and pentoses [ 7 , 10 ]) and mixotrophically, refixing, via the WL pathway, the CO 2 liberated upon the decarboxylation of pyruvate to acetyl-CoA in glycolysis ( 7 , 11 ). A genetic system has been developed for C. ljungdahlii ( 12 , 13 ), and it has been demonstrated to be capable of heterologous gene expression ( 12 ). In this paper, we will show that it is possible to couple the Embden-Meyerhof-Parnas (EMP) and the Wood-Ljungdahl (WL) pathways to a genetically engineered eukaryotic MVA pathway in C. ljungdahlii , enabling the conversions of fructose and syngas to mevalonate and isoprene. Syngas is likely to be a less expensive feedstock than a hexose sugar on a mass basis, as it can be generated by the gasification (via steam or oxygen reforming) of a wide range of inexpensive materials, e.g., natural gas, agricultural waste, coal, and municipal solid waste ( 6 ). Low-cost CO-enriched waste streams are also generated as a consequence of industrial processes such as steel manufacture and oil refining, a source that has been exploited by LanzaTech for the generation of biofuels and other biochemicals using acetogenic microorganisms ( 6 ). That syngas can be derived from waste gas or from waste materials means that there is potentially an environmental benefit, in addition to a cost benefit, associated with the use of such feedstocks.\n\nPlasmids used for the introduction of the MVA pathway genes. Four different replicons found in Gram-positive organisms are represented in the set of Clostridium-E. coli shuttle plasmids used here ( 14 ) (see Materials and Methods). Most of the C. ljungdahlii transformants were obtained with the pCB102 replicon (typically 50 to 100/μg plasmid DNA), approximately one order of magnitude fewer were obtained with pBP1, and none were obtained with pCD6 and pIM13.",
"discussion": "DISCUSSION We have demonstrated that all of the heterologous enzymes of the eukaryotic MVA pathway plus Idi and IspS can be synthesized in vivo at high levels and in active forms using Clostridium-E. coli shuttle plasmids in C. ljungdahlii . In transformants cultivated on fructose and on syngas, heterologous MvaE and MvaS ( Fig. 2 ) catalyzed the production of mevalonate from acetyl-CoA, and heterologous Idi and IspS catalyzed the production of isoprene from IPP and DMAPP, produced via the endogenous DOXP/MEP pathway ( 17 ). Mvk, Pmk, Mvd, Idi, and IspS ( Fig. 2 ) together were able to catalyze in vivo the synthesis of isoprene from exogenous mevalonate. However, the quantities of product (mevalonate and isoprene) produced here, while a proof of concept, are far below what is required for commercial production. Increased isoprene production is likely to require enhanced levels of heterologous enzyme synthesis and further refinements directed at flux optimization, redox balance, and ATP supply. Comparison of cultivation conditions. The isoprene measurements give some indication of the limitations in pathway flux. Table 2 indicates that considerably more isoprene (18- to 90-fold/cell) accumulates when generated from mevalonate using the lower MVA pathway plus Idi and IspS than when only Idi and IspS are produced in fructose-grown cells. This difference likely results from a higher rate of synthesis and higher intracellular concentrations of IPP and DMAPP produced from mevalonate in the MVA pathway than those produced by the endogenous DOXP/MEP pathway ( 17 ) in cells grown on fructose or syngas. An accumulation of toxic phosphorylated intermediates derived from mevalonate, particularly IPP ( 5 ), is also a possible explanation for the inhibition of cell growth at ≥10 mM mevalonate. Growth inhibition by ≥10 mM mevalonate itself also remains a possibility. As shown earlier, pJF100-transformed C. ljungdahlii was able to produce 2 mM mevalonate while growing to an OD 600 of 2 ( Fig. 1 ). Even the level of isoprene produced in the absence of added mevalonate was higher upon the synthesis of the lower mevalonate pathway plus Idi and IspS than in the other cases described in Table 2 . MES-F medium contains a very low concentration of mevalonate, 0.36 μg/ml or 2.4 μM ( Fig. 4 ), that is not inhibitory to cell growth and yet is in 20-fold molar excess relative to the amount of isoprene produced (∼1.3 nmol). Considerably more mevalonate accumulates with fructose as the carbon source than with syngas. With the cells that grew to an OD 600 of 2 in both cases, ∼300 μg/ml accumulated in the presence of fructose as opposed to ≤68 μg/ml under syngas growth conditions. Growth on syngas is also substantially slower (2- to 4-fold) than it is on fructose. This difference in metabolism is likely a reflection of differences in the cellular energy charge in the two cases, 2 to 3 times higher in fructose-grown as opposed to syngas-grown C. ljungdahlii , as determined in a metabolomics study on C. ljungdahlii (B. A. Diner, Y. Xu, and P. Mauvais, unpublished). Glycolysis produces a net 2 ATP molecules by substrate-level phosphorylation per 2 acetyl-CoA molecules produced in the EMP pathway. Net ATP production from syngas to acetyl-CoA is the result of the operation of the Wood-Ljungdahl pathway (which consumes one ATP/acetyl-CoA molecule) and the chemiosmotic Rnf complex, coupled by the electrochemical membrane potential to the cytoplasmic membrane-bound ATP synthase ( 19 ). Assuming that 3.66 H + are required per ATP in the ATP synthase and assuming that all of the electron-bifurcating and Rnf proton pumping mechanisms for energy conservation invoked by Mock et al. ( 20 ) ( Fig. 2 ) are operating in C. ljungdahlii in association with the Wood-Ljungdahl pathway, one can calculate that 0.05 ATP are consumed per acetyl-CoA molecule from H 2 plus CO 2 , and 0.50 ATP per acetyl-CoA are produced from CO. The reason for the difference between H 2 and CO as the source of reducing equivalents is because CO (E 0′ CO 2 /CO = −520 mV) ( 21 ) is more reductive than H 2 (E 0′ H + /H 2 = −414 mV) ( 22 , 23 ), thereby increasing on a mole basis the amount of reduced ferredoxin (Fd red ) that can be produced by CO relative to H 2 and consequently increasing the amount of ATP that can be generated via the proton motive force maintained by the Rnf complex coupled to the reaction Fd red + NAD + → Fd ox + NADH (with a presumed 2H + /2e − , see references 19 and 20 ). The 2 g yeast extract present per liter MES medium had little impact on the flux of carbon to the products in C. ljungdahlii . Syngas-adapted cells cultivated on MES-0F under an N 2 atmosphere showed no sustained growth. We also found that >2 times as many moles of CO were consumed as moles of CO 2 produced ( Table 3 ) when C. ljungdahlii was grown on MES-0F and syngas, indicating that a significant fraction of the carbon from CO was converted to biomass and product. Bioenergetics considerations for syngas conversion to isoprene. A total of 6 ATP molecules is required for the synthesis of isoprene from syngas, including 3 for the synthesis of 3 acetyl-CoA molecules via the Wood-Ljungdahl pathway and another 3 for the conversion via the MVA pathway of the 3 acetyl-CoA molecules to isoprene. Bertsch and Müller ( 24 ) have argued that in Acetobacterium woodii grown on syngas, an insufficient amount of ATP is generated chemiosmotically to enable the synthesis of isoprene. Some substrate-level ATP production via phosphotransacetylase and acetyl kinase from acetyl-CoA ( Fig. 2 ) is required. The electron-bifurcating redox complexes in C. ljungdahlii differ in numbers and cofactor specificities relative to those in A. woodii ( 20 , 24 ) and offer some bioenergetic advantage over those in A. woodii . However, our own calculations (not shown) indicate some shortfall in ATP generation in C. ljungdahlii , even if this organism were to use the same energy-conserving electron-bifurcating complexes suggested to be operating in Clostridium autoethanogenum ( 20 ) ( Fig. 2 ), including MTHFR, the presence of which remains speculative ( 25 ). As in the redox argument above, it is advantageous in the metabolic pathways to produce NADPH (NADP + /NADPH, E′ = −370 mV under physiological conditions) ( 21 ) and to consume NADH (NAD + /NADH, E′ = −280 mV under physiological conditions) ( 26 ), as NADPH (but not NADH) can generate Fd red by Nfn electron bifurcation ( Fig. 2 ) ( 27 ). As an example, replacing the NADPH-dependent 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase (MvaE) with one that is NADH dependent ( 28 ) adds on a mole basis 1 Fd red . In addition, assuming 3.3 H + (Na + )/ATP ( 24 ) or 3.66 H + /ATP ( 20 ) in the ATP synthase, then the ATP chemiosmotically generated using the Rnf complex is close to the 6 (we calculate 6.2 and 5.7, respectively) molecules necessary for isoprene synthesis from CO. Allowing some substrate-level phosphorylation, derived from acetyl-CoA to acetate ( Fig. 2 ), would provide additional ATP but at a further cost to the mass yield of isoprene synthesis. There are a number of options, yet to be explored, that might further enhance ATP synthesis. A H + -dependent pyrophosphatase has been identified in prokaryotes ( 29 – 31 ), which if introduced into C. ljungdahlii and directed to the cytoplasmic membrane, could contribute (2 H + translocated per PP i hydrolyzed [ 32 ]) to the membrane electrochemical proton potential. Doing so would require the use of internal sources of pyrophosphate (e.g., RNA, DNA, polysaccharide, and fatty acyl-CoA synthesis), as an exogenous supply would likely require an expenditure of energy for PP i import. Potential for further improvement in the energy balance. Coelho and coworkers ( 33 ) have proposed four anaerobic pathways, beginning with 2,3-dihydroxyisovalerate (DHIV), for the synthesis of isoprene. DHIV can be generated from syngas and the WL pathway by acetyl-CoA → pyruvate → 2-acetolactate → DHIV. While these pathways consume only 1 or 2 ATP molecules as opposed to 3 for the MVA pathway conversion of acetyl-CoA to one isoprene, their realization would require the discovery of 5 to 7 new enzymes. It was originally shown in Clostridium aceticum and Moorella thermoacetica that the 2 CO 2 molecules produced per glucose in the EMP pathway could be refixed by the WL pathway using the eight reducing equivalents derived from glycolysis ( 7 ). There are sufficient reducing equivalents (24 e − ) and ATP (six molecules) produced in the glycolysis of 3 fructose molecules to convert the 6 acetyl-CoA molecules generated to 2 isoprenes via the MVA pathway. However, there is then no residual ATP and insufficient reducing equivalents to enable the 6 CO 2 molecules, generated at the same time, to be refixed and converted to isoprene. There are potential solutions to both deficiencies. It was recently demonstrated in C. ljungdahlii that H 2 can provide additional reducing equivalents, enabling, through mixotrophic metabolism, further CO 2 refixation in the production of acetone ( 11 ). A number of pathway modifications of the MVA and glycolytic pathways could enable an enhancement in the generation of ATP as well. These include, as above, the replacement of the NADPH-dependent HMG-CoA reductase with its NADH-dependent homologue in the MVA pathway and, in the EMP pathway, the replacement of the NAD-dependent with an NADP-dependent glyceraldehyde-3-phosphate dehydrogenase (e.g., see reference 34 ) and replacement of the ATP-dependent phosphofructokinase with its pyrophosphate-dependent homologue (e.g., from Methylomonas methanica [ 35 ]) ( Fig. 2 ). The combination of supplemental reducing equivalents, cofactor replacement, and three rather than two ATP molecules per hexose sugar converted to acetyl-CoA via glycolysis ( 36 , 37 ) could enable, by CO 2 refixation, the conversion on a mole basis of 3 fructoses to 3 isoprenes. Our calculations (not shown) indicate that these changes in glycolytic substrate-level phosphorylation plus chemiosmotic ATP synthesis could satisfy the overall ATP demand for producing 3 isoprene molecules from 3 fructose molecules with a theoretical mass yield of 37%, an improvement over the 25% yield of the classical EMP and MVA pathways."
} | 4,514 |
39668401 | PMC11848123 | pmc | 6,964 | {
"abstract": "ABSTRACT Several recently discovered small proteins of less than 100 amino acids control important, but sometimes surprising, steps in the metabolism of cyanobacteria. There is mounting evidence that a large number of small protein genes have also been overlooked in the genome annotation of many other microorganisms. Although too short for enzymatic activity, their functional characterization has frequently revealed the involvement in processes such as signaling and sensing, interspecies communication, stress responses, metabolism, regulation of transcription and translation, and in the formation of multisubunit protein complexes. Cyanobacteria are the only prokaryotes that perform oxygenic photosynthesis. They thrive under a wide variety of conditions as long as there is light and must cope with dynamic changes in the environment. To acclimate to these fluctuations, frequently small regulatory proteins become expressed that target key enzymes and metabolic processes. The consequences of their actions are profound and can even impact the surrounding microbiome. This review highlights the diverse functions of recently discovered small proteins that control cyanobacterial metabolism. It also addresses why many of these proteins have been overlooked so far and explores the potential for implementing metabolic engineering strategies to improve the use of cyanobacteria in biotechnological applications.",
"conclusion": "3 Conclusion Several small proteins in cyanobacteria mediate important structural functions in photosynthesis, others fulfil regulatory functions in the integration of the photosynthesis‐derived cellular redox status with the carbon/nitrogen balance and various metabolic activities. Several small proteins are potent regulators of metabolic activities. Those that directly affect the activities of abundant enzymes or multiprotein complexes, such as IF7 and NirP1, which act on glutamine synthetase and nitrite reductase, respectively, or AtpΘ, which targets ATP synthase, are of critical importance. Because these targets are abundant and stable proteins in the cell, the control exerted by these small protein regulators cannot be replaced by transcriptional regulation or by posttranscriptional mechanisms targeting the translation of the corresponding enzymes. However, those regulatory mechanisms frequently coexist. An overview of the roles of small proteins in cyanobacteria is given in Figure 6 . There are 83 annotated genes in Synechocystis encoding proteins of 60 aa or less, and 54 of these are currently without a known function. Therefore, it is obvious that many small proteins remain functionally uncharacterized, and it is likely that some of them are involved in fundamental processes. Moreover, many more sORFs at unorthodox sites such as out‐of‐frame within annotated genes or in non‐coding RNAs likely wait to be discovered (Figure 1 ). A major obstacle is the lack of comprehensive large datasets and the significant bioinformatic and experimental challenges. Obvious strategies to address these challenges are ribosome profiling and proteogenomic screens [ 30 , 31 , 32 , 33 , 34 , 35 ]. These scans may be combined with data from other genome‐wide approaches, such as CRISPR interference (CRISPRi) screens, in which the relevance of every targeted locus can be inferred over time or under different conditions. For cyanobacteria, genome‐wide CRISPRi screens were developed for Synechocystis . Initially, all 3546 annotated protein‐coding genes and 1871 potential non‐coding RNA segments on the chromosome were addressed and in a follow‐up study also plasmid‐located genes were included. The growth of all library members was followed by deep sequencing and fitness estimates were derived [ 83 , 84 ]. In this way, condition‐specific importance could be assigned to previously uncharacterized loci, including some non‐coding RNAs. Another, more recently developed approach, called “ProTInSeq”, involves sequencing of transposon insertions which can only express a selection marker when they were inserted in‐frame in a protein‐coding gene [ 85 ]. When ProTInSeq was applied to Mycoplasma pneumoniae , it identified 153 sORFs that had not been previously annotated [ 85 ]. It is of great relevance to make such comprehensive large datasets accessible. Regarding ribosome profiling, the RIBOBASE data repository has been established that now hosts these data for analyses of the bacteria E. coli and Sinorhizobium meliloti , of the archaea Haloferax volcanii and Methanosarcina mazei and likely of several more prokaryotes once the corresponding analyses are concluded [ 31 , 32 , 34 , 86 ]. For proteogenomic datasets, the concept of integrated proteogenomics databases (iPtgxDBs) has been developed [ 87 ]. FIGURE 6 Overview of the roles of small proteins in cyanobacteria with their potential impact on the microbiome. Small proteins (highlighted in blue letters) in cyanobacteria play diverse roles ranging from regulating internal metabolic processes to influencing the surrounding microbiome. Secreted metabolites of metabolic pathways controlled by small proteins can serve as products for other organisms. Thus, small proteins play a pivotal role in the formation of both internal and external metabolic networks, thereby underscoring their importance for the ecosystem and the interactions between organisms within the microbiome. To fully understand the diverse functions of small proteins and their contribution to cellular and physiological processes, it is crucial to overcome these limitations. In this respect, the combination of proteogenomic analyses with ribosome profiling is powerful and will lead to the identification of many more small proteins [ 30 , 31 , 32 , 33 ]. Cyanobacteria could play an important role in biotechnology as they require only sunlight, CO 2 , and simple inorganic nutrients for growth. They are producers of several natural products like terpenes that serve as substrates in pharmaceuticals, cosmetics, or in biofuel industries [ 82 ]. Deepening our understanding of the regulatory mechanisms that control the carbon/nitrogen balance in cyanobacteria is essential not only for the progress of fundamental research but also for the further development of biotechnological applications. Identifying novel regulatory players involved in carbon and nitrogen metabolism, as well as their connections and interactions with the overall control of growth and metabolic networks, has the potential to enhance the production of target compounds. Finally, analyzing small proteins in cyanobacteria gives clues about their possible origin. Some are clearly widely conserved, beyond cyanobacteria, such as the ribosomal proteins Rpl34, Rpl36, Rps21, and Rps32. Consistent with the conservation of the photosynthetic machinery, many small proteins associated with photosynthesis are conserved between cyanobacteria and the chloroplasts of plants and eukaryotic algae. But some small proteins play more specific roles and appear to have originated more recently, such as SliP4 and AcnSP (Figures 4 and 5 ). SliP4 is a single‐transmembrane helix protein and might have functional homologs that lack sequence conservation. In contrast, AcnSP derived from a partial gene fragment of the aconitase gene (AcnB) that was likely duplicated and inserted into pSYSA, one of the plasmids of Synechocystis . Gene duplications are an accepted source of new genes; therefore, this example highlights a possible mechanism also for the origin of small protein‐coding genes by duplicating just a few codons encoding a relevant protein domain or functional sequence stretch.",
"introduction": "1 Introduction 1.1 Small Proteins in Cyanobacteria At least 2.7 billion years ago, cyanobacteria evolved the ability to perform oxygenic photosynthesis. With that, cyanobacteria are the only prokaryotes that are capable of carrying out oxygenic photosynthesis. This process uses the light energy to split water molecules to generate chemical energy for CO 2 fixation into carbohydrates, thereby releasing molecular oxygen [ 1 ]. Due to their long evolutionary history, cyanobacteria can be found in almost all habitats of life. They occur in the sea, freshwater, soil, biological soil crusts and even in snow, hot and arid deserts, or thermal springs. Cyanobacteria are therefore a diverse group of prokaryotes whose remarkable characteristic is their ability to adapt to extreme and changing conditions [ 2 , 3 , 4 , 5 ]. In contrast to all other bacteria, the physiology of cyanobacteria is based on the assimilation of organic compounds from inorganic carbon through oxygenic photosynthesis, which poses many challenges. Although the inorganic carbon supply is relatively stable, the concentrations of nitrogen‐containing nutrients vary a lot over the course of the seasons. Therefore, one of these challenges is the integration of carbon and nitrogen metabolism with each other and with the availability of reducing power from photosynthesis [ 6 ]. It is meanwhile well established that several small proteins are involved in the control of these processes [ 7 , 8 , 9 ]. Due to the endosymbiotic origin of chloroplasts from a cyanobacterial ancestor [ 10 ], and the associated massive influx of cyanobacterial genes, some of these small protein regulators have conserved homologs in plants and eukaryotic algae, such as the Calvin Cycle regulator CP12 [ 11 ]. Small proteins are considered particularly useful in this regard, as they provide an efficient means of governing the metabolism, without a large investment of amino acids and energy. The classification as a small protein is not commonly standardized. Often, proteins with less or equal to 50 amino acids are considered small proteins or microproteins [ 12 ], but an upper limit of 60–100 aa can also be found in the literature. In general, small proteins are polypeptides too small to function as enzymes, but they often have structural functions or act as regulators [ 12 , 13 , 14 , 15 ]. In cyanobacteria, many small proteins were characterized previously due to the meticulous analysis of the large photosynthetic apparatus. For example, the monomeric photosystem I and II complexes in cyanobacteria consist of 12 and at least 20 different subunits encoded by psa and psb genes, respectively [ 16 , 17 ]. Twenty proteins of less than 60 aa were characterized in photosystem I (genes psaM, psaJ, psaI ) [ 18 ], photosystem II ( psbM, psbT ( ycf8 ), psbI, psbL, psbJ, psbY, psbX, psb30 ( ycf12 ), psbN, psbF, psbK ) [ 19 , 20 ], photosynthetic electron transport (cytochrome b \n 6 \n f complex; petL, petN, petM, petG ) [ 21 , 22 , 23 ], or as having accessory functions ( hliD [ scpE ], hliC [ scpB ]) [ 24 , 25 ]. Among the photosystem I subunits, the two small proteins, PsaE (8.0 kDa) and PsaJ (4.4 kDa), are essential for the stable assembly and structural integrity of the entire complex [ 26 ]. Some of these proteins are very small, for instance, the cytochrome b \n 6 \n f complex subunit VIII, encoded by petN , is with only 29 aa the shortest conserved protein in cyanobacteria [ 27 ]. These photosynthesis‐related proteins are widely conserved throughout the cyanobacterial phylum and most have homologs in the chloroplasts of plants and eukaryotic algae. The majority of these proteins play structural roles, but many other small proteins often have regulatory functions, and generally, the various functions of small proteins are still largely unknown, even in model bacteria. 1.2 Are Small Proteins Still Too Small to Be Detected? In recent years, more and more small proteins have been discovered, but why have they been overlooked for so long? The answer to this question is that the identification of small proteins and short protein‐coding genes poses particular challenges in bioinformatic analyses and experimental methods. Although genomic data are now available for hundreds of thousands of bacteria, even with completely sequenced genomes, the genes for small proteins are often incorrectly annotated or not annotated at all. The algorithms for detecting small proteins were previously often set to a threshold of 100 aa, so that smaller open reading frames (sORFs) were not recognized (Figure 1 ). Although the quality of gene prediction algorithms has improved remarkably by software such as Prodigal [ 28 ], substantial shortcomings exist particularly for finding very short and nonconserved genes or sORFs in unorthodox places, such as antisense RNAs or within other genes. In addition to the complexity of in‐silico analysis, the limitations of experimental approaches also pose a major challenge. Certain amino acids in proteins are important for detection by the respective method. Coomassie Blue, the dye used for staining proteins in gels, binds primarily to arginine and to some extent to other basic and hydrophobic residues [ 29 ]. In mass spectrometry (MS), proteins are usually proteolytically digested with trypsin at lysine residues. Because the number of the corresponding amino acid residues in small proteins can be biased, they may be stained less clearly on gels, and the small number of expected peptide fragments makes them more difficult to detect by MS (Figure 1 ). In addition, the MS‐based mapping of peptide fragments will fail if the corresponding small protein‐coding genes were not modeled during the genome annotation. The identification of small proteins and their genes has only recently been improving, and increasing numbers are discovered with the advent of ribosome profiling and progress in proteogenomic methods, both in bacteria as well as in archaea [ 30 , 31 , 32 , 33 , 34 , 35 ]. FIGURE 1 Schematic overview of problems and challenges that have resulted in small proteins frequently being overlooked. Bioinformatic challenges, as well as limitations in biochemical analyses, have contributed to the frequent neglect of small proteins in research. Bioinformatic tools used for gene prediction and protein annotation are generally optimized for larger proteins and fail in the identification of protein‐coding potential (nested genes) within larger reading frames (ORFs), or in atypical locations such as in antisense transcripts. Some mRNAs were initially classified as non‐coding RNA (often used as a collective term for antisense RNAs, sRNAs, guide RNAs, and other types of transcripts lacking coding potential), or not detected at all due to the atypically short coding sequences and the lack of conserved domains. During heterologous expression, small proteins are often only weakly expressed, secreted into the medium, or are toxic to their host cells. Because of their small size, amino acid composition, and low abundance of detectable peptides, small proteins are often difficult to stain or resolve in protein gels and can be easily missed in mass spectrometry (MS) analyses. Here, first the impact of transcriptome analyses in improving the identification of small protein‐coding genes is to be mentioned. Following the genome‐wide mapping of transcription start sites and the identification of transcriptional units, these can be scanned for the presence of unassigned candidate sORFs and their potential conservation evaluated. Indeed, transcripts previously described as non‐coding RNAs were found to contain sORFs. In the model cyanobacterium Synechocystis sp. PCC 6803 (hereafter Synechocystis ), this strategy led to the identification of 293 candidate sORFs ≤80 codons including the known sORFs for small proteins involved in photosynthesis. From these, 146 were shared with the Synechocystis strain PCC 6714, 42 with the land plant Arabidopsis thaliana and 25 with Escherichia coli , and after tagging, five were validated to exist [ 13 ]. Ribosome profiling, or ribo‐seq, is a method for analyzing the association of ribosomes with transcripts in a transcriptome‐wide fashion [ 36 ]. In ribosome profiling, the cellular lysate is treated with a nuclease that digests nucleic acids not bound to ribosomes. The ribosome‐bound mRNA fragments are then enriched by sucrose gradient ultracentrifugation, after which the associated RNA is recovered, converted to cDNA, and sequenced. Ribosome profiling is now well‐established for bacteria [ 32 , 37 ], and the methodology has been further improved by introducing the use of antibiotics that stall ribosomes at initiation or termination codons. These antibiotics are retapamulin that stalls ribosomes at translation initiation sites [ 38 ], and apidaecin that stalls ribosomes at or close to stop codons [ 39 ]. Subsequent sequencing then identifies the RNA fragments bound by ribosomes, and the comparison with the total transcriptome reveals the positions where ribosomes were actively translating an mRNA, revealing previously unrecognized translation initiation and termination sites, and thus unknown reading frames and the encoded proteins. In parallel, bioinformatic workflows have been developed to analyze bacterial ribosome profiling data [ 40 ]. In summary, these ribosome profiling approaches have truly revolutionized the identification of small proteins and their genes [ 41 ], and have also been applied to cyanobacteria [ 42 , 43 ]. Many small proteins, most of them related to photosynthesis, were previously analyzed in cyanobacteria in more detail, providing a kind of paradigm for the functional analysis of the microbial small proteome [ 13 ]. Therefore, in the following we will put emphasis on the more recent advancements in the functional analysis of small proteins in cyanobacteria, with a focus on Synechocystis as a model."
} | 4,413 |
29719788 | PMC5925392 | pmc | 6,967 | {
"abstract": "The cultivation of Panax plants is hindered by replanting problems, which may be caused by plant-driven changes in the soil microbial community. Inoculation with microbial antagonists may efficiently alleviate replanting issues. Through high-throughput sequencing, this study revealed that bacterial diversity decreased, whereas fungal diversity increased, in the rhizosphere soils of adult ginseng plants at the root growth stage under different ages. Few microbial community, such as Luteolibacter , Cytophagaceae, Luteibacter , Sphingomonas , Sphingomonadaceae, and Zygomycota, were observed; the relative abundance of microorganisms, namely, Brevundimonas , Enterobacteriaceae, Pandoraea , Cantharellales, Dendryphion , Fusarium , and Chytridiomycota, increased in the soils of adult ginseng plants compared with those in the soils of 2-year-old seedlings. Bacillus subtilis 50-1, a microbial antagonist against the pathogenic Fusarium oxysporum , was isolated through a dual culture technique. These bacteria acted with a biocontrol efficacy of 67.8%. The ginseng death rate and Fusarium abundance decreased by 63.3% and 46.1%, respectively, after inoculation with B. subtilis 50-1. Data revealed that microecological degradation could result from ginseng-driven changes in rhizospheric microbial communities; these changes are associated with the different ages and developmental stages of ginseng plants. Biocontrol using microbial antagonists alleviated the replanting problem.",
"conclusion": "5 Conclusions Ginseng cropping induced changes in rhizospheric microbial communities and decreased bacterial diversity. These effects could collectively cause microecological degradation, which consequently results in replanting problems. However, inoculation with a biocontrol bacterial strain alleviated the replanting problem and improved the growth of ginseng ( Fig. 7 ). Given that the replanting issues that underlie this work are common to many perennial medicinal plants, our work provides crucial insight into the replanting problem within the framework of rhizospheric microbial dynamics. Moreover, we confirmed that inoculation with microbial antagonists is an effective soil bioremediation method that alleviates the replanting problem associated with Chinese medicinal plants. Figure 7 Schematic model demonstrated ginseng plants droved the imbalance of microbial community and biological bacterium alleviated replanting problem. Fig. 7",
"introduction": "1 Introduction Panax ginseng C.A. Meyer demonstrates neuroprotective effects against ischemic stroke 1 . Ginseng plants are mainly distributed in Asia, particularly in China and South Korea 2 . The current annual global market value of this species is approximately 3.5 billion dollars 3 . Wild ginseng resources have dwindled because of excessive and predatory exploitation; thus, wild ginseng has been gradually substituted with cultivated ginseng in the mainstream market 4. , 5. . Ginseng is continuously cultivated in fixed plots for 4–5 years; however, subsequent replanting commonly fails because of obstacles to continuous cropping 6 . Decades of crop rotation are needed for successful replanting. The replanting issue is a severe drawback that hinders the development of the ginseng industry and thus requires urgent resolution. The replanting problem is caused by the deterioration of soil physicochemical properties, allelopathy/autotoxicity, outbreak of soil-borne diseases, and changes in soil microbial communities 7. , 8. , 9. . The change in soil microbial community is a major factor that hinders crop replantation 10. , 11. . Imbalances in rhizospheric microbial communities disrupt ginseng cultivation 12 . Microbial communities change during ginseng cultivation 13 , and the increased abundance of pathogenic microorganisms is related to the occurrence of soil-borne disease 14 . Collective changes in the rhizospheric microbial community may cause replanting issues. Plants of different ages can alter microbial community 15 . The continuous cropping of Panax quinquefolius L. changes the microbial community in arable soil 16 . Ginseng plants of different ages drive changes in microbial community. Specifically, rhizospheric and nonrhizospheric soil microbial communities in a particular site become drastically different with ginseng growth 4 . The diversity and relative activity of soil microbial communities change throughout plant development 17 . However, the mechanism through which Panax plants of different ages and developmental stages mediate microbial community is unclear. Root rot is a severe disease that hinders the replantation of Panax plants 18 . Fusarium oxysporum is the main pathogenic fungus of root rot in Panax plants 14. , 19. . The relative abundance of F. oxysporum increases with notoginseng growth and is significantly related with the death rate of ginseng seedlings 14 . The application of biocontrol bacteria could effectively alleviate the occurrence of root rot. Biological control using microbial antagonists has attracted interest as an effective method to decrease the abundance of plant pathogens due to its nontoxic nature 20 . Biocontrol bacteria have important roles in plant defense, and many isolates have shown antagonistic activity against phytopathogenic fungi 21 . In tomato, Bacillus amyloliquefaciens RWL-1 inhibits the growth of F. oxysporum 22 . Nevertheless, microbial antagonists against ginseng root rot are rare. Herbgenomics has been utilized in recent investigations on medicinal plants. It involves the use of genomic tools, including metagenomic sequencing technology, to facilitate the analysis of rhizospheric microecology 23 . In the present study, 16S and 18S rRNA genes were analyzed through high-throughput sequencing to illustrate the changes in microbial diversity and composition in the rhizosphere soil of ginseng seedlings at different ages and developmental stages. Furthermore, biocontrol bacteria against F. oxysporum were isolated through a dual culture technique, and their inhibitory activity against the target pathogen in replanting soil was confirmed. The results of this study provide insight into the reasons that underlie the replanting issues caused by rhizospheric microbial communities. These data may provide an effective soil bioremediation method to replanting issues associated with Chinese medicinal plants.",
"discussion": "4 Discussion The replanting problem is a common and severe issue faced in the cultivation of medicinal plants. The biomass and tumor quality of Rehmannia glutinosa decrease as a result of replanting problems 42 . The survival rate of ginseng seedlings is lower than 25% after 3 years of replantation 43 . The replanting problem is caused by multiple factors, and changes in the soil microbial community influence soil health and crop yield 44. , 45. . The composition of the soil microbial community is governed by plant species and growth 46 . We previously used high-throughput sequencing technology to show that microbial diversity and composition change in soils cultivated with American ginseng relative to those in soils cultivated with traditional crops 16 . In the present study, we revealed that ginseng seedlings of different ages and developmental stages drive changes in the rhizospheric microbial community. We also screened for antagonistic bacteria against root rot and confirmed that biological control is a remarkably potent approach to decreasing the occurrence of root rot. The H' and Chao1 values obtained in this study revealed that bacterial diversity was low, whereas fungal diversity was high, in the rhizosphere of adult ginseng seedlings in the root growth stage. A similar study reported that increasing ginseng cultivation ages decreases bacterial diversity and increases fungal diversity 12 . The diversity of microbial communities in the rhizosphere of Pseudostellaria heterophylla decreases with the increasing number of cropping years 47 . The developmental stage of crops is an important drive of microbial community structure 17 . Additionally, microbial diversity is critical to the maintenance of soil health and quality, as well as serves as a sensitive bioindicator of soil health 48 . For example, the death rate of notoginseng and fungal diversity are significantly and negatively correlated, suggesting that fungal diversity is a potential bioindicator of soil health 14 . Microbial diversity and root disease suppression are related 49. , 50. . The decrease in bacterial diversity in response to adult plants in the root growth stage is a possible indicator of ecological variations and functional impairment. The microbial compositions of the rhizospheres of ginseng plants at different ages and developmental stages were different. Changes in microbial dynamics occur in the rhizosphere during ginseng growth 4 . Soybeans in the vegetative stage of growth affect the structure of bacterial communities, with bacterial communities are likely further altered during later growth stages 51 . Microbial community structures are highly divergent during the young plant stage of tomato but become homogeneous during the flowering and senescence stages of the plant 46 . In our study, the relative abundance of microbial community decrease, such as Luteolibacter , Cytophagaceae, Luteibacter , Sphingomonas , and Sphingomonadaceae, while that of Brevundimonas , Enterobacteriaceae, Pandoraea , Cantharellales, Dendryphion , Fusarium , and Chytridiomycota increased in the soils of adult ginseng plants compared with those in the soils of 2-year-old seedlings. Another paper reported similar results, i.e ., the population of microbe decreases, whereas that of microorganisms increases with the increasing number of cropping years 47 . The soil microbial community is an important bioindicator of soil function 52 . Changes in functional groups revealed that the microecological environment of the rhizosphere gradually degrades with the increasing age of ginseng. Soil characteristics and plant species can influence the rhizospheric microbial community 53. , 54. . In our study, field-grown ginseng plants appeared healthy without any signs of disease during the growing season ( Supplementary Information Fig. S7 ). The pH, available K , and organic matter contents of rhizosphere soils from ginseng plants of different ages and developmental stages were not significantly different ( Supplementary Information Table S5 ). Plant species could influence rhizobacterial communities 55 . Root exudates are a main drive of the changes in rhizospheric microbial communities during plant growth 53 . The richness of the rhizospheric microbial community in an Arabidopsis system is enhanced by the high cumulative levels of sugars secreted during the early developmental stages of the plant 56 . Root exudates also contain allelochemicals that disturb the balance of a microbial community 15 . Our results showed that the diversity of rhizospheric microbial communities markedly changed as the ginseng plants entered the root growth stage. This phenomenon likely resulted from the influence of different root types and root exudates. The composition of Arabidopsis root exudates changes throughout the plant development; for example, root exudates produced by tomato plants during the reproductive stage are more phytotoxic than those produced during the vegetative stage 57. , 58. . The ginseng root rapidly grows after 3 years of cropping prior to harvest, especially during the root growth stage. The root growth stage of ginseng is possibly characterized by a specific but distinctive root exudation pattern that drives different bacterial communities. In the present study, B. subtilis 50-1 was isolated and acted as an effective antagonist against F. oxysporum . Inoculation results revealed that the biocontrol bacterium decreased ginseng morbidity and alleviated the replanting problem. Numerous studies have reported that Bacillus strains are potent biological control agents of plant diseases 22. , 59. . The biocontrol efficacy of B. amyloliquefaciens 54 against bacterial fruit blotch has been proven in greenhouse experiments 60 . B. megaterium (B5), B. cereus sensu lato (B25), and Bacillus sp. (B35) exhibit antagonistic activity against F. verticillioides 61 . Bioactive constituents produced by microbial strains can attenuate the negative effects of pathogens and abiotic stresses on plants 22 . Thus, microbial antagonists could be used for the efficient and environmentally friendly control of plant pathogens. Moreover, the genome sequencing analysis of Ganoderma lucidum revealed key genes that encode for cytochrome P450s, which are involved in secondary metabolism 62 . The genome sequence of strain 50-1 would help provide insight into the pathways of functional bacteria and facilitate their exploration. However, information related to the biocontrol functions of strain 50-1 requires further analysis."
} | 3,248 |
24452305 | PMC3899592 | pmc | 6,973 | {
"abstract": "The spin memristive devices combining memristance and tunneling magnetoresistance have promising applications in multibit nonvolatile data storage and artificial neuronal computing. However, it is a great challenge for simultaneous realization of large memristance and magnetoresistance in one nanoscale junction, because it is very hard to find a proper spacer layer which not only serves as good insulating layer for tunneling magnetoresistance but also easily switches between high and low resistance states under electrical field. Here we firstly propose to use nanon composite barrier layers of CoO-ZnO to fabricate the spin memristive Co/CoO-ZnO/Co magnetic tunnel junctions. The bipolar resistance switching ratio is high up to 90, and the TMR ratio of the high resistance state gets to 8% at room temperature, which leads to three resistance states. The bipolar resistance switching is explained by the metal-insulator transition of CoO 1−v layer due to the migration of oxygen ions between CoO 1−v and ZnO 1−v .",
"discussion": "Discussion We also fabricated glass/Cr(2 nm)/Ag(30 nm)/Co(10 nm)/CoO-ZnO(2 nm)/Au(60 nm) reference junction in which an inactive Au electrode replaced the electrochemically active top Co/Ag bilayer. The bipolar switch of this junction is shown in Figure 4a , which is similar to that observed in Figure 2a . This means that the top ferromagnetic metal Co layer is not required for the observation of the bipolar switch, though it is necessary for tunneling magnetoresistance. Figure 2a and Figure 4a further indicate that the amplitude of the set voltage is always bigger than that of the reset voltage. This means that the HRS is more stable than the LRS. Similar resistive switching phenomena were attributed to the formation/rupture of metal filaments. Some experiments suggested that the metallic ions injected from the anode to the insulator may be responsible for the filament channel 5 26 27 . Since the Au electrode is inactive and the Ag ions at the bottom electrode cannot migrate into CoO-ZnO composite layers under the positive bias voltage, we can easily exclude the possibility of metallic ions migration in our case. Other experiments suggested that the electrical switching is due to the filament formation/rupture by a redox process 20 28 29 , which was caused by the migration of oxygen ions in the oxide. This filament scenario can also be excluded in our case based on the following analysis. In order to elucidate the possible conduction mechanism of the Co/CoO-ZnO/Co MTJ, we examined the temperature dependence of HRS and LRS resistance. As shown in Figure 4b , the HRS resistance decreases with increasing temperature, which is a typical feature of tunneling transport through a continuous insulating barrier 30 . Correspondingly, the nonlinear I-V curve in the HRS in Figure 2a also shows tunneling transport through a continuous insulating barrier. On the contrary, the LRS resistance increases with increasing temperature, showing metallic-like transport behavior. Moreover, the linear I-V curve in the LRS in Figure 2a shows pure ohmic transport behavior. However, the LRS resistance does not increases linearly with increasing temperature, which means that the conducting paths are still not the pure metal or alloy filaments. Figure 4c further indicates the time dependence of the electrical switching by applying a voltage less than the reset voltage in amplitude to obtain the HRS. The switching shows multiple plateaus, which are similar to the multilevel switch in TiN/ZnO/Pt 31 . The time dependence of the electrical switching shows two obvious features: first, within each plateau, the resistance (R = V/I) increases very slowly with increasing time; second, the sharp resistance jumps between plateaus indicate that the metal-insulator transition happens at least in some areas of the CoO 1−v layer and finally in the whole CoO 1−v layer. Taking into account all the above experimental results, we proposed a new scenario of migration of oxygen ions and resulting metal-insulator transition of CoO 1−v to explain the observed electrical switching and magnetoresistance, which is depicted schematically in Figure 5 . ZnO is usually an n-type semiconductor ZnO 1−v and has relatively small resistivity at room temperature due to the existence of O vacancies. By contrast, CoO is usually a p-type semiconductor Co 1−x O (0 ≤ x < 1 denotes Co vacancies) due to the existence of Co vacancies (Co-deficient or O-excess) 22 . However, in an O-deficient case like Co/CoO-ZnO/Co junctions, CoO can be in the form of CoO 1−v and has a trend to become more insulating by obtaining O ions. In this case, the change in junction resistance mainly depends on the CoO layer rather than ZnO layer. Assuming the initial HRS is the insulating CoO and/or O-deficient CoO 1−v which has much larger resistance than that of the ZnO 1−v , the resistance is mainly due to the direct tunneling of electrons through CoO and indirect tunneling through the oxygen vacancies. Under a positive bias voltage, oxygen ions (O 2− ) leave the CoO 1−v layer to ZnO 1−v layer, and make the CoO 1−v layer have more oxygen vacancies. This increases indirect tunneling paths through oxygen vacancies and reduces the resistance. With increasing the concentration of oxygen vacancies, the initial localized wave functions of the electrons trapped in the oxygen vacancies overlap and become delocalized. In this case, the insulating or semiconducting CoO 1−v of HRS becomes metal-like CoO 1−v of LRS. Since this metal-insulator transition may happen in the whole CoO 1−v , both the HRS and LRS show electrical transport properties through a whole junction area. Vice versa, a negative bias voltage can turn the metal-like CoO 1−v of LRS into the insulating or semiconducting CoO 1−v of HRS. According to the above model, two features of Figure 4c can be well explained. First, within each plateau, the resistance increases very slowly with increasing time. This means that the migration of oxygen ions (corresponding to the migration of oxygen vacancies) is much slower at a voltage less than the reset voltage in amplitude. In fact, the current due to the migration of oxygen vacancies is negligibly small as compared with the current due to the electron movement in the metal-like LRS. The decrease of the oxygen vacancies in the metal-like CoO 1−v layer mainly reduces the carrier density, which leads to a slow increase of the resistance. Second, the sharp resistance jumps between plateaus indicate that the metal-insulator transition happens at least in some areas of the CoO 1−v layer and finally in the whole CoO 1−v layer. When the concentration of oxygen vacancies in the CoO 1−v layer is less than a critical value, the delocalized wave functions of the electrons suddenly become localized and the conducting electrons were trapped in the oxygen vacancies. In this case, the metal-like CoO 1−v of LRS becomes insulating or semiconducting CoO 1−v of HRS. Finally, we briefly discussed the potential applications of the spin memristive Co/CoO-ZnO/Co MTJ devices. The spin memristive MTJ devices may have applications in multibit nonvolatile data storage and artificial neuronal computing. It is well known that MTJ is the basic storage element of magnetic random access memory with high-speed and nonvolatile memory, and meanwhile resistance switching is utilized to develop resistive random access memory. By both electrical and magnetic controlling, the multiple resistance states of the spin memristive MTJ devices can greatly increase the data storage densities. Another exciting application is that the spin memristive MTJ may work as nanoscopic synapse-neuron system. A detailed discussion about the memory effects in complex materials and nanoscale systems can be found in the recent review article 32 . In conclusions, we have successfully fabricated spin memristive Co/CoO-ZnO/Co junctions with CoO-ZnO nanon composite barrier layers, which simultaneously realize large memristance and tunneling magnetoresistance. The bipolar resistance switching ratio is high up to 90, and the TMR ratio of the high resistance state gets to 8% at room temperature. The bipolar resistance switching is explained by the migration of oxygen ions and resulting metal-insulator transition of CoO 1−v layer. The spin memristive devices have promising applications in multibit nonvolatile data storage and artificial neuronal computing."
} | 2,120 |
34442548 | PMC8401315 | pmc | 6,974 | {
"abstract": "Several marine bacteria of the Roseobacter group can inhibit other microorganisms and are especially antagonistic when growing in biofilms. This aptitude to naturally compete with other bacteria can reduce the need for antibiotics in large-scale aquaculture units, provided that their culture can be promoted and controlled. Micropatterned surfaces may facilitate and promote the biofilm formation of species from the Roseobacter group, due to the increased contact between the cells and the surface material. Our research goal is to fabricate biofilm-optimal micropatterned surfaces and investigate the relevant length scales for surface topographies that can promote the growth and biofilm formation of the Roseobacter group of bacteria. In a preliminary study, silicon surfaces comprising arrays of pillars and pits with different periodicities, diameters, and depths were produced by UV lithography and deep reactive ion etching (DRIE) on polished silicon wafers. The resulting surface microscale topologies were characterized via optical profilometry and scanning electron microscopy (SEM). Screening of the bacterial biofilm on the patterned surfaces was performed using green fluorescent staining (SYBR green I) and confocal laser scanning microscopy (CLSM). Our results indicate that there is a correlation between the surface morphology and the spatial organization of the bacterial biofilm.",
"conclusion": "5. Conclusions In this study, we established a process for the fabrication of honeycomb patterns with different periodicities and different morphologies in silicon surfaces. Our results suggest that such structures may promote biofilm formation, but the effect of the surface morphology parameters was inconclusive. Our results are encouraging, as the applied fabrication methods allow a wide range of patterns and aspect ratios and are scalable with respect to surface area. A huge parameter space, including possible different choices of materials, awaits investigation. Further improvements of the proposed method and further optimization of surface structures and surface chemistries may hold promise for significantly enhanced bacterial biofilm growth. In addition, it is also necessary to understand how those patterned surfaces can be effectively integrated into industrial applications for the promotion of beneficial bacterial biofilms, before scaling the fabrication process to large areas.",
"introduction": "1. Introduction Marine aquaculture industries are continuously expanding due to the increasing demand for fish production to feed the growing world population [ 1 ]. Infectious bacterial diseases on the rearing units of many fish species are common; therefore, the use of antibiotics is increasing [ 2 , 3 ]. Consequently, the risk of antibiotic resistance is higher, and new sustainable alternatives for disease control are essential [ 4 , 5 ]. Several studies have shown that marine bacteria of the Roseobacter group, such as the well-studied Phaeobacter inhibens , can inhibit other microorganisms, such as the fish pathogenic Vibrionaceae , by producing the antibacterial compound tropodithietic acid (TDA) [ 6 , 7 , 8 , 9 ]. Several of the probiotic bacteria are especially antagonistic when growing in biofilms [ 10 , 11 ]. One of the main challenges in using probiotic bacteria is how to optimally introduce the beneficial bacteria into the rearing units. Bentzon-Tilia et al. (2016) suggested the integration of the probiotic bacteria into synthetic bacterial communities of pre-established biofilms of biological aerated filters (BAFs) in recirculating aquaculture systems (RAS). Moreover, biofilms on synthetic communities could be released and allow the spread of probiotics throughout the system [ 12 ]. A better understanding of how a given surface can promote or inhibit biofilm attachment is crucial for potential applications. In this study, we focus on how to manipulate surface topography in order to enhance beneficial bacterial biofilms relevant to aquaculture systems. Our focus is on the enhancement of biofilm formation of the Roseobacter group, especially the TDA-producing Phaeobacter species. Several studies have shown that many factors can influence bacterial attachment on surfaces [ 13 , 14 , 15 , 16 , 17 , 18 ]. Most of these studies have focused on antibacterial properties and surface morphologies. Based on microscale topographies, micropatterned surfaces may promote biofilm formation due to increased contact between the cells and the surface material [ 19 , 20 , 21 ]. On the other hand, an emphasis on surface morphology and the promotion of biofilms is crucial. In this case, the aim is to identify how microcavities can be integrated in industrial applications for the promotion of beneficial bacterial biofilms. In our work, we explore the attachment of Phaeobacter inhibens on micropatterned silicon surfaces in order to comprehend how a particular surface morphology can influence bacterial attachment to the surface. By applying UV lithography and DRIE, we investigate how microfabricated silicon surfaces with different morphologies, such as pits or pillars, aspect ratios, and length scales of honeycomb-patterned surfaces, can maximize the attachment of biofilm in comparison to planar reference surfaces [ 22 , 23 ].",
"discussion": "4. Discussion We tested whether a surface feature length scale could increase bacterial attachment by modifying the surface area. Studies comprising surfaces with micro- and nanostructures that influence bacterial adhesion have mainly focused on antibacterial properties and surface morphologies [ 13 , 14 , 15 , 17 , 33 , 34 , 35 , 36 ]. Here, we fabricated hexagonal micropatterned surfaces with different polarities and different length scales, such as side lengths, trench widths, and depths ( Figure 1 ). The fabricated surfaces were based on a honeycomb pattern, which was chosen because it is a widely used, naturally occurring pattern [ 37 ] that has been previously studied for bacterial attachment [ 19 , 21 ]. The fabricated substrates were silicon surfaces, and the patterned designs were etched via a DRIE etching Bosch process ( Section 2.1 , Table S1 ). The etched surfaces after this process were not as smooth as a pristine planar flat surface. As shown in Figure 1 a,b, after the dry etching, the pillars and sidewalls of the pits had a certain roughness due to periodic undulations, also known as scallops [ 38 ]. Surface roughness can prevent or promote the attachment of biofilm on fabricated surfaces [ 18 , 39 ]. The formation of scallops increases the surface area, so bacterial attachment may be enhanced. In our study, the effect of the produced scallops on the honeycomb pits may also have been beneficial. By employing confocal microscopy to quantify the bacterial growth, we demonstrated that both honeycomb pillar and pit arrays resulted in higher biomass than the planar reference surface over time. Specifically, a pillar array surface with a = 10 µm and d = 5 µm showed greater growth, both after 48 and after 72 h ( Figure 3 a), compared to the flat surface. Also, in Figure 3 b, we see that the average biomass on the honeycomb pits with side length a = 10 µm and d = 5 µm was higher than that on the planar surface after 72 h. These results indicate that the bacterial biofilm volume corresponding to a surface (biomass) can be controlled by changing the trench width between honeycomb pillars and the side length of honeycomb pits. Scheuerman et al. (1998) studied bacterial attachment on microscale topographies and stated that bacterial attachment is independent of groove width [ 20 ]. Furthermore, their study suggested that only the motility of strains allows cells to accumulate on the bottom of grooves [ 20 ]. Our preliminary study is thus in disagreement with the findings of Scheuerman et al. (1998). In Figure 5 , we can observe that different side lengths and cavities might also influence bacterial biofilm attachment on patterned surfaces. Our observations are supported by those reported in a recent review, which showed that there is an interaction of bacterial adhesion on patterned surfaces [ 15 ]. A limitation in this preliminary study is concerned with the experimental setup, which consisted of a closed system, without a continuous supply of nutrients, and the fixed orientation of the silicon slides. Growth media were added only at the beginning of the experiment, and this may have influenced biofilm maturing and aging, as overgrown biofilms were encountered in some cases ( Figure S5 ). Furthermore, even though the patterned surfaces where randomized on the silicon microscope slides and the stirring was kept constant during the experiment; the orientation of the slides might have had an influence on the bacterial attachment ( Figure S2 ). Overall, overcoming the mentioned limitations could lead to a better understanding of biofilm growth promotion by means of microfabricated patterns. As previously mentioned, the material used in our pilot experiment was pure silicon. Another material with a different surface energy could very well result in enhanced biofilm attachment [ 40 , 41 , 42 , 43 , 44 , 45 ]. Finally, in this study, we used patterned surfaces with heights of around 13–14 µm. Hence, additional optimization of the structure height and surface chemistry [ 23 , 40 , 46 ] may yield better control and enhancement of bacterial colonization on the surfaces."
} | 2,358 |
31110560 | PMC6511200 | pmc | 6,975 | {
"abstract": "Background Microalgae are attracting much attention as a promising feedstock for renewable energy production, while simultaneously providing environmental benefits. So far, comparison studies for microalgae selection for this purpose were mainly based on data obtained from batch cultures, where the lipid content and the growth rate were the main selection parameters. The present study evaluates the performance of native microalgae strains in semi-continuous mode, considering the suitability of the algal-derived fatty acid composition and the saponifiable lipid productivity as selection criteria for microalgal fuel production. Evaluation of the photosynthetic performance and the robustness of the selected strain under outdoor conditions was conducted to assess its capability to grow and tolerate harsh environmental growth conditions. Results In this study, five native microalgae strains from Tunisia (one freshwater and four marine strains) were isolated and evaluated as potential raw material to produce biofuel. Firstly, molecular identification of the strains was performed. Then, experiments in semi-continuous mode at different dilution rates were carried out. The local microalgae strains were characterized in terms of biomass and lipid productivity, in addition to protein content, and fatty acid profile, content and productivity. The marine strain Chlorella sp. showed, at 0.20 1/day dilution rate, lipid and biomass productivities of 35.10 mg/L day and 0.2 g/L day, respectively. Moreover, data from chlorophyll fluorescence measurements demonstrated the robustness of this strain as it tolerated extreme outdoor conditions including high (38 °C) and low (10 °C) temperature, and high irradiance (1600 µmol/m 2 s). Conclusions Selection of native microalgae allows identifying potential strains suitable for use in the production of biofuels. The selected strain Chlorella sp. demonstrated adequate performance to be scaled up to outdoor conditions. Although experiments were performed at laboratory conditions, the methodology used in this paper allows a robust evaluation of microalgae strains for potential market applications.",
"conclusion": "Conclusions Five native Tunisian microalgae strains were isolated and identified. These were Scenedesmus sp., Tetraselmis sp., Chlorella sp., Amphora sp. and Nitzschia sp. Four isolates were successfully cultivated in semi-continuous mode at different dilution rates under laboratory conditions, which simulated outdoor conditions, in order to evaluate their potential for biofuel production. Chlorella sp. demonstrated the best performance—at 0.20 1/day, it achieved a lipid productivity of 35.10 mg/L day; while at 0.3 1/day, it achieved a FAME productivity of 20.30 mg/L day. Furthermore, when imposing severe outdoor conditions, the selected strain showed acceptable tolerance, with a photosynthetic parameters assessment using chlorophyll-fluorescence analysis revealing prompt culture recovery.",
"discussion": "Results and discussion Phylogenetic analysis A fragment of the 18S rRNA gene and the 3′-end of the rbc L gene segment ( rbc L-3′P) were used to identify the Strain WT4 and Strain WT7 diatom isolates. The rbc L-3P and SSU rDNA regions were sequenced and compared with sequences already available in GenBank database. Phylogenetic analyses based on SSU rDNA and rbc L-3P sequences revealed that Strain WT4 and Strain WT7 isolates were members of the Bacillariophyceae class (pennate diatoms) and were grouped within the Thalassiophysales and Bacillariales orders, respectively (Fig. 1 a, b). The SSU rDNA region and the rbc L-3P protein sequences revealed that Strain WT4 and Strain WT7 isolates were related to the Amphora subtropica [ 55 ] and Nitzschia draveillensis [ 56 ] species, respectively. These results demonstrated that these isolates (Strain WT4 and Strain WT7) belong to the Catenulaceae and Bacillariaceae families and were closely related to the genus Amorpha and Nitzschia, respectively. The Strain WT3 isolate was identified based on the 18S rRNA sequence and both the internal transcribed spacer regions combined with the 5.8S gene (ITS-1 + 5.8S + ITS-2). PCR amplification and subsequent DNA sequencing allowed the determination of approximately 1800 bp of the 18S rRNA gene. Strain WT3 displayed the highest 18S rRNA sequence relatedness with the green alga Tetraselmis striata (99% similarity) [ 57 ]. Strain WT3 was grouped among the Chlorodendrophyceae class and Chlorophyta phylum (Fig. 2 a). The sequence of the nuclear rDNA spacers confirmed that the Strain WT3 isolate was closely related to Tetraselmis striata (Fig. 2 b). The 18S rRNA gene sequence from the Strain WT1 isolate presented a 99% similarity with Chlorella sp. Strain SAG 211-18 [ 58 ] and Chlorella sp. strain KAS012. This strain was placed among the Parachlorella-clade in the Trebouxiophyceae class and Chlorophyta phylum (Fig. 2 a). As previously reported [ 26 ], phylogenetic analyses of the 18S rRNA gene sequences placed StrainWT6 in the Scenedesmaceae (Sphaeropleales, Chlorophyceae) family (Fig. 2 c). This family has a monophyletic lineage within the Sphaeropleales order consisting of autosporic green algae from the genus Scenedesmus and its relatives. The 18S rRNA gene sequence from Strain WT6 showed ≥ 99.9% similarity with Scenedesmus ( Acutodesmus ) rubescens CCAP 232/1 [ 59 ], Scenedesmus ( Acutodesmus ) dissociatus UTEX 1537 [ 60 ], Scenedesmus ( Acutodesmus ) littoralis [ 61 ] , Scenedesmus ( Acutodesmus ) distendus Hegewald 1975-267 [ 60 ] and Scenedesmus deserticola BCP-EM2-VF3 [ 62 ]. Fig. 1 a Dendrogram based on the 18S rRNA gene sequence. Bootstrap values are given at the nodes. The scale bar represents the substitution percentage. Thalassiosira concaviuscula was used as the outgroup. GenBank accession numbers follow the species name in parentheses. b Dendrogram based on the rbc L-3P sequence. Bootstrap values are given at the nodes. The scale bar represents the substitution percentage. Vaucheria repens was used as the outgroup. GenBank accession numbers follow the species name in parentheses \n Fig. 2 a Dendrogram based on the 18S rRNA gene sequence. Bootstrap values are given at the nodes. The scale bar represents the substitution percentage. Nitzschia communis was used as the outgroup. b Dendrogram based on the nuclear rDNA spacers’ sequence (ITS-1, 5.8 S gene and ITS-2). Bootstrap values are given at the nodes. The scale bar represents the substitution percentage. Stichococcus bacillaris was used as the outgroup. c Dendrogram based on the 18S rRNA gene sequence. Bootstrap values are given at the nodes. The scale bar represents the substitution percentage. Ankistrodesmus stipitatus was used as the outgroup. GenBank accession numbers follow the species name in parentheses \n Performance of native microalgae strains As a contribution to the survey of microalgae strains which are native to the south-western Mediterranean, the five different newly isolated microalgae strains from north-eastern Tunisia were evaluated in semi-continuous mode at different dilution rates to determine the optimal dilution rate for each strain and to select the best performing strain that could be used in further experiments. Experiments were performed in semi-continuous mode at dilution rates ranging from 0.08 to 0.90 1/day. The diatom Amphora sp. exhibited a very low biomass concentration during the first batch phase as well as operational cultivation difficulties such as biofouling; hence, it was not cultivated further in semi-continuous mode. The influence of the dilution rate on biomass concentration and productivity for each strain of the four isolates that grew adequately is displayed in Fig. 3 a, b, respectively. For all the strains, the results demonstrated a typical trend of light-limited culture, where the biomass concentration decreased as the dilution rate increased; with a maximum biomass concentration of 2.00 g/L being measured (Fig. 3 a). The maximum biomass productivity (0.25 g/L day) was obtained with Tetraselmis sp. at 0.65 1/day, while the lowest (0.01 g/L day) was with Nitzschia sp. The highest biomass productivity determined in this study (0.25 g/L day) was lower than the 0.87 and 0.76 g AFDW/L day obtained from S. obliquus and C. sorokiniana DEO1412, respectively—cultivated in batch mode for 200 h [ 63 ], and the 0.47 g/L day yielded by C. pyrenoidosa in 1-L flask laboratory cultures after 16 days of incubation [ 64 ]. However, the productivity obtained is comparable to previously reported values; for instance, San Pedro et al. [ 25 ] obtained a biomass productivity of 0.23 g/L day from Tetraselmis suecica cultivated semi-continuously at 0.47 1/day. Nonetheless, the productivity reported here is higher than the 47.71, 39.85 and 31.55 mg/L day reported from 15-day batch-cultivated Chlorella ellipsoidea in 2-L, 20-L (indoor) and 200-L (outdoor) bubble-column bioreactors, respectively [ 65 ]. Fig. 3 Changes in biomass concentration ( a ), biomass productivity ( b ) and photosynthetic efficiency ( c ) of Tetraselmis sp., Chlorella sp., Nitzschia sp. and Scenedesmus sp. once steady state of the cultures was achieved as a function of the dilution rate applied The very low productivity obtained from the continuous cultivation of the Nitzschia sp. strain could be attributable to an inadequate culture medium or unsuitable cultivation conditions. Nitzschia sp. is a benthic species that tends to grow adhered to solid substrates, and when cultivated in suspension tends to form lumps. In addition, temperature and pH greatly affect the growth of this microalga [ 25 , 64 , 66 , 67 ]. In fact, microscopic observations of this strain carried out during the experiment revealed the presence of cell agglomerates as well as cell morphology alterations. Further experiments focusing on culture condition optimization should be carried out to enhance the biomass productivity of this diatom strain. The maximum photosynthetic activity of PSII for the different strains was measured daily as an indicator of the cells’ physiological status and the steady-state data are presented in Fig. 3 c. Of all the dilution rates applied, the Fv/Fm ratio for Tetraselmis sp. was the highest compared to the other strains, and remained constant in the 0.71–0.74 range, reflecting the high photosynthetic efficiency of this strain. Similarly, the marine species Chlorella sp. presented constant photosynthetic activity (0.69–0.62) and dropped to 0.52 at the washing-out dilution rate, suggesting that the decrease in biomass concentration at that dilution rate (0.50 1/day) was probably associated with high light intensity leading to the photoinhibition of the cells. The Fv/Fm ratio of the freshwater species Scenedesmus sp. did not vary significantly at any of the dilution rates and were in the 0.61–0.59 range whilst Nitzschia sp. expressed low Fv/Fm ratio values—as low as 0.57—which is in agreement with the biomass productivity obtained, and confirms the hypothesis that this strain’s culture was subject to stress. Nevertheless, the values reported here are similar to those found in previous studies [ 26 , 67 , 68 ]. Effect of the dilution rate on the biochemical composition of the biomass The biochemical composition of each native microalgae strain was evaluated once steady state was achieved (washing the reactor volume three times). Results shown in Fig. 4 a reveal that the protein content is strain dependent and varied from one genus to another. Scenedemus sp. presented the largest amount of proteins, increasing from 15.30% d.wt. at 0.10 1/day, to remain constant at all the higher dilution rates (ranging from 53 to 49.50% d.wt.). This would make it potentially interesting as a feedstock for protein-based food processes, following oil extraction, particularly for aquaculture species such as shrimps, molluscs and fish [ 69 , 70 ]. Similarly, the dilution rate seemed to have no effect on the protein content of Tetraselmis sp., which ranged between 16.70 and 19.30% d.wt., although a slight increase (up to 21.60% d.wt.) was noticed at 0.40 1/day. The values reported here were lower than those obtained by Khatoon et al. [ 71 ], who cultivated Tetraselmis sp. in batch mode at different salinities, yet were comparable to those reported by Kim et al. [ 72 ] for Tetraselmis sp. KCTC 12236BP after batch mode cultivation in an aerated photobioreactor. As for Chlorella sp., the protein content remained constant (25.60–28.10% d.wt.) at the first three dilution rates, then continuously declined as the dilution rate increased, dropping to 12.80% d.wt. at 0.50 1/day. An approximate range of values was observed by Wu and Miao [ 64 ], who cultivated C. pyrenoidosa indoors in batch mode at different nitrate concentrations. A different trend was observed for Nitzschia sp., whereby the protein content increased as the dilution rate increased from 15.10 to 20% d.wt., with a slight decrease at 0.25 1/day. This trend is in accordance with the data found by San Pedro et al. [ 25 ] using Nannochloropsis gaditana . Fig. 4 Variations in protein ( a ) and lipid ( b ) contents, and lipid productivity ( c ) of Tetraselmis sp., Chlorella sp., Nitzschia sp. and Scenedesmus sp. at steady state, as a function of the dilution rate applied Regarding lipid content and productivity, the variation as a function of the dilution rate was studied for the different isolates and the data are shown in Fig. 4 b, c. As previously pointed out [ 23 , 68 ], the crucial criterion in the selection of potential candidates for biofuel production is the lipid productivity rather than the lipid content. For all the isolates, no specific trend was observed relating to the dilution rate applied. For the marine isolate Chlorella sp., the total lipid content ranged from 12.30 to 17.90% d.wt. and the lipid productivity was in the 1.70–35.10 mg/L day range—the maximum (17.90%, 35.10 mg/L day) was recorded at 0.20 1/day. A similar lipid content range was determined by Selvarajan et al. [ 73 ], who cultivated five microalgae strains of the Chlorella genus in batch mode for 21 days, whereas Song et al. [ 23 ]. determined a lower lipid productivity (7.96 mg/L day) than that reported in this study for the same genus strain. The lipids content of Scenedesmus sp. ranged from 17.80 to 22.90% d.wt. and the lipid productivities were in the range of 28.30–37.40 mg/L day, with the lipid content peaking at 0.20 1/day. A comparable lipid content range was obtained by Yin-Hu et al. [ 74 ], who cultivated Scenedesmus sp. under phosphorus-starvation and -repletion conditions in batch mode. Nonetheless, the lipid productivities reported here were higher than those determined for Scenedesmus sp. in previous studies [ 23 , 75 , 76 ]. Nitzschia sp. demonstrated a notable capability for accumulating lipids (20.60–30.70% d.wt.), which decreased as the dilution rate increased. However, the lipid productivity was only 2.90–6.80 mg/L day due to the low biomass productivity. A similar result was observed by Song et al. [ 23 ] for diatom strains. Tetraselmis sp. exhibited lipid productivities ranging from 17.90 to 33.40 mg/L day, whilst the lipid content was in the 8.80–15.90% d.wt. range, where it continuously decreased at dilution rates higher than 0.40 1/day. This lipid range was lower than that previously reported for the same genus, Tetraselmis [ 71 , 72 ]. The differences in the biochemical composition results for the various isolates from the same phylum and genus demonstrate that they are not only caused by differences in culture conditions but also because the biochemical composition of the microalgae strain is an intrinsic characteristic and species specific. In the light of the results obtained, the marine isolate Chlorella sp. demonstrated the best compromise of lipid and biomass productivity at 0.20 1/day; therefore, this microalgae strain was selected for further experiments. Another critical criterion for the selection of potential microalgae candidates for biodiesel production (i.e., the direct transesterification of the biomass lipid fraction) is the qualitative fatty acid composition and the saponifiable lipids productivity (i.e., the fatty acid methyl esters [FAME] productivity). In fact, the fatty acid profile and structure significantly influence and dictate the physical and chemical properties of the biodiesel produced, including the octane number, viscosity, cold flow, oxidative stability and lubricity [ 68 , 77 ]. The saponifiable fatty acid content and profile (Fig. 5 a and Table 2 , respectively) varied widely amongst the different isolates with no specific trend being observed in relation to the distinct dilution rates applied. It is worth mentioning that the dilution rate is not the only variable impacting on lipid content and lipid productivity in continuous cultures. In fact, the relationship between the lipid productivity, the dilution rate, the medium composition and the average irradiance in microalgae cultures has been previously reported [ 78 ]. Besides, the culture growth status, i.e., the average irradiance within the reactor, influences the fatty acid profile and content, because the fatty acid content is a function of the light availability inside the culture and the extent of photolimitation–photoinhibition phenomena [ 78 ]. Considering SFA and MUFA in combination, Tetraselmis sp. presented a range of 33.40–53.90% of total saponifiable fatty acids—comparatively higher than the other two Chlorophyte isolates. The highest saturated and monosaturated FAME content, however, was observed for the diatom Nitzschia sp. (81.40%). As for FAME productivity, the highest value (20 mg/L day) was recorded for Chlorella sp. at 0.30 1/day followed by Tetraselmis sp. (16.90 mg/L day) at 0.20 1/day and then Scenedesmus sp. (16.30 mg/L day) at 0.50 1/day (Fig. 5 b). In a study conducted by San Pedro et al. [ 24 ], the selected strain for biodiesel production showed higher productivity of accumulated fatty acids up to 51 mg/L day comparing to the one here determined. Similarly, these reported results are far from those obtained by Münkel et al. [ 79 ] who cultivated the freshwater strain Chlorella vulgaris in batch mode achieving a fatty acid productivity of 0.39 g/L day. However, it is worth noting that these authors carried out a two-stage culture strategy where the second stage is nitrate or/and phosphate-depleted phase to enhance lipid accumulation; whereas in comparison with the present work, the cultivation was performed under sufficient nutrients conditions. Fig. 5 Variations in fatty acids content ( a ) and saponifiable lipids productivity ( b ) of Tetraselmis sp., Chlorella sp., Nitzschia sp. and Scenedesmus sp. at steady state, as a function of the dilution rate applied \n Table 2 Summary of fatty acid compositional profiles at steady state for algal lipids from the different isolates cultivated in semi-continuous mode at different dilution rates Chlorella sp. Scenedesmus sp. D, 1/day 0.08 0.20 0.30 0.38 0.45 0.50 0.10 0.20 0.30 0.40 0.50 14:0 – – – – – – – – – – – 16:0 – – – 17.40 15.40 14.80 – 16.40 16.50 16.30 16.00 16:1n7 33.10 16.60 14.50 2.30 – – 18.80 2.90 2.60 2.60 2.70 16:2n4 6.30 12.90 5.30 4.40 4.20 4.30 3.20 2.30 – 3.00 – 16:3n4 5.00 6.90 10.00 9.50 11.20 10.70 2.90 – – – – 16:4n1 – – 3.70 2.40 – – 9.10 11.80 13.50 13.70 13.80 18:0 1.80 – 3.50 – 2.80 3.20 – – 7.10 6.80 – 18:1n9 8.40 – 2.80 6.90 2.40 2.80 11.70 7.80 – – 6.90 18:1n7 – – – – – – – – – – – 18:2n6 – – – 14.20 13.40 12.90 – 14.10 11.30 9.70 8.40 18:3n3 9.30 13.20 – 28.10 30.00 28.30 16.20 24.50 28.10 29.90 31.90 18:4n3 16.40 19.50 – – – – 20.20 1.80 1.80 1.80 – 20:4n6 – – – – – – – – – – – 20:1n9 3.00 3.00 – 2.20 3.20 3.20 – – – – – 22:5n3 – – – – – – – – – – – SFA 1.80 – 3.50 17.40 18.20 18.00 – 16.40 23.50 23.10 16.00 MUFA 44.50 19.60 19.40 11.40 5.60 3.20 30.50 10.70 2.60 2.60 9.60 PUFA 37.00 52.50 57.50 58.70 58.90 56.30 51.60 54.60 54.80 58.10 54.10 Nitzschia sp. Tetraselmis sp. D, 1/day 0.10 0.20 0.30 0.10 0.20 0.30 0.40 0.50 0.60 0.80 0.90 14:0 6.60 7.60 6.30 – – – – – – – – 16:0 23.70 25.20 20.80 14.90 14.60 14.80 16.00 11.40 17.00 17.90 18.00 16:1n7 47.70 43.60 34.40 3.20 3.00 2.30 2.40 2.30 1.90 1.70 1.40 16:2n4 1.80 2.20 4.20 8.00 8.00 7.30 6.00 5.60 5.00 4.50 4.00 16:3n4 1.60 2.40 5.10 0.50 5.30 6.40 8.80 8.90 10.20 11.40 10.30 16:4n1 – – – 6.40 12.40 11.70 11.50 11.00 9.80 10.10 10.80 18:0 1.40 1.50 – 12.70 2.30 2.10 0.90 13.90 12.20 10.60 9.80 18:1n9 1.70 1.60 1.80 2.50 17.90 16.40 14.10 14.70 16.50 17.50 20.20 18:1n7 – – – – 13.00 14.50 13.30 – – – – 18:2n6 – – – 17.00 – – – 2.80 – – – 18:3n3 8.00 7.80 19.00 10.90 0.70 0.90 1.60 1.60 2.00 2.30 2.40 18:4n3 – – – 0.80 0.90 0.90 1.20 1.20 – 1.40 1.60 20:4n6 – – – 2.80 1.00 1.20 1.70 1.90 2.30 2.80 3.10 20:1n9 – – – – 3.00 3.10 2.60 2.40 2.20 2.20 1.80 22:5n3 7.70 7.90 13.00 – – – – – – – – SFA 31.80 34.40 27.00 27.60 16.90 16.90 16.90 25.30 29.30 28.50 27.80 MUFA 49.60 45.20 36.20 5.80 37.00 36.40 32.40 19.40 20.60 21.40 23.50 PUFA 11.00 12.50 22.40 46.50 28.30 28.40 31.00 33.20 29.40 32.60 32.30 Data are mean value of two repetitions SFA: Saturated fatty acids; MUFA: mono unsaturated fatty acids; PUFA: poly unsaturated fatty acids Palmitic and palmitoleic acids were the predominant fatty acids in most of the algal isolate extracts followed by C18:3n3, C18:1n9, C16:4n1 and C16:3n4 (Table 2 ). Similar classes of fatty acids were previously reported [ 80 ] with some differences in content that might be attributable to the diversity of microalgae strains studied and the culture conditions. The Chlorophyceae Scenedesmus sp. and the Trebouxiophyceae Chlorella sp. contained high C18:3n3 contents in the 16.20–31.90% range and 9.30–30% of total fatty acids, respectively. The highest amount of C16:1n7 was attributed to the diatom Nitzschia sp. (47.70% of total fatty acids at 0.10 1/day), whereas a considerable amount of C18:1 (2.50–30.90% of total fatty acids) was obtained from Tetraselmis sp. It is worth noting that, of all the different isolates, only Nitzschia sp. presented broad classes of fatty acids from lighter species such as C:14 (6.30–7.60% of total fatty acids) to heavier species like C22:5 (7.70–13% of total fatty acids). Comparing these FAME profiles with those of common and current worldwide biofuel feedstocks [ 77 ], the Trebouxiophyceae Chlorella sp. and the freshwater Scenedesmus sp. presented similar fatty acid profiles at high dilution rates to camelina, characterized by the high C18:3n3 content. Whereas, the flagellate Tetraselmis sp. showed (at almost all the dilution rates tested) a fatty acid composition more similar to tallow [ 77 ], marked by an abundance of C16:0 and C18:1 as well as a considerable C18:0 content (Table 2 ). In addition, a biodiesel quality estimation was conducted for the different isolates. In this study, the most important fuel properties frequently reported in the literature for the assessment of FAME suitability as a fuel (Table 3 )—kinematic viscosity, specific gravity, cloud point (CP), cetane number (CN), iodine value (IV) and high heating value (HHV)—were empirically determined using predictive models based on fatty acid composition [ 77 ]. The estimated values of CN, CP, HHV, specific gravity and kinematic viscosity for the different FAMEs derived from the isolates complied with almost all the common quality specifications, in accordance with standards ASTM D6751 in the US and EN 14214 in Europe. Furthermore, they are within the value ranges obtained from the widely used biodiesel feedstock (Table 3 ). The European standard is more stringent, designating a minimum CN value of 51; whereas, the US standard requires a minimum of 47. The CN is a measure of ignition quality in a diesel engine [ 68 , 77 ]. The results showed a range of 48.80–58.50 for the different strains, in compliance with the standards requirements. The iodine value (IV) is a measure of unsaturation, and thus reflects a biodiesel’s oxidative stability. ASTM D6751 does not designate any IV specification whereas EN 14214 includes a maximum of 120 g I 2 /100 g FAME. Tetraselmis sp. and Nitzschia sp. exhibited an IV range of 60.90–67.00 and 99.00–112.30 g I 2 /100 g FAME, respectively. This is lower than the prescribed limits, compared to the most common vegetable oils used for biodiesel production and lower than camelina (152.80 g I 2 /100 g FAME), soy (125.50 g I 2 /100 g FAME) and sunflower (128.70 g I 2 /100 g FAME) oils (Table 3 ) [ 77 ]. The kinematic viscosity is a measure of a biodiesel’s flow resistance and this increases as the saturation increases, leading to poor combustion, high emissions and oil dilution [ 77 ]. The results presented in Table 3 showed that kinematic viscosity values varied between 3.90 and 4.80 mm 2 /s, which falls within the specifications range. The diatom Nitzschia sp. demonstrated the highest kinematic viscosity values, of 4.80 mm 2 /s, probably due to its high SFA and MUFA contents (up to 81.30% of total fatty acids). Table 3 Comparison of estimated biodiesel properties of algal lipids from the different isolates as a function of the dilution rate along with the US and European specifications (B100), and those of common vegetable oils Chlorella sp. Scenedesmus sp. Nitzschia sp. D, 1/day 0.08 0.20 0.30 0.38 0.45 0.50 0.10 0.20 0.30 0.40 0.50 0.10 0.20 0.30 KV, at °C mm 2 /s 4.20 4.00 3.90 4.10 4.20 4.20 3.90 4.10 4.10 4.00 4.00 4.80 4.80 4.70 Specific gravity, Kg/L 0.88 0.88 0.88 0.88 0.88 0.88 0.88 0.88 0.88 0.88 0.88 0.87 0.87 0.87 CP, °C − 2.10 − 4.60 − 8.00 − 2.90 − 2.00 − 1.00 − 8.20 − 2.90 − 2.80 − 4.00 − 3.70 11.20 11.30 10.20 CN, min 51.80 50.60 48.90 51.40 51.90 52.40 48.80 51.40 51.50 50.90 51.00 58.50 58.50 58.00 IV, g I 2 /100 g 135.80 149.40 169.00 140.20 135.20 130.00 170.00 140.50 139.80 146.40 144.70 61.50 60.90 67.00 HHV, MJ/Kg 41.40 41.80 42.20 41.50 41.40 41.30 42.20 41.50 41.50 41.70 41.60 39.70 39.70 39.80 Tetraselmis sp. ASTM D6751-08 EN 14214 Camelina Sunflower Soy D, 1/day 0.10 0.20 0.30 0.40 0.50 0.60 0.80 0.90 KV, at °C mm 2 /s 4.40 4.40 4.40 4.40 4.40 4.50 4.40 4.40 1.9–6.0 3.5–5 3.8 4.42 4.26 Specific gravity, Kg/L 0.87 0.87 0.87 0.87 0.87 0.87 0.87 0.87 – 0.85–0.9 0.88 0.87 0.88 CP, °C 2.60 2.80 2.80 2.10 3.30 4.50 2.70 2.40 – – 3 2 – CN, min 54.20 54.30 54.30 53.90 54.60 55.10 54.30 54.00 47 51 50.4 51.1 51.3 IV, g I 2 /100 g 109.50 108.20 108.00 112.30 105.30 99.0 108.80 110.90 – 120 152.8 128.7 125.5 HHV, MJ/Kg 40.80 40.80 40.80 40.90 40.70 40.60 40.80 40.80 – – 45.2 40.6 39.7 Data are mean value of two repetitions KV: Kinematic viscosity; CP: cloud point; CN: cetane number; IV: iodine value; HHV: high heating value In this study, the estimated specific gravity (also referred to as fuel density), which affects engine performance [ 77 ], varied between 0.87 and 0.88 kg/L. This is in accordance with the globally accepted standard EN 14214. In contrast, the US and European fuel standards do not state any specification regarding the CP. This property, related to a biodiesel’s low-temperature performance, increases along with an increase in the presence of long-chain SFA resulting in poor cold flow properties [ 77 ]. The estimated CP value results for biodiesel from the isolates revealed a wide range, from − 8.20 to 11.30 °C, where the highest CP was determined for the diatom Nitzschia sp., indicating poor low-temperature performance. No specifications include HHV values; however, the values determined for the different isolates in the present study (39.70–42.20 MJ/kg) were within the range of common vegetable oil biodiesels and previously reported values from microalgae [ 73 , 77 ]. Sun et al. [ 81 ]. studied the properties of biodiesel from nine Chlorella strains. The authors found higher CP values comparing to those here reported; whereas lower IV values were determined. Also, Miao et al. [ 64 ] reported the biodiesel properties of Scenedesmus obliquus and Chlorella pyrenoidosa and showed that they meet several of ASTM specifications. In these both studies, similar range of CN was recorded comparing to that determined in the present work. Influence of outdoor conditions on the performance of the selected microalgae strain WT1 Chlorella sp. The effect of temperature and light prevailing outdoor conditions on the performance of the selected microalgae strain Chlorella sp. was studied. The effect of high and low temperature (38 °C and 10 °C) is shown in Fig. 6 a, b. Multivariate analyses based on Pillai’s Trace test showed that both temperature, at the two levels (38 °C and 10 °C), and the length of exposure to the tested temperature (time), had a highly significant effect on the photosynthetic parameters ( p < 0.001). The interaction between these factors also showed a synergy (Pillai’s trace value = 1.314; F = 2.516; p < 0.001). As can be observed in Fig. 6 a, the Fv/Fm decreased due to high temperature (38 °C) and recovery came about only after 155 min. In addition, the curve of light saturation intensity (Ik) exhibited a notable decrease starting at 125 min. On the other hand, Fig. 6 b displayed a different low-temperature-effect pattern (10 °C), where alpha constantly dropped from 0.15 to 0.09 at 215 min and then suddenly increased. Furthermore, the saturation irradiance showed a variable trend during the experiment time. Total Fv/Fm variability was mainly governed by the exposure time (74%); whereas, the temperature effect accounted for 90% and 98% on Ik and alpha, respectively. However, temperature and time exposure almost equally controlled ETR (58% and 42%, respectively). Fig. 6 Effect of temperature at 38 °C ( a ), at 10 °C ( b ) and high irradiance (1600 µmol/m 2 s) ( c ) on different photosynthetic parameters of the WT1 Chlorella sp. strain as a function of time Regarding the light effect (1600 µmol/m 2 s), the results for the photosynthetic parameter variations as a function of time are displayed in Fig. 6 c. They reveal that highlight intensity had a major impact on ETR and alpha ( p < 0.001) while influencing Ik less ( p < 0.005). The exposure time to high irradiance greatly affected ETR and Fv/Fm, according to multivariate test analyses using Pillai’s Trace test. Similarly, the interaction effect between these two factors was found to be highly significant (Pillai’s Trace value = 1.769; F = 2.677; p < 0.001). Additionally, a highly significant effect of this synergy was noted on ETR and Fv/Fm ( p < 0.001). The figure shows that both alpha and Fv/Fm sharply decreased after 35 min and 5 min, respectively, of exposure to high irradiance. Similar results were found by [ 82 ], where Fv/Fm reduced after 5 min of exposure to high irradiance. In fact, some studies have reported on photo-protective mechanisms that aim to protect photosynthetic reaction centers from over excitation caused by high irradiance or temperature variations. The determination for the corrected total sum of square partitioning indicated that ETR and Fv/Fm were 81% and 83% controlled by high irradiance, respectively, while alpha was controlled by the length of exposure to high light intensity (68%). Consequently, after evaluating all these results, the marine strain Chlorella sp. was selected as the most promising candidate for renewable fuel production due to its considerable biomass and lipid productivities (0.20 g/L day and 35.10 mg/L day, respectively, at 0.20 1/day), the suitability of its fatty acid profile with respect to biofuel properties, and its important FAME productivity. The investigation into this selected marine isolate’s photosynthetic characteristics showed accordance with the results, demonstrating that the Chlorella sp. strain is sufficiently tolerant and robust when faced with severe outdoor conditions."
} | 8,012 |
25221465 | null | s2 | 6,976 | {
"abstract": "Currently, most of the integrated sorting modules in the microfabricated DEP-based and fluorescent-activated cell sorters (μFACS) still suffer from low-throughput operation and require complex fabrication process (e.g. embedded electrodes) and high power consumption (e.g. electrokinetically-driven sorters). In this paper, we demonstrate an easy-to-fabricate, low-powered and high-speed sorting module (at a single cell level) using an on-chip integrated piezoelectric (PZT) actuator. By controlling the bending motion of the PZT actuator, we have investigated and verified the high-speed flow-switching and sorting capabilities both theoretically (dynamic simulation) and experimentally using beads and biological agents."
} | 180 |
21036663 | null | s2 | 6,977 | {
"abstract": "Protists account for the bulk of eukaryotic diversity. Through studies of gene and especially genome sequences the molecular basis for this diversity can be determined. Evident from genome sequencing are examples of versatile metabolism that go far beyond the canonical pathways described for eukaryotes in textbooks. In the last 2-3 years, genome sequencing and transcript profiling has unveiled several examples of heterotrophic and phototrophic protists that are unexpectedly well-equipped for ATP production using a facultative anaerobic metabolism, including some protists that can (Chlamydomonas reinhardtii) or are predicted (Naegleria gruberi, Acanthamoeba castellanii, Amoebidium parasiticum) to produce H(2) in their metabolism. It is possible that some enzymes of anaerobic metabolism were acquired and distributed among eukaryotes by lateral transfer, but it is also likely that the common ancestor of eukaryotes already had far more metabolic versatility than was widely thought a few years ago. The discussion of core energy metabolism in unicellular eukaryotes is the subject of this review. Since genomic sequencing has so far only touched the surface of protist diversity, it is anticipated that sequences of additional protists may reveal an even wider range of metabolic capabilities, while simultaneously enriching our understanding of the early evolution of eukaryotes."
} | 347 |
21418498 | null | s2 | 6,981 | {
"abstract": "The Columbia River is a major source of dissolved nutrients and trace metals for the west coast of North America. A large proportion of these nutrients are sourced from the Columbia River Estuary, where coastal and terrestrial waters mix and resuspend particulate matter within the water column. As estuarine water is discharged off the coast, it transports the particulate matter, dissolved nutrients and microorganisms forming nutrient-rich and metabolically dynamic plumes. In this study, bacterial manganese oxidation within the plume and estuary was investigated during spring and neap tides. The microbial community proteome was fractionated and assayed for Mn oxidation activity. Proteins from the outer membrane and the loosely bound outer membrane fractions were separated using size exclusion chromatography and Mn(II)-oxidizing eluates were analysed with tandem mass spectrometry to identify potential Mn oxidase protein targets. Multi-copper oxidase (MCO) and haem-peroxidase enzymes were identified in active fractions. T-RFLP profiles and cluster analysis indicates that organisms and bacterial communities capable of oxidizing Mn(II) can be sourced from the Columbia River estuary and nearshore coastal ocean. These organisms are producing up to 10 fM MnO₂ cell⁻¹ day⁻¹. Evidence for the presence of Mn(II)-oxidizing bacterial isolates from the genera Aurantimonas, Rhodobacter, Bacillus and Shewanella was found in T-RFLP profiles. Specific Q-PCR probes were designed to target potential homologues of the Aurantimonas manganese oxidizing peroxidase (Mop). By comparing total Mop homologues, Aurantimonas SSU rRNA and total bacterial SSU rRNA gene copies, it appears that Aurantimonas can only account for ~1.7% of the peroxidase genes quantified. Under the broad assumption that at least some of the peroxidase homologues quantified are involved in manganese oxidation, it is possible that other organisms oxidize manganese via peroxidases."
} | 489 |
29467397 | PMC6018805 | pmc | 6,982 | {
"abstract": "A critical step in the biogeochemical cycle of sulfur on Earth is microbial sulfate reduction, yet organisms from relatively few lineages have been implicated in this process. Previous studies using functional marker genes have detected abundant, novel dissimilatory sulfite reductases (DsrAB) that could confer the capacity for microbial sulfite/sulfate reduction but were not affiliated with known organisms. Thus, the identity of a significant fraction of sulfate/sulfite-reducing microbes has remained elusive. Here we report the discovery of the capacity for sulfate/sulfite reduction in the genomes of organisms from 13 bacterial and archaeal phyla, thereby more than doubling the number of microbial phyla associated with this process. Eight of the 13 newly identified groups are candidate phyla that lack isolated representatives, a finding only possible given genomes from metagenomes. Organisms from Verrucomicrobia and two candidate phyla, Candidatus Rokubacteria and Candidatus Hydrothermarchaeota, contain some of the earliest evolved dsrAB genes. The capacity for sulfite reduction has been laterally transferred in multiple events within some phyla, and a key gene potentially capable of modulating sulfur metabolism in associated cells has been acquired by putatively symbiotic bacteria. We conclude that current functional predictions based on phylogeny significantly underestimate the extent of sulfate/sulfite reduction across Earth’s ecosystems. Understanding the prevalence of this capacity is integral to interpreting the carbon cycle because sulfate reduction is often coupled to turnover of buried organic carbon. Our findings expand the diversity of microbial groups associated with sulfur transformations in the environment and motivate revision of biogeochemical process models based on microbial community composition.",
"conclusion": "Conclusions By the Proterozoic Eon, sulfate reduction had become a significant biological process in the oceans [ 70 , 71 ]. Based on phylogenomic arguments and isotopic records, it was suggested that the capacity to reduce sulfite to sulfide emerged in thermophilic archaea around 3.5 billion years ago, and that mesophilic sulfate reducers evolved only after the rise in atmospheric oxygen level [ 2 , 72 ]. Our findings indicate a complex evolutionary history of this capacity involving extensive LGT of dsr genes. Consequently, it may be impossible to constrain the specific lineage in which this metabolism first appeared. The ability for a DsrAB-based dissimilatory sulfur metabolism is now predicted in a much wider diversity of mesophilic bacterial and archaeal groups than was recognized previously. We conclude that many groups of microorganisms now known to have genes involved in dissimilatory sulfur metabolism impact biogeochemical processes in marine and terrestrial sediments, aquifers, wetlands, methane seeps, coastal marshes and estuaries, as well as agricultural and human microbiomes. Many are organisms from well-studied phyla, but still novel at the genus to class levels, but others are organisms from candidate phyla known only based on their genomes. The results underline the value of genomic analyses for prediction of key ecosystem capacities that cannot be made based on rRNA gene surveys and motivate targeted cultivation strategies for organisms currently lacking laboratory tractable representatives. Finally, these findings will better inform future microbial trait-based ecosystem models that can predict the outcomes of global change on biogeochemical processes and planetary elemental cycles [ 73 ].",
"introduction": "Introduction The cycling of sulfur is one of Earth’s major biogeochemical processes. Sulfate reduction in conjunction with sulfur disproportionation may be an early evolved microbial metabolism, given evidence for biological fractionation of sulfur isotopes around 3.5 billion years ago [ 1 , 2 ], and it remains an important energy metabolism for anaerobic life [ 3 ]. In natural ecosystems, human microbiomes, and engineered systems, this process is important because the product hydrogen sulfide (H 2 S) can be toxic [ 4 ], can corrode steel [ 5 ], and sour oil reservoirs [ 6 ]. Overall, sulfate reduction is a primary driver in the carbon cycle, and is responsible for a large part of the organic carbon flux to CO 2 in marine sedimentary environments [ 7 , 8 ] and in wetlands [ 9 ]. Importantly, the coupling of sulfate/sulfite reduction to oxidation of H 2 , small chain fatty acids, or other carbon compounds limits the availability of these substrates to other organisms like methanogens and alters the energetics via syntrophic interactions [ 10 , 11 ]. All of these processes also impact methane production. Given the many reasons why the biological conversion of sulfate/sulfite to sulfide is important, it is vital that we understand which organisms can carry out the reactions and the pathways involved. The canonical microbial pathway for dissimilatory sulfate reduction involves the initial reduction of sulfate to sulfite by a combination of sulfate adenylyltransferase (Sat) and adenylyl-sulfate reductase (AprBA) followed by reduction of sulfite by sulfite reductases. Sulfite reductase genes catalyze the rate-limiting steps in the global sulfur cycle [ 12 , 13 ] and confer bacteria and archaea the ability to grow via reduction of sulfite, and can function in reverse in some organisms that disproportionate or oxidize elemental sulfur [ 14 – 16 ]. Four different groups of sulfite reductases function in dissimilatory sulfur metabolism. Of these, siroheme-dependent dissimilatory sulfite reductase (dsr), siroheme-dependent anaerobic sulfite reductase (asr) genes, and octaheme cytochrome c sulfite reductase ( mccA ) catalyze the reduction of sulfite to sulfide, while reverse dissimilatory sulfite reductase genes (rdsr) are involved in sulfur oxidation. All of these sulfite reductases except for mccA constitute an ancient lineage of enzymes that may predate the separation of Bacteria and Archaea [ 17 ]. The taxonomic distribution of dissimilatory sulfite reductases has been considered to be restricted to organisms from selected bacterial and archaeal phyla [ 18 ]. Only organisms from nine microbial phylum-level lineages, namely Deltaproteobacteria , Firmicutes , Thermodesulfobacteria , Actinobacteria , Nitrospirae , Caldiserica , Euryarchaeota , Crenarcheota , and Aigarchaeota are known to possess the genetic capacity to reduce sulfite to sulfide using the dsr system. The asr enzymes have a far more limited distribution and are known to be present only in organisms from four phylum-level lineages, Gammaproteobacteria , Firmicutes , Spirochaetes , and Fusobacteria . The distribution of MccA enzymes is restricted to organisms from Epsilonproteobacteria [ 19 ] and Gammaproteobacteria [ 20 ]. Finally, the rdsr enzyme complex for sulfur oxidation is associated with organisms from five phylum-level lineages including Alphaproteobacteria , Betaproteobacteria , Gammaproteobacteria , Deltaproteobacteria , and Chlorobi . This diversification of sulfite reductases was likely driven by speciation and functional divergence, and to a lesser extent, lateral gene transfer (LGT) [ 21 ]. The recent availability of thousands of genomes from organisms belonging to many newly sampled phyla has provided the opportunity to test for the presence of sulfite reductase genes in bacteria and archaea that have not previously been associated with dissimilatory sulfur metabolism [ 22 ]. Here we use shotgun metagenomic sequencing and recovery of metagenome-assembled genomes (MAGs) from a diverse set of marine and terrestrial environments to show that organisms from novel lineages contain sulfite reductases that implicate them in the dissimilatory cycling of sulfur. In total, we more than doubled the number of microbial lineages that can catalyze dissimilatory sulfate/sulfite reduction or sulfur oxidation in the environment. We shed light on the complicated evolutionary history of dissimilatory sulfite reductases and show that LGT of catalytic sulfite reductase genes is much more common than previously thought.",
"discussion": "Results and discussion To investigate the diversity of microorganisms that contain sulfite reductases involved in dissimilatory sulfate/sulfite reduction or sulfur oxidation in the environment, we analyzed genomes reconstructed from metagenomic sequence datasets recovered from six distinct terrestrial and marine subsurface environments where geochemical conditions have suggested active microbial sulfur cycling. Our sampling sites included an aquifer adjacent to the Colorado River, USA [ 24 , 26 ], a deep subsurface CO 2 geyser in Utah, USA [ 27 ], a deep borehole in Japan [ 23 ], an acidic sulfide mine waste rock site in Canada, deep subseafloor basaltic crustal fluids of the hydrothermally active Juan de Fuca ridge flank in the Pacific Ocean [ 25 ], and an acidic peatland in Germany [ 50 ]. In total, we searched in excess of 4000 near-complete MAGs for the presence of sulfite reductase genes. Identification of dissimilatory sulfur cycling organisms from MAGs We identified sulfite reductase genes in 123 near-complete microbial genomes (Supplementary Table 1 ). Phylogenetic analyses using a set of 16 concatenated ribosomal proteins (RP) and the small subunit ribosomal (SSU) RNA gene show that these genomes belong to organisms from 20 distinct phylum-level lineages (Table 1 ), 13 of which were not known to have dsr genes [ 18 ]. In addition, we identified anaerobic sulfite reductase (asr) genes required for sulfite reduction in three bacterial groups not previously reported to have this capacity [ 51 ]. All of the identified catalytic protein subunits (DsrA, DsrB, and AsrC) contained all conserved sulfite reductase residues and secondary structure elements for the formation of α helices and β sheets [ 36 ] (Supplementary Figs. 1 – 3 ). Table 1 Details of lineages involved in dissimilatory sulfur cycling as identified in this study Phylum-level lineage No. of genomes reported Potential contribution to sulfur cycle Mechanism Source Electron donor Hydrogen Fatty acid metabolism Organic C \n Acidobacteria \n \n 3 \n \n Sulfate/sulfite reduction \n \n dsr \n \n A, D \n \n Yes \n \n Yes \n \n Yes \n \n Actinobacteria \n 3 Sulfate/sulfite reduction, or sulfur oxidation, or both dsr – Yes No Yes \n Armatimonadetes \n \n 1 \n \n Sulfate/sulfite reduction \n \n dsr \n \n C \n \n Yes \n \n No \n \n Yes \n \n Candidatus \n Desantisbacteria \n \n 4 \n \n Sulfate/sulfite reduction \n \n dsr \n \n B, C \n \n Yes \n \n No \n \n Yes \n \n Candidatus \n Falkowbacteria \n \n 8 \n \n Unknown \n \n Unknown \n \n A \n \n Yes \n \n No \n \n Yes \n \n Candidatus \n Hydrothermarchaeota \n \n 4 \n \n Sulfate/sulfite reduction, or sulfur oxidation, or both \n \n dsr \n \n E \n \n Yes \n \n No \n \n Yes \n \n Candidatus \n Lambdaproteobacteria \n \n 5 \n \n Sulfite reduction, or sulfur oxidation, or both \n \n dsr \n \n A \n \n Yes \n \n Yes \n \n Yes \n \n Candidatus \n Muproteobacteria \n \n 14 \n \n Sulfur oxidation \n \n rdsr \n \n A \n \n Yes \n \n Yes \n \n Yes \n \n Candidatus \n Omnitrophica \n \n 2 \n \n Sulfite reduction \n \n asr \n \n A, B \n \n Yes \n \n Yes \n \n Yes \n \n Candidatus \n Riflebacteria \n \n 4 \n \n Sulfite reduction \n \n asr \n \n A, B \n \n Yes \n \n Yes \n \n Yes \n \n Candidatus \n Rokubacteria \n \n 8 \n \n Sulfur oxidation \n \n dsr \n \n A \n \n No \n \n Yes \n \n Yes \n \n Candidatus \n Schekmanbacteria \n \n 1 \n \n Sulfate/sulfite reduction \n \n dsr \n \n A \n \n No \n \n Yes \n \n Yes \n \n Candidatus \n Zixibacteria \n \n 2 \n \n Sulfate/sulfite reduction \n \n dsr \n \n B \n \n Yes \n \n Yes \n \n Yes \n \n Chloroflexi \n 2 Sulfate/sulfite reduction dsr A Yes Yes Yes \n Deltaproteobacteria \n 34 Sulfate/sulfite reduction dsr A, C Yes Yes Yes \n Ignavibacteria \n \n 5 \n \n Sulfate/sulfite reduction \n \n dsr \n \n A, B \n \n Yes \n \n Yes \n \n Yes \n \n Nitrospinae \n \n 3 \n \n Sulfur oxidation \n \n rdsr \n \n A \n \n Yes \n \n Yes \n \n Yes \n \n Nitrospirae \n \n 2 \n \n Sulfur oxidation \n \n rdsr \n \n A \n \n No \n \n Yes \n \n Yes \n \n Nitrospirae \n 19 Sulfate/sulfite reduction dsr A, C Yes No Yes \n Planctomycetes \n \n 1 \n \n Sulfate/sulfite reduction \n \n dsr, asr \n \n A, B \n \n Yes \n \n Yes \n \n Yes \n \n Verrucomicrobia \n \n 1 \n \n Sulfate/sulfite reduction, or sulfur oxidation, or both \n \n dsr \n \n F \n \n Yes \n \n No \n \n Yes \n Sampling sources are indicated by letters: A—Aquifer at Rifle, Colorado, USA; B—Deep subsurface in Japan; C—CO 2 geyser at Green River, Utah, USA; D—Glencore Mine, Canada; E—Juan de Fuca ridge flank marine subsurface fluids; F—Natural Peatland in Germany. Newly identified lineages are shown in bold. Contribution to sulfur cycle for DsrAB-containing organisms were decided as described in Table 2 . Dissimilatory sulfite reductase containing organisms Given our interest in identifying organisms with the capacity to produce sulfide, we initially searched the genomes for operons that contained genes encoding DsrD [ 52 ]. This gene was considered a marker for sulfite reduction because it is absent in bacteria that use the rdsr pathway for sulfur oxidation [ 53 ]. It is however important to note that the dsrD gene is present and highly expressed in sulfur disproportionating organisms like Desulfurivibrio alkaliphilus that cannot be distinguished from canonical sulfate-reducing bacteria using gene synteny or other genomic features [ 16 ]. Although the exact function of the DsrD protein is unclear, the presence of winged-helix domains in its structure and its association with other core proteins of the dsr complex ( dsrABC ) suggest a regulatory role in bacterial sulfite reduction [ 37 ]. We identified 78 genomes that encode at least dsrABCD (Supplementary Fig. 4 ). A multiple alignment of DsrD sequences confirmed highly conserved residues, indicating that the proteins are likely active (Supplementary Fig. 5 ). These putative sulfate/sulfite-reducing microorganisms affiliate with eight distinct phyla not previously reported to be capable of these processes. Four are phyla with isolated representatives ( Acidobacteria , Armatimonadetes , Ignavibacteria , Planctomycetes ) and four are uncultivated candidate phyla ( Candidatus Zixibacteria, Candidatus Schekmanbacteria, Candidatus Desantisbacteria, Candidatus Lambdaproteobacteria) (Fig. 1a ). Fig. 1 A. Concatenated DsrAB protein tree showing the diversity of organisms involved in dissimilatory sulfur cycling using the dsr system. Lineages in blue contain genomes reported in this study. Phylum-level lineages with first report of evidence for sulfur cycling are indicated by blue letters. Only bootstrap values >50 are shown. The complete tree is available with full bootstrap support values as Additional Data File S2. b Concatenated AsrABC protein tree showing the diversity of organisms that possess the anaerobic sulfite reductase system. Lineages in colors were identified in this study. Only bootstrap values >50 are shown Importantly, organisms from Verrucomicrobia and two candidate phyla, Candidatus Rokubacteria and Candidatus Hydrothermarchaeota, lack dsrD genes and their dsrAB sequences form completely novel lineages outside the four known main phylogenetic DsrAB clusters, namely the reductive bacterial-type, the oxidative bacterial-type, the reductive archaeal-type, and the second dsrAB copies of Moorella species (Fig. 1a ). To determine the earliest evolved and most basal lineages in the DsrAB tree, we performed paralogous rooting analysis on a representative subset of sequences. In accordance with previous reports, our results show that the second copies of dsrAB in Moorella spp. likely represent the most basal DsrAB branch [ 18 ]. This was followed by the newly identified sequences from Candidatus Rokubacteria, Verrucomicrobia , and Candidatus Hydrothermarchaeota (Fig. 2 ). Interestingly, Candidatus Hydrothermarchaeota sequences were not monophyletic with one sequence (JdFR-18 JGI24020J35080_1000005) clustering with Verrucomicrobia and Candidatus Rokubacteria, while the remaining were affiliated with bacterial-type DsrAB. Other organisms lacking dsrD genes cluster together with organisms known to be sulfur oxidizers in the dsrAB tree. Based on this clustering, the group implicated in sulfur oxidation using the rdsr pathway now includes bacteria from three additional phylum-level lineages: Nitrospirae , Nitrospinae, and Candidatus Muproteobacteria (Fig. 1a ). Fig. 2 Paralogous rooting analysis of DsrAB. Bayesian inference tree showing the phylogenetic relationship between DsrA and DsrB (50 sequences, 377 alignment positions). Arrow indicates outgroup of other sulfite, non-DsrAB reductase superfamily (COG2221) sequences. Branch supports (posterior probability) higher than 0.9 are indicated by black circles. DsrA/DsrB sequences from this study are marked in bold. Assignment of oxidative/reductive, bacterial/archaeal-type DsrAB is according to Müller et al. [ 18 ] Anaerobic sulfite reductase-encoding organisms The asr pathway for sulfite reduction was found in three bacterial phyla not previously known to possess this pathway, namely Planctomycetes and members of two candidate phyla, Candidatus Omnitrophica and Candidatus Riflebacteria. Concatenated protein trees of all three subunits AsrA, AsrB, and AsrC showed that sequences from these phyla clustered with those from Firmicutes , suggesting that they were acquired by LGT (Fig. 1b ). Investigations into the operon structure of the asr complex revealed that while organisms from Planctomycetes and Candidatus Riflebacteria had a canonical gene organization in the order asrA , asrB , and asrC , Candidatus Omnitrophica had a fourth gene ( asrD ) as an insertion between asrB and asrC subunits. Analyses of conserved domains show that AsrD is related to the family of formate and nitrite transporters (pfam01226, COG2116, TIGRfam00790). We hypothesize that this may in fact serve as a bisulfide channel associated with dissimilatory sulfite reduction using the asr enzyme complex as observed in Clostridium difficile [ 54 ]. dsrD genes in candidate phyla radiation organisms Surprisingly, we identified dsrD genes in eight genomes of organisms affiliating with Candidatus Falkowbacteria, putatively symbiotic bacteria within the Parcubacteria superphylum of the candidate phyla radiation (CPR) [ 55 ]. There is no indication of the presence of other dsr genes in these genomes. Given the predicted close physical and metabolic interactions between CPR bacteria and their hosts, we suggest that this small protein could modulate host metabolism, as sometimes occurs with viruses/phage and their hosts [ 56 ]. CPR organisms are common in aquifers where conditions oscillate between oxic and anoxic [ 24 , 26 ]. The predicted Falkowbacteria DsrD protein sequences cluster with sequences from well-characterized Deltaproteobacteria capable of sulfate reduction, suggesting that deltaproteobacterial sulfate reducers served as dsrD -donors during LGT to these CPR bacteria (Supplementary Fig. 6 ). Considering the presence of dsrD genes in CPR organisms and putative sulfur-oxidizing/sulfur disproportionating bacteria [ 16 ], we propose that dsrD is not a good marker for sulfite reduction. Therefore, we suggest an alternate set of rules for utilizing a combination of dsr genes to distinguish DsrAB-based sulfite reduction from sulfur oxidation on the basis of genomic features (Table 2 ). Table 2 Suggested rules for determination of direction of dissimilatory sulfur metabolism for uncultivated organisms Contribution to sulfur cycle Suggested rules Sulfate reduction (to sulfide) aprBA , sat are present, DsrAB cluster with reductive DsrAB, dsrD gene is present, and dsrEFH are absent Sulfite reduction (to sulfide) DsrAB cluster with reductive DsrAB, dsrD gene is present, and dsrEFH are absent Sulfate reduction (to sulfite) aprBA and sat are present Sulfur oxidation (to sulfite) DsrAB do not cluster with reductive DsrAB, dsrEFH are present, and dsrD is absent Sulfur oxidation (to sulfate) DsrAB do not cluster with reductive DsrAB, dsrEFH are present, and dsrD is absent, aprBA and sat are present Sulfite reduction, or sulfur oxidation, or both dsrD and dsrEFH are present. Sulfur oxidation (to sulfite) as a part of sulfur disproportionation DsrAB cluster with reductive DsrAB, dsrD gene is present, and dsrEFH are absent Sulfur disproportionating organisms cannot be differentiated from sulfite-reducing organisms on genomic features alone. Details of individual dsr genes are specified in Supplementary Table 6 . Lateral gene transfer of DsrAB sulfite reductases Prior analyses have suggested that LGT has influenced the evolution of dsrAB among extant microorganisms but only by comparably few events among major taxonomic lineages [ 14 , 18 , 57 ]. We used a comparison of 16S ribosomal RNA and concatenated DsrAB protein trees to reevaluate the extent to which LGT has influenced the organismal distribution of dsrAB genes (Fig. 3 ). Mismatching branching pattern between the two trees indicates that dsrAB has been introduced into most of the candidate phyla members by multiple independent LGT events. Our analyses show that organisms from five bacterial and archaeal phyla, Deltaproteobacteria , Nitrospirae , Candidatus Hydrothermarchaeota, Actinobacteria , Chloroflexi likely acquired dsrAB genes in multiple events. Amongst these, Nitrospirae and Deltaproteobacteria displayed the highest number of LGT involving five independent events spanning across both reductive and oxidative branches of the DsrAB tree. These findings provide evidence for the complex evolutionary history of dsr genes. Currently, it may not be possible to identify the specific lineage in which sulfite reduction first appeared; however, our extensive dataset sets the stage for future studies to investigate the evolution of dissimilatory sulfur metabolism. Fig. 3 Comparison of 16S rRNA and concatenated DsrAB trees for sulfate/sulfite-reducing microorganisms. Sequences are grouped at the phylum level. Trees were constructed using a consensus of neighbor-joining and maximum-likelihood phylogenies with 1000 bootstrap re-samplings each. Each phylum is colored differently to identify LGT based on inconsistent branching patterns. Phylum names with an asterisk represent sulfate/sulfite-reducing lineages that were discovered in this study. Numbers indicate number of independent LGT events associated with the specific phylum. Complete phylogenetic trees with bootstrap values are available in Data Files S3–S6. LGT events involving oxidative-type DsrAB for Nitrospirae (2 events) and Deltaproteobacteria (1 event) are not shown Sulfate vs. sulfite reduction in organisms To determine whether these newly identified organisms reduce sulfate vs. sulfite to sulfide we looked for the genes involved in the reduction of sulfate to sulfite, specifically adenylyl-sulfate reductase reductase subunits A and B ( aprBA ), sulfate adenylyl transferase ( sat ), and quinone-interacting membrane-bound oxidoreductase subunits A, B, and C ( qmoABC ) [ 58 – 60 ]. Organisms from three phyla, the dsr-containing Candidatus Lambdaproteobacteria, and asr-containing Candidatus Riflebacteria, and Candidatus Omnitrophica, lacked genes for the reduction of sulfate to sulfite suggesting that they were sulfite reducers. Sulfite utilized by these organisms may derive from the environment or is produced inside the cell as part of other sulfur metabolism pathways such as tetrathionate or thiosulfate reduction, sulfur disproportionation, or by organosulfonate respiration. This suggests that recent genome-based observations supporting potential “metabolic handoffs” between organisms (transfer of metabolites associated with energy metabolism) in the oxidative cycle of sulfur [ 24 , 26 , 61 ] likely extend to the reductive cycle as well [ 62 ]. Interestingly, Candidatus Rokubacteria whose DsrAB sequences represent a novel deep-branching lineage in the DsrAB tree also have apr , sat , and qmo genes that are required for sulfate reduction or sulfite oxidation. Phylogenetic analyses of the individual dsr proteins shows that the sulfate reduction system of Candidatus Rokubacteria is of mosaic evolutionary origin (Fig. 4 ) (Supplementary Figs. 7 – 9 ). Fig. 4 Dsr operon structure and enzymatic roles of proteins involved in sulfate reduction in Candidatus Rokubacteria. Purple: genes involved in sulfate reduction to sulfite. Orange: putative enzymatic roles of genes; blue: microbial lineages with closest homologs as determined by phylogeny/BLAST against NCBI GenBank. APS refers to adenosine-5′-phosphosulfate. Green: genes involved in sulfite reduction to sulfide. This is the first case in which dsrE , dsrF , and dsrH genes are present in organisms other than sulfur-oxidizing bacteria Prevalence of dsrT in dsrAB -containing microorganisms In addition to dsrD , we sought evidence for hypothetical genes in proximity to known dsr genes that may help in distinguishing between DsrAB-based sulfate/sulfite reduction and sulfur oxidation pathways. We identified a hypothetical gene that encodes for the N-terminal domain of an anti-sigma factor antagonist protein [ 63 ] that almost always occurs within the operon encoding dsr genes (Fig. 5 ; Supplementary Fig. 4 ). This hypothetical protein is part of a protein family that includes the Bacillus subtilis RsbT co-antagonist protein rsbRD, which are important components of the stressosome and function as negative regulators of the general stress transcription factor sigma-B [ 64 ]. This gene is unique to DsrAB-based sulfite-reducing organisms and is mostly absent in recognized sulfur-oxidizing organisms, except for those within the phylum Chlorobi (Supplementary Fig. 10 ). We refer to this gene as ‘ dsrT ’ in accordance with homologous genes in phototrophic green sulfur bacteria from the phylum Chlorobi [ 65 ]. This gene always precedes the electron transport components encoded by dsrMKJOP genes [ 66 ] and is fused with dsrM in some organisms (Supplementary Fig. 11 ). Fused dsrT - dsrM genes are oriented with dsrT in the N-terminal and dsrM in the C-terminal, thereby maintaining the gene order observed in canonical dsr operons: dsrT , dsrM , dsrK . From structural predictions and conserved motifs, we hypothesize that it likely performs a regulatory function (Supplementary Fig. 12 ). Fig. 5 Dsr operon structure in previously reported (black names) and newly reported groups (blue names). Interestingly, and in contrast to the previously studied organisms for which the operon is interrupted (=SS=), the entire dsr pathway (including electron transport chain and ancillary proteins) is often encoded in a single genomic region dsrEFH in the newly discovered dsrAB -containing microorganisms Recent studies looking into the distribution of genes associated with the dsr operon have suggested that dsrE , dsrF , and dsrH are unique to sulfur oxidizing microorganisms and are absent in sulfate/sulfite-reducing and putative sulfur disproportionating microorganisms [ 16 , 38 , 67 ]. In sulfur-oxidizing microorganisms, DsrEFH can serve as an effective sulfur donor for DsrC [ 68 ]. On the other hand, co-located dsrE , dsrF , and dsrH genes are present in ~24% of the newly identified DsrAB-encoding microorganisms (30 out of 123 genomes) (Supplementary Fig. 4 ). These dsrEFH genes were identified in organisms from six phylum-level lineages, Actinobacteria , Candidatus Rokubacteria, Candidatus Lambdaproteobacteria, Candidatus Muproteobacteria, Nitrospirae , and Nitrospinae . Phylogenetic analysis of all identified DsrEFH shows that they cluster with well-characterized sulfur-oxidizing organisms (Fig. 6 ). The presence of dsrEFH genes (with well-known roles in sulfur oxidation) in organisms from Actinobacteria and Candidatus Lambdaproteobacteria is perplexing since these organisms also possess the dsrD gene that is unique to DsrAB-containing sulfite-reducing organisms. Further, the presence of dsrEFH genes in organisms within the phylum-level lineage Candidatus Rokubacteria (with a novel deep-branching clade of DsrAB) suggests that these organisms are likely involved in sulfur oxidation rather than sulfite reduction. Finally, sequences from Candidatus Muproteobacteria formed two distinct clades (Group 1, Group 2) and clustered with two separate groups, Alphaproteobacteria and Gammaproteobacteria respectively. All C andidatus Muproteobacteria with dsrEFH sequences in Group 1 also possessed a second copy of these genes that clustered with sequences from Magnetofaba australis and Candidatus Rokubacteria. Fig. 6 Concatenated DsrEFH protein tree inferred by maximum likelihood. Phylum-level lineages with first report of the presence of dsrEFH genes are shown in blue (from organisms with unknown-type DsrAB) and orange (from organisms with oxidative type DsrAB). Homologous TusBCD from E. coli and S. enterica were used to root the tree. Only bootstrap values >50 are shown Electron donors and other metabolic potential associated with sulfur cycling In order to better understand the energy metabolism and ecology of these newly identified organisms, we investigated potential electron donors for putative sulfate/sulfite reduction. Specifically, we targeted genes involved in the oxidation of hydrogen [ 69 ] (Ni–Fe hydrogenase groups I, IIa, IIb, IIIa, IIIb, IIIc, IIId, Fe–Fe hydrogenase groups A, B1/B2) and transformation of organic carbon compounds (genes involved in breakdown of cellulose, hemicellulose, chitin, pectin, starch, amino sugars, other monosaccharides, and polysaccharides) [ 49 ] and short chain fatty acids. Our analyses show that organisms from 12 of the 13 newly identified putative sulfate/sulfite-reducing DsrAB-containing lineages identified in this study possess the ability to utilize hydrogen as an electron donor (Supplementary Table 2 ). On the other hand, organisms from all 13 lineages possessed the ability to breakdown complex carbon compounds although the diversity of genes encoding for specific carbohydrate-active enzymes varied greatly across phyla (Supplementary Table 3 ). Organisms from 8 of the 13 putative sulfate/sulfite-reducing lineages possessed the ability to oxidize short chain fatty acids by beta-oxidation (Supplementary Table 4 ). In order to understand if these newly identified organisms were heterotrophs or autotrophs, we looked at the carbon fixation potential encoded in the genomes. We identified three different carbon fixation mechanisms, the Calvin–Benson cycle (CBB), the reverse (reductive) tricarboxylic acid (rTCA) cycle, and the Wood–Ljungdahl pathway in ~50% of all organisms (Supplementary Table 5 ). In total, 11 organisms contained genes encoding for the CBB cycle with 10 possessing the Form I RuBisCO and 1 organism possessed the Form II RuBisCO. Genes for the Wood–Ljungdahl pathway were encoded in 42 genomes while genes for the rTCA cycle were encoded in 7 genomes. We propose that sulfate/sulfite reduction by organisms from these newly identified lineages likely serves an important control on carbon cycling in the terrestrial and marine subsurface."
} | 7,827 |
34739314 | PMC8570599 | pmc | 6,984 | {
"abstract": "Positive interactions between bacteria, often described to be rare, occur commonly and primarily as parasitisms.",
"introduction": "INTRODUCTION Microbial communities are composed of multiple species that interact with one another in a variety of ways as part of the “struggle for existence” ( 1 ). Interactions between species can be negative, where a species inhibits another species’ growth through nutrient exploitation and chemical warfare ( 2 ), or positive, where a species promotes another species’ growth by increasing nutrient availability and creating new niches ( 3 ). The bidirectional interaction between two species is determined by the two one-way interactions between the species. For example, a mutualism occurs when two species positively affect each other. The overall distribution of positive and negative interactions within a microbial community profoundly affects the community’s structure, stability, and productivity ( 4 – 6 ). These properties, in turn, shape a community’s ability to perform vital functions for the environment ( 7 – 10 ) and for host organisms ( 11 – 14 ). Despite the importance of the distribution of microbial interactions, the relative prevalence of positive and negative interactions in nature remains largely unknown. Positive interactions are generally thought to be rare. Experimental evidence from coculture studies points to a dominance of negative interactions ( 2 , 15 ). For example, evidence of positive interactions was found in <10% of pairs of bacteria isolated from tree holes ( 15 ). However, these results are subject to strong experimental biases, such as the use of a single environment, although microbial interactions can differ markedly across environmental conditions ( 16 – 18 ), and the use of strains that each grow individually in the environmental conditions being tested. Metabolic modeling, which can simulate millions of interactions across myriad environments, as well as limited experimental evidence, suggests that positive interactions emerge via environment-dependent secretions and can facilitate otherwise nongrowing species ( 18 – 21 ). In addition, evolutionary theories such as the Black Queen hypothesis argue that such secretion-mediated positive interactions are selected for ( 22 ). Together, these findings suggest that positive interactions among microbes may be common and play an important role in shaping microbial communities, but these theories have not yet been thoroughly tested experimentally. Quantifying the prevalence of positive interactions and determining the conditions in which they occur could improve our ability to predict and control the ecology of microbial communities ( 23 , 24 ). Positive interactions are predicted to enhance a community’s diversity and productivity but decrease its stability ( 6 , 25 , 26 ). Therefore, a better understanding of these interactions would enhance our ability to manipulate and manage communities, with widespread applicability in environmental conservation ( 27 ), crop health ( 28 ), and human health ( 29 ). Nevertheless, the data required for quantifying the distribution of interactions across environments are still lacking because of methodological limitations that frustrate comprehensive sampling of interactions under many conditions ( 30 ). Inferring interactions from metagenomic sequencing remains an outstanding challenge ( 31 , 32 ), and directly measuring interactions at scale is difficult to perform using existing experimental paradigms. To gain a broad understanding of how species interact across a wide range of environments, we used a combinatorial screening platform called kChip ( 33 – 35 ) to measure >150,000 bidirectional bacterial cocultures among 20 different soil bacterial strains across 40 environments with differing carbon source identity or concentration. The kChip generates cocultures at an ultrahigh-throughput scale by rapidly and randomly combining droplets containing microbial cultures and/or medium components within microwells ( Fig. 1, A to D , and fig. S1) ( 33 ). Here, we paired unlabeled (wild-type) and green fluorescent protein (GFP)–labeled versions of 20 Gammaproteobacteria isolated from soil (6 Enterobacterales and 14 Pseudomonadales; table S1 and figs. S2 and S3). We selected these bacterial strains from our larger pool of fluorescently labeled soil isolates, maximizing for phylogenetic diversity (table S1 and Materials and Methods). We cocultured each strain pair in each of the 33 single-carbon sources [0.5% (w/v)], a mix of these, 5 of the 33 at reduced concentration [0.05% (w/v)], and a no-carbon control. Carbon sources were drawn from several biochemical classes including carbohydrates, amino acids, and carboxylates (table S2). We measured the effect of each unlabeled strain on the growth of each labeled strain in each carbon source, giving a total of 17,600 combinations [20 labeled strains, (20 + 2 control) unlabeled strains, and (39 + 1 control) carbon sources]. Each combination appeared >10 times on average, and only 3% appeared <3 times and were excluded from further analysis (fig. S4). Measured one-way interactions were used to classify each bidirectional interaction qualitatively ( Fig. 1E ) and quantitatively ( Fig. 1F ). Fig. 1. High-throughput interaction assay and analysis. ( A ) Steps to assay the effect of multiple unlabeled species on a single-label species across carbon sources on each kChip. Color-coded droplets, each containing either a labeled + unlabeled coculture or a single carbon source, were generated (step 1), pooled together (step 2), and loaded onto a kChip (step 3). Each kChip contained an array of microwells that randomly paired coculture droplets with carbon source droplets. After imaging the color codes to identify the inputs per microwell, droplet pairs were merged via exposure to an electric field (step 4), and the growth of the labeled strain was measured at 0, 24, and 72 hours (step 5). ( B ) Overall size of the kChip screen. ( C ) Using data across kChips, bidirectional interactions were deduced by combining data where each strain within a given pair was the labeled strain. ( D to F ) Framework for kChip data analysis. Each bidirectional interaction was described qualitatively (interaction classification) and quantitatively (interaction strength, m ; interaction type, ϴ). Our data provide direct experimental evidence that positive interactions are indeed common and occur primarily as parasitisms, where one species’ growth is improved at the expense of the other’s. More broadly, we found that interactions strongly depend on the environment via differences in the carbon consumption preferences of the interacting strains. Notably, we found that strongly growing partners consistently enabled the growth of strains that were unable to grow in monoculture (85%), suggesting a simple strategy for cultivation, microbiome engineering, and design of microbial consortia.",
"discussion": "DISCUSSION In this study, we performed comprehensive pairwise coculturing of 20 culturable soil microbes across 40 different carbon source environments. By measuring interactions across many environments, our study produced many instances in which at least one of the two cocultured strains grew poorly or not at all in monoculture, a regime in which we found that positive interactions were far more likely to occur than previously measured ( 2 , 15 ). The study also produced instances where both strains grew well and antagonism was common, a result consistent with previous large-scale studies of bacterial interactions ( 37 , 38 ). Our study unearthed the wealth of positive interactions that occur in our system. While mutualisms were relatively rare (5%), commensalisms (12%) and parasitisms (18%) were common and accounted for the majority of cases where total coculture yield was greater than the sum of monocultures (24%) (fig. S22), a criterion previously used to classify cooperative interactions ( 15 ). The prevalence of these positive interactions corroborates predictions from large-scale metabolic models ( 19 , 20 , 39 , 40 ). Our results are also consistent with the predictions of theories such as the Black Queen hypothesis, which asserts that interspecies cross-feeding of “leaky” public goods is evolutionarily selected for ( 22 , 41 ). Last, our results generalize smaller-scale demonstrations that cocultured strains ( 42 ) and spent media ( 43 ) can induce growth of fastidious bacteria. Together, positive interactions increasingly appear to play a dominant role in driving community properties, such as resistance to invasion and productivity ( 3 , 22 ), and in supporting microbial biodiversity ( 44 ). Positive interactions can also shape the phylogenetic composition of communities. In particular, common facilitation among phylogenetically distinct strains, as shown in our data, may promote phylogenetic diversity within communities, potentially leading to phylogenetic overdispersion ( 45 , 46 ). Interactions in our system varied significantly across environments and time points (fig. S20), suggesting that properties of natural communities can display considerable spatial and temporal variability. While interactions did not appear to significantly depend on properties intrinsic to the environment itself, they nonetheless strongly depended on the environment via the ability of each strain to individually grow in it: Negative interactions were frequent between strong growers, while positive interactions occurred commonly between strong and weak growers across all time points and environments. Given the widespread differences in growth that occur among bacteria, these results suggest that positive interactions may occur commonly in nature. A variety of mechanisms could explain the prevalence of positive interactions in our system. First, facilitated strains might have grown on components of accumulating dead cells, although this is unlikely given the time scale of the coculture experiment ( 47 ). Second, the facilitator might have secreted carbon source–degrading enzymes that increased overall carbon availability. This mechanism is consistent with the general prevalence of positive interactions in dimeric and trimeric sugars ( Fig. 3A ) but may not explain positive interactions in simple carbon sources such as monomeric sugars and TCA cycle intermediates ( Fig. 3, A and B ). Third, the facilitator may have excreted incompletely oxidized metabolites that were used by the facilitated strain ( 20 , 41 , 48 ). Such “overflow metabolism” would allow strains to indirectly benefit from the biochemical transformation capabilities of their facilitators. Exploitation of newly created niches could explain the positive interactions that we observed on simple carbon sources (e.g., the excretion of short-chain fatty acids as a by-product of incomplete monosaccharide oxidation). It may also explain the rarity of positive interactions on lower carbon source concentrations since respiration is known to be favored over fermentation under such conditions and overflow is less likely to occur ( 48 ). Despite the high throughput of our experiment, our system did not capture real-world bacterial diversity or environmental complexity. Our strain library was limited to two taxonomic orders isolated from topsoil. We only chose strains that grew on a minimal medium as part of our culturing protocol, possibly biasing our dataset in a variety of ways, e.g., against obligate facilitations for interactions involving amino acid or vitamin auxotrophies, which are known to be common ( 49 ). Strains were also pregrown on glucose as the sole carbon source before construction of coculture/carbon source combinations, imposing glucose consumption as a requisite for inclusion in our strain library and potentially affecting bacterial physiology (e.g., lag phase) and interactions. Moreover, while our carbon source library represented a variety of carbon source types, it was limited to soluble compounds, excluding many polymers on which metabolically driven positive interactions may be more common. Whether our results extend across additional phylogenetic groups (i.e., those occurring within soil and in other microbial ecosystems) and nutrient environments (i.e., across different and/or multiple carbon sources, concentrations thereof, and noncarbon nutrient requirements) should be investigated in follow-up studies to generalize trends observed in our system. Our results indicate that knowledge of how strains grow individually in an environment can be strongly predictive of how they interact in that environment. In contrast, knowing how the same strains interact in a different environment or how different strains interact in the same environment does not appear to be very informative. Last, our results suggest that a potential strategy for inducing the growth of a nongrowing or weakly growing strain, independent of growth medium, is to coculture it with a strongly growing strain. Here, we uncovered several general, statistical rules governing microbial community structure and function. These rules deepen our understanding of microbial community ecology and are crucial to enable the efficient design and control of beneficial microbial communities."
} | 3,319 |
28596334 | null | s2 | 6,985 | {
"abstract": "Many organisms and tissues display the ability to heal and regenerate as needed for normal physiology and as a result of pathogenesis. However, these repair activities can also be observed at the single-cell level. The physical and molecular mechanisms by which a cell can heal membrane ruptures and rebuild damaged or missing cellular structures remain poorly understood. This Review presents current understanding in wound healing and regeneration as two distinct aspects of cellular self-repair by examining a few model organisms that have displayed robust repair capacity, including "
} | 146 |
25992898 | PMC4439118 | pmc | 6,988 | {
"abstract": "Fast growth represents an effective strategy for microbial organisms to survive in competitive environments. To accomplish this task, cells must adapt their metabolism to changing nutrient conditions in a way that maximizes their growth rate. However, the regulation of the growth related metabolic pathways can be fundamentally different among microbes. We therefore asked whether growth control by perception of the cell’s intracellular metabolic state can give rise to higher growth than by direct perception of extracellular nutrient availability. To answer this question, we created a simplified dynamical computer model of a cellular metabolic network whose regulation was inferred by an optimization approach. We used this model for a competing species experiment, where a species with extracellular nutrient perception competes against one with intracellular nutrient perception by evaluating their respective average growth rate. We found that the intracellular perception is advantageous under situations where the up and down regulation of pathways cannot follow the fast changing nutrient availability in the environment. In this case, optimal regulation ignores any other nutrients except the most preferential ones, in agreement with the phenomenon of catabolite repression in prokaryotes. The corresponding metabolic pathways remain activated, despite environmental fluctuations. Therefore, the cell can take up preferential nutrients as soon as they are available without any prior regulation. As a result species that rely on intracellular perception gain a relevant fitness advantage in fluctuating nutrient environments, which enables survival by outgrowing competitors.",
"introduction": "Introduction One of the most essential aspects of living cells is growth and its associated control to fit the organisms’ needs. In human, selection for fast and selfish growth can result in cancer, while it represents a very effective evolutionary strategy for microorganisms to survive in a competitive environment. The reproductive success of microbial organism depends on the fast and precise adjustment of their growth rate to the actual environmental condition [ 1 ]. The reason is that most microbes live in a highly competitive environment where fast and effective transfer of available nutrients into biomass can give a significant fitness advantage [ 2 ]. Selection for fast growth leads to phenomena such as overflow metabolism [ 3 – 5 ], where fast but wasteful conversion of glucose into biomass can be of advantage in comparison to the effective use of nutrients. The overflow metabolism of E.coli is also known as Crabtree effect in S. cerevisiae and as Warburg effect in cancer cells [ 6 ]. Another regulatory phenomena that is associated with fast growth and is commonly used among many bacteria and other microbes is carbon catabolite repression (CCR) [ 7 – 9 ]. To grow fast microbes selectively utilize preferred carbon sources in a hierarchical manner. In the presence of a preferred sugar such as glucose, CCR causes metabolic enzymes of alternative carbon sources to be expressed at low rate and can additionally reduce their activity. There is strong evidence that growth dependent phenomena such as overflow metabolism or CCR are the consequence of a metabolic regulation or growth control in response to extracellular nutrient availability. Further, it seems possible that the perception of extracellular nutrient availability plays an important role in growth control [ 10 ], as it is the primary information cellular response is based on. We define two distinct types of perception, termed intracellular and extracellular perception. In the case of extracellular perception the cell regulates its metabolism exclusively in response to extracellular nutrient information, while in the case of intracellular perception microbes indirectly recognize nutrient availability by perceiving the intracellular metabolic state. The intracellular perception is motivated by experimental observations [ 11 – 13 ] of microbes, e.g. E.coli , which do not possess any extracellular carbohydrate receptors, like the Rgt2 and Snf3 glucose sensors of yeast [ 14 , 15 ]. These microbes should be capable of perceiving extracellular nutrient availability indirectly from intracellular metabolic states. Intuitively, the extracellular perception should lead to a more precise and fast adaptation to nutrient availability, since changes in the environment can be perceived faster and to higher accuracy. Here, the question arises whether exclusive intracellular perception can result in a growth benefit in presence of fast fluctuating nutrient concentrations. Following this question, we are interested in which frequency regimes the exclusive perception of intracellular nutrient concentration is evolutionary more beneficial than the exclusive perception of extracellular nutrient concentrations. Furthermore, what are the regulatory principles causing this benefit in average growth rate or fitness and can the regulatory phenomenon of carbon catabolite repression be understood by means of nutrient perception? To give an answer to these questions and an explanation how the integration of the perception strategies for growth control contribute to shape growth rate in microorganisms, we will introduce a simplified replicator model for microbial growth. The replicator model consists of a minimal metabolic network, ribosomes, and a controller that can detect intracellular and extracellular metabolite concentrations. Optimal growth control is realized by minimizing the difference between the actual intracellular concentrations of metabolites and precursors and their desired concentrations, which is determined by the perceived nutrient availability. Using this simplified model we are able to show that growth control by perception of extracellular nutrient concentrations is of selective advantage if environmental conditions change slowly over time. If environmental conditions change fast in comparison to the minimum generation time, gene regulation and protein turnover will lag behind and the model predicts that in this case sensing the intracellular precursor state is of advantage.",
"discussion": "Discussion This study indicates that indirect intracellular perception of extracellular nutrient availability can give rise to a growth benefit under situations where the up and down regulation of pathways cannot follow the fast changes of the nutrient environment. Although intracellular perception carries less information about the actual environmental conditions, this regulatory mechanism enables exponentially growing organisms to gain maximal average growth if nutrient concentrations fluctuate on timescales comparable to the minimum generation time. In our simulation, a system with intracellular perception responds to strong fluctuations by keeping preferential nutrient pathways activated and non-preferential pathways inactivated. As a result the cell can take up preferential nutrients as soon as they are available without any prior regulation. This regulatory strategy is a good example for minimal adjustment . According to Schuetz et al. [ 1 ] there is a trade-off between optimality under one given condition and minimal adjustment between different conditions, i.e. Pareto optimality [ 32 ]. In other words, cells will tune metabolic pathways to obtain optimal growth if surrounded by a constant environment. Contrarily, in a fluctuating environment, cells will regulate their pathways to respond to environmental changes by minimal adjustment of pathways. In this sense, intracellular perception gives rise to a regulation of minimal adjustment , which is dominant under fast environmental changes. Additionally, our results show that the notion of optimality is also given under fluctuating conditions, since minimal adjustment is a consequence of maximizing an objective function averaged over the range of conditions. Moreover, our model of intracellular perception is in agreement with the phenomenon of carbon catabolite repression [ 7 , 9 ], if cells are not able to distinguish between different conditions anymore, i.e. the fluctuation frequency approaches infinity. This situation is equivalent to a mixed constant environment. While carbon catabolite repression reflects the cell’s affinity to preferential sugars in a stable mixed nutritional surrounding, our results indicate that this mechanism holds under fast fluctuations (around the minimum generation time) as well. To our knowledge, CCR has not been obtained from an mathematical optimization process, before. Furthermore, our simulation of the growth dynamics produced a break-even point, where the average growth rate of the IPS and EPS are equal ( Fig 6 ). At this point the growth benefit of the IPS in the preferential environment matches the growth loss in the non-preferential environment. Growth benefit and loss arise from the exclusive adaptation to the PS environment ( Fig 7A ). This is in agreement with the experimental work of Mitchell et al. [ 33 ], who have observed anticipation of environmental changes in the sugar metabolism of E.coli and S.cerevisiae . Mitchell et al. classified the regulatory response to environmental changes into direct and anticipatory regulation, whereas the former regulates its metabolism in direct response and the latter in advanced preparation. Further, they state that an anticipatory response will be evolutionary beneficial if “the benefit gained from anticipation exceeds the cost of early preparation”. We can identify the anticipatory regulation with the IPS and the direct regulation with the EPS. As we have shown intracellular perception yields a preparation for the PS environment during the NPS environment, which can be regarded as an anticipatory behavior. Especially, the hyper-up-regulation of the PS uptake transporter in the presence of NPS environment, which results in the resonance peak of the average growth rate, serves as a good example for anticipatory regulation. This course of action is only beneficial for fluctuating environments with frequency smaller than the break-even frequency. Thus, anticipatory behavior in a highly predictable fluctuating environment can be understood by limited and delayed intracellular perception. Using our phenomenological computer model, we further showed that extracellular perception is of selective advantage under slow environmental fluctuations. However, it is reasonable to assume that intracellular perception always contributes to some extent to growth control. This hypothesis is supported by the observations of New et al. [ 34 ], who have shown that wild S. cerevisiae strains divide into sub-populations of specialist and generalists according to their growth rate related response time (lag phase). Generalist will adapt faster to a new carbon environment than specialists if the environment changes from a preferential to a non-preferential carbon source. Our results in Fig 7A for the non-preferential regime exhibit the same relation between growth regulation by means of extracellular perception (EPS) and intracellular perception (IPS). The EPS, like the generalists, adapts faster to the non-preferential environment than the IPS. In this context generalist could be seen as microbes whose growth control mainly depends on extracellular perception, while the contribution of intracellular perception has an bigger impact on the specialist’s growth control. Although, both perception types can be utilized by microorganisms, their contribution to growth control can be differently depending on the individual evolutionary background. Regarding the IPS, an interesting result of our simulation is the existence of a resonance peak for fluctuations around the minimum generation time. At this peak, the time delay in nutrient perception equals the switching time between environments resulting in optimal fitness. The data-based mathematical model of Mitchell and Pilpel [ 35 ] supports our finding as their cellular system shows a fitness peak around 1–2.5 generation times. To summarize, our work indicates that intracellular perception is of selective advantage and gives rise to CCR in oscillating environments, so that microbes specialize on the preferential nutrient and anticipate it in its absence. In general, intracellular perception could be a fundamental regulatory principle of minimal adjustment to changing conditions. Although our study is limited to a purely qualitative conclusion, due to the simplicity of our approach, the presented model is sufficient to gain insight in the fundamental differences of microbial growth control. In following projects, it would be worthwhile to test our simulation with real metabolic networks, like from the model organisms E.coli or S.cerevisiae . Moreover, experimental evidence, i.e. competing species experiments, is needed to confirm our theory of the dominance of intracellular perception under fast fluctuations."
} | 3,251 |
36133119 | PMC9416852 | pmc | 6,989 | {
"abstract": "Being one of the most common forms of energy existing in the ambient environment, acoustic waves have a great potential to be an energy source. However, the effective energy conversion of an acoustic wave is a great challenge due to its low energy density and broad bandwidth. In this work, we developed a new piezoelectric nanogenerator (PENG), which is mainly composed of a piece of piezoelectric nanofiber/polymer composite membrane. As an energy harvester, the PENG can effectively scavenge a broad low-frequency (from 50 Hz to 400 Hz) acoustic energy from the ambient environment, and it can even scavenge a very weak acoustic energy with a minimum pressure of only 0.18 Pa. When a drum was used as an excitation source, the maximum open-circuit voltage and short-circuit current density of the PENG reached 1.8 V and 1.67 mA m −2 , respectively. In addition, the PENG had a good stability and its output frequency and amplitude were closely related to the driving sound wave, which made the PENG capable of detecting acoustic signals in the living environment and have the potential to be applied as a self-powered active acoustic detector.",
"conclusion": "Conclusions In summary, we develop a piezoelectric nanofiber/polymer composite membrane-based PENG for harvesting sound wave energy from ambient environments. The lowest sound intensity for driving the PENG is 79 dB which corresponds to a sound pressure of only 0.18 Pa. The generator has a wideband characteristic, which can respond to sound waves ranging from 50 Hz to 400 Hz. Meanwhile, by adjusting the resonant frequency of the device via tension control, the PENG adapts well to different environments. Owing to its high sensitivity and broad bandwidth characters, the newly designed PENG can convert the output current signal back to the sound signal, which implies its potential to be an active acoustic detector for sound distinguishing and recording.",
"introduction": "Introduction Acoustic waves are one of the most common energy forms existing in the environment. If acoustic energy could be effectively harvested, it has the potential to power millions of widely-distributed sensor network in internet of things. However, its low energy density and broad bandwidth make acoustic energy difficult to harvest. There are three main mechanisms that have been utilized to fabricate sound wave energy harvesters: electromagnetic induction, 1 piezoelectric effect, 2–6 and electrostatic effect. 7–11 An electromagnetic acoustic wave generator has a high sensitivity and output performance, but the complicated and expensive manufacturing technique make it not suitable for large-scale and dispersive applications. The other two kinds of acoustic energy harvesters are the piezoelectric nanogenerator (PENG) and triboelectric generator (TENG). Energy conversion in a TENG is achieved by coupling the triboelectric effect and the electrostatic effect. 12 Though a relative higher output has been reported, the performance of the TENG was largely determined by the charge density, which must be high enough for a normal output through long accumulation time and is easily dissipated by moisture, dust, and intermittently driven force. 13–15 Therefore, it is difficult for TENGs to achieve their best performance in a real working environment. Compared with TENG, the working mechanism for the PENG is based on the piezoelectric property of piezoelectric materials. When the PENG is deformed by external forces, a piezoelectric potential is created, 16 which is not susceptible to the environment. In addition, our previous work demonstrated the PENG's ability to convert mechanical energy into electric energy and its sensitivity to tiny forces. 17 Consequently, its performance is barely affected by the environment, which makes PENGs a promising candidate for outdoor applications to harvest sound energy. To realize the harvesting of acoustic energy, the PENG must meet the following requirements. First, it should be very sensitive to tiny forces, which means that its response to sound pressure should be fast and accurate. Second, it should be able to harvest the noises of broadband frequency around our living environment. Third, its working frequency should be regulated to adapt to different situations. Among previous research, the piezoelectric effect was mostly focused on acoustic sensing rather than energy harvesting and utilization, where the accurate voltage response of the acoustic signals represented by the intensity and frequency is of a greater concern. 18–20 However, when paying attention to the effective harvesting of acoustic energy, a higher output energy and conversion efficiency are highly desired, especially the output current. 21 Although the sound pressure created by acoustic wave is generally weak, greater deformation along the tension direction of a flexible membrane is possible to increase the output current. On the other hand, lead zirconate titanate (PZT) is widely used in the production of acoustic harvesting devices due to its compatibility with nano/micro-fabrication techniques and its easy development using a sol–gel process. 5,22 Unfortunately, PZT has also aroused intensive concerns for human health and the environment. 23,24 Therefore, it is important to explore lead-free piezoelectric membrane. 0.5Ba(Zr 0.2 Ti 0.8 )O 3 –0.5(Ba 0.7 Ca 0.3 )TiO 3 (BZT–BCT) nanofiber is a lead-free piezoelectric ceramic with a high piezoelectric coefficient (620 pC N −1 ), 25–28 which has been reported as a biocompatible PENG. 29 However, they have not been explored as an acoustic energy harvester until now. In this paper, we demonstrate a membrane PENG device, in which piezoelectric BZT–BCT nanofibers serve as the energy conversion material. Polydimethylsiloxane (PDMS) and BZT–BCT nanofibers were assembled into a piece of membrane. This composite nanofiber membrane showed an ability to harvest energy from ambient noise. The membrane PENG responded to acoustic waves ranging from 50 Hz to 400 Hz as well as produced an open-circuit voltage of 1.8 V and a peak output current density of 1.67 mA m −2 . The minimum pressure response was only 0.18 Pa. Furthermore, the restoration of the driving sound signals via the output current implied that the device has the potential to be used as an active acoustic detector.",
"discussion": "Results and discussion BZT–BCT nanofiber piezoelectric membrane In our work, the PENG was made into a monolayer membrane structure as shown in Fig. 1a . Herein, the BZT–BCT nanofibers were used to fabricate the monolayer membrane structure. Fig. 1b shows an optical photograph of a piece of the BZT–BCT vibrating membrane with a side length of 2 cm and thickness of only 5 μm. The spiral silver electrode was deposited on the BZT–BCT membrane through magnetron sputtering using a shadow mask ( Fig. 1c ). The detailed fabrication process is depicted in the Experimental section. The nature frequency of the square membrane with its four sides fixed can be expressed as follows: 1 where T is the uniform surface tension per unit length, M is the surface mass density, m and n are arbitrary integers, and L is the side length of the membrane. The compound parameter, , refers to the wave propagation velocity in the membrane. Eqn (1) shows that the membrane could be induced resonance at many frequencies with different values of m and n . In addition, the nature frequency could be easily adjusted by changing the parameters of the membrane, such as the surface tension T , the surface mass density M or the side length L . This adjustment enabled our devices to work in a wider frequency range and improved its adaptability for harvesting noise energy from different environments. The working mechanism of the PENG is shown in Fig. 1d and e . When the acoustic source was set near the PENG and turned on, the BZT–BCT membrane vibrated along the vertical direction of the membrane surface. Consequently, the strain in the membrane changed periodically, radially, symmetrically, and isotropically in all directions. Therefore, the BZT–BCT nanofibers between the two electrodes were stretched and recovered alternately with the changing strain. During this process, the electrons in the external circuit flowed continuously between the two electrodes to respond to the change in the piezoelectric potential. The detailed working mechanism is depicted in (Fig. S1 † ). Fig. 1f shows the microstructure of the sintered BZT–BCT textile, where the BZT–BCT nanofibers were distributed randomly with a diameter of about 200 nm. The BZT–BCT nanofibers were prepared via an electrospinning method as discussed in a previous report. 30 The detailed preparation process for the BZT–BCT nanofiber textile is depicted in the Experimental section. The BZT–BCT textile was fabricated into a thin vibrating membrane with almost no loss of its thin and light features by filling with PDMS which provided the textile with more support to prevent the tearing of the nanofibers. The detailed preparation process for the BZT–BCT vibrating membrane is depicted in the Experimental section. Fig. 1g shows a Scanning Electron Microscopy (SEM) image of the BZT–BCT vibrating membrane, which demonstrates that the PDMS filled in the gap of the BZT–BCT textile and formed a reinforced concrete structure together with BZT–BCT nanofibers. This structure not only enhanced the tensile strength of the membrane but also kept the thin and light features of the BZT–BCT textile. Fig. 1 Morphology and structure of the PENG. (a) Schematic illustration of the PENG. (b and c) Optical photographs of the BZT–BCT membrane and BZT–BCT membrane with electrodes in the PENG. (d and e) Working mechanism of the PENG. (f and g) SEM images of the BZT–BCT nanofibers and a BZT–BCT/PDMS nanofiber composite membrane. Performance of the membrane PENG for noise energy harvesting and active acoustic wave detection \n Fig. 2 shows the output current and voltage of the PENG when a drum was used as an excitation source. The PENG was fixed on an iron support that was about 2 cm above the drum head. When we beat the drum head with a drumstick, the vibration of the drum spread through the air to the PENG. Fig. 2a and b show five current/voltage wave envelopes corresponding to five beats on the drum head. The maximum output current and voltage of this PENG reached 0.67 μA and 1.8 V, respectively (the current signal and voltage signal were measured through a Stanford Research System SR560 low-noise current preamplifier and SR570 low-noise voltage preamplifier. Their input resistances are 100 Ω and 100 MΩ, respectively.). The whole driven process was completely manual. The knocking force cannot be controlled the same. Therefore, for all of the beats, there was a slight deviation between the peak values for the output current/voltage. Fig. 2c and d show the details for the output current/voltage wave envelope. These damped waves reflect the damped vibration on the drum head. It was found that the frequency of the output electric signal was about 300 Hz, which was in accordance with the natural frequency of the drum. Fig. 2 (a and b) The output voltage/current of a PENG excited by drumbeats. Each wave envelope corresponds to one beat of a drumstick. The device was placed 2 cm above the drum. The maximum output voltage and current of the PENG reached 1.8 V and 0.67 μA, respectively. (c and d) The details for an output voltage/current wave envelope, respectively. In order to further investigate the performance of the PENG, including the frequency and intensity response characteristics, a mini speaker was applied as the excitation source, which sent out sounds with different frequencies controlled by a LabVIEW program. This measurement system generated a sine acoustic wave with a frequency ranging from 50 Hz to 400 Hz. The sound intensity was monitored by a noise meter. In the testing process, the PENG was fixed above the speaker and kept a certain distance from the speaker. First, the generator was driven under the sound of 126 Hz and 104 dB for one second. Fig. 3a and b show that the peak value of the output current and voltage reached 132 nA and 1.0 V, respectively. The PENG device also exhibited a very fast response speed, which reached a stable output within only 0.08 seconds. For comparison, a TENG often takes several minutes to achieve a relatively stable output. 15 When the driving sound stopped, the vibration membrane gradually stabilized after 0.22 seconds of damping vibration. As shown in the enlarged view, the frequency of the voltage/current curves was 125.6 Hz (the acoustic wave was 126 Hz). The voltage curve was similar to the sinusoidal form. However, the current curve was seriously deformed, which was chiefly due to the smaller resistance of the current preamplifier leading to a faster current change. Fig. 3 (a and b) The output voltage/current of a PENG driven by a sound of 126 Hz and 104 dB. The details for the output voltage/current wave are marked by the red line. (c) The output voltage/current of a PENG driven by an acoustic source with different frequencies (the loudness was kept at 92 dB). (d) The output voltage of two different devices with different membrane tension. (e) The output voltage of the PENG in different environments (the frequency was kept at 126 Hz). (f) The relationship between the peak voltage and sound intensity in semi-logarithmic coordinates. The generator was driven at a series of frequencies ranging from 50 Hz to 400 Hz while keeping the sound intensity at 92 dB. The corresponding peak values for the voltage and current are shown in Fig. 3c . In this test, the output signal changed with the frequency of the driving acoustic wave and there was an explicit relationship between the output signal and the driving frequency, which was used as a criterion to detect the acoustic wave. In this frequency range, the maximum output voltage and current were 0.45 V and 44 nA at 115 Hz, respectively. The minimum output voltage and current were 9 mV and 3 nA at 400 Hz, respectively. Therefore, the resonant frequency of this device was 115 Hz and the PENG showed a certain selectivity to the frequency of the acoustic wave. In order to improve the detection performance of the PENG, its resonant frequency was adjusted to a value close to the working frequency in certain circumstances. As shown in eqn (1) , if we increased the tension degree T , the resonant frequency of the PENG moved to higher values. For example, when the BZT–BCT membrane's tension was increased in the device, the resonant frequency increased accordingly to 200 Hz as shown in Fig. 3d . Fig. 3e shows the peak value of the voltage signals of the PENG driven by 126 Hz acoustic waves with the sound intensity increasing from 79 dB to 104 dB. The output signals increased monotonically. In fact, a simple expression between the voltage peak and sound intensity (SIL) was obtained (the derivations for eqn (2) and the energy conversion efficiency are shown in Derivation S1 and Derivation S2, respectively, in the ESI † ): 2 log 10 V = 0.05 × SIL + 0.5 × (log 10 k + log 10 RI 0 ) where V is the peak value of voltage, SIL is the sound intensity, SIL = 10 × log 10 I / I 0 , I is the energy flux density of the sound, I 0 is a basic parameter that equals to 10 −12 W m −2 , R is the load resistance, which equals to 100 MΩ, and k is a scale factor that is related to the energy conversion efficiency. Eqn (2) shows that the logarithm of the voltage is linearly proportional to the SIL. After taking the logarithm of the voltage peak in Fig. 3e , the result is shown in Fig. 3f . These data points show a strict linear relationship with the slope of the fitting being 0.049 and the intercept of the fitting being −4.76. By the fitting intercept, the scale factor k was calculated as 0.05. Furthermore, the energy conversion efficiency μ was also calculated as 0.86% for our device. It is worth mentioning that the sound pressure dropped exponentially from 3.17 Pa to 0.18 Pa when the sound intensity decreased from 104 dB to 79 dB. The low sound pressure (0.18 Pa) caused the PENG's response at the sound intensity of 79 dB. The output current and voltage were 10 nA and 70 mV, respectively. Considering the sensitivity to low sound pressure, the response bandwidth feature, and the adjustability of a membrane PENG, it should be suitable to harvest ambient noise energy. From Fig. S2, † the PENG successfully harvested noise at different environments such as a noisy workshop, a helicopter taking off, an alarm, and a man's voice, and converted them into electricity. Fig. 4a and b show the current responses and the original waveform of a PENG to a Chinese popular song, “ Su Xiu ”, respectively. Their corresponding frequency spectra are shown in Fig. 4c and d , respectively. It can be seen that the current response of the PENG exhibited some distortion in the waveform compared with the original waveform, which was mainly caused by its response defects to the high-frequency signals. This suggested that an improved design was needed. However, after converting the current signals in Fig. 4a back to sound signals (Video S1 † ) using a MATLAB program, the melody and lyrics were still easily distinguishable and similar to the original song. These results implied the potential value of the PENG as an active acoustic wave detector. Fig. 4 (a) Current responses of a PENG to the sound of a Chinese popular song, “ Su Xiu ”. (b) The original waveform of this popular song. (c and d) The corresponding frequency spectra for (a) and (b), respectively. Moreover, benefiting from the reinforced composite structure of the BZT–BCT membrane, the PENG had a great mechanical robustness and long-time working ability. Fig. 5 shows the stability of the PENG driven by a 120 Hz and 95 dB acoustic wave. The PENG almost exhibited no decay in output even after working continuously for over 2 hours. Fig. 5 The electrical output of a PENG working for more than 2 hours at a frequency of 120 Hz. (Since the amount of data was too large at a sampling rate of 5000, the data were only acquired for 3 minutes every 20 minutes)."
} | 4,558 |
35775078 | PMC9237525 | pmc | 6,991 | {
"abstract": "The spidroin N-terminal domain (NT) is responsible for high solubility and pH-dependent assembly of spider silk proteins during storage and fiber formation, respectively. It forms a monomeric five-helix bundle at neutral pH and dimerizes at lowered pH, thereby firmly interconnecting the spidroins. Mechanistic studies with the NTs from major ampullate, minor ampullate, and flagelliform spidroins (MaSp, MiSp, and FlSp) have shown that the pH dependency is conserved between different silk types, although the residues that mediate this process can differ. Here we study the tubuliform spidroin (TuSp) NT from Argiope argentata , which lacks several well conserved residues involved in the dimerization of other NTs. We solve its structure at low pH revealing an antiparallel dimer of two five-α-helix bundles, which contrasts with a previously determined Nephila antipodiana TuSp NT monomer structure. Further, we study a set of mutants and find that the residues participating in the protonation events during dimerization are different from MaSp and MiSp NT. Charge reversal of one of these residues (R117 in TuSp) results in significantly altered electrostatic interactions between monomer subunits. Altogether, the structure and mutant studies suggest that TuSp NT monomers assemble by elimination of intramolecular repulsive charge interactions, which could lead to slight tilting of α-helices.",
"introduction": "Introduction Orb weaving spiders stand out among silk producers found in nature as they can produce up to seven different types of silk, each tailored for a specific function. The main components of the spider silk threads are large proteins called spidroins. Major ampullate spidroin (MaSp) is the main component of dragline silk, the strongest and toughest of all spider silk types ( Ayoub et al., 2007 ). Spider silk formed from minor ampullate spidroin (MiSp) is used for auxiliary spiral stabilization ( Gosline et al., 1986 ; Colgin and Lewis, 1998 ). Aciniform spidroin (AcSp) is the main component of wrapping silk used to immobilize prey ( Tremblay et al., 2015 ). Pyriform spidroin (PySp) makes up the attachment discs, which lash the joints of the web and attaches dragline silk to surfaces ( Perry et al., 2010 ). Flagelliform spidroin (FlSp) forms capture spiral silk, which can extend up to 500% of its length ( Eisoldt et al., 2011 ). Aggregate spidroin (AgSp), coats the FlSp threads and is used as a glue to capture prey on the web ( Opell and Hendricks, 2010 ; Collin et al., 2016 ). Tubuliform spidroin (TuSp) forms the outer layer of spider egg cases and protects the eggs from the external environment ( Tian and Lewis, 2005 ; Eisoldt et al., 2011 ). Each spidroin is produced in a specific gland. The major ampullate gland, which has been studied most extensively, has a long, winding, and narrow tail, a wider ampulla or sac, and an S-shaped narrowing duct connected to the sac via a funnel ( Andersson et al., 2013 ). After secretion in the tail segment, spidroins are held in the ampulla at up to 50% (w/w) concentrations ( Hijirida et al., 1996 ) and upon silk spinning, pulled through the tapered duct. Here, changes in the surrounding environment conditions (i.e., reduced pH, altered ion composition, increased partial CO 2 pressure) and increasing shear forces cause them to assemble and form solid threads ( Knight and Vollrath, 2001 ; Rousseau et al., 2004 ; Andersson et al., 2014 ; Sparkes and Holland, 2017 , 2019 ). The tubuliform gland, on the other hand, is long, noodle-shaped, and smooth without a distinguishable ampulla-shaped storage sac ( Chaw and Hayashi, 2018 ). Also, the conditions in the tubuliform gland are not known. The conditionally high solubility and regulation of spider silk formation is mediated by two conserved spidroin terminal domains–the N-terminal domain (NT) and the C-terminal domain (CT) ( Askarieh et al., 2010 ; Hagn et al., 2010 ), whereas the highly variable central repetitive domain (Rep), is responsible for the silk properties ( Gosline et al., 1999 ). MaSp NTs from different spider species were early on found to be pH sensitive ( Gaines et al., 2010 ; Landreh et al., 2010 ). In the ampulla at pH > 7 the NT forms a monomeric five-helix bundle showing a distinct dipolar distribution of charged amino acids ( Askarieh et al., 2010 ; Jaudzems et al., 2012 ). Under these conditions, it likely promotes solubility of the spidroin Rep domain by forming the shell of micelle-like particles with the aggregation-prone regions sequestered in their core ( Kronqvist et al., 2017 ). As spidroins are pulled through the spider silk duct, the pH is reduced to <5.7 ( Andersson et al., 2014 ), which causes sequential protonation of a cluster of glutamic acids on the NT’s surface. This disrupts several charge interactions leading to movement of α-helices, relocation of a wedged W10 side chain from a buried to a surface exposed conformation, and subsequent dimerization of the domain. The NT dimerization results in firmly interconnected spidroins (CT is dimeric already during storage), which ensure propagation of pulling forces during the silk fiber formation ( Kronqvist et al., 2014 ; Ries et al., 2014 ). The pH dependent dimerization mechanism seems conserved among NTs from different silk types and species ( Heiby et al., 2017 ); however, the exact structural details are divergent ( Otikovs et al., 2015 ). Through studies of site-directed mutants bearing glutamate to glutamine substitutions, the residues E79, E84 and E119 were identified to be protonated in the Euprosthenops australis MaSp NT dimer ( Kronqvist et al., 2014 ). In Araneus ventricosus MiSp NT, E84 is substituted by a serine and the nearby E73 was instead found to be protonated ( Otikovs et al., 2015 ). Similar mutants were investigated for Nephila clavipes FlSp NT and Latrodectus hesperus MaSp NT, however, the exact carboxylates to be protonated could not be identified ( Bauer et al., 2016 ; Sarr et al., 2022 ). AcSp from Nephila antipodiana displays a very different charge distribution on the protein surface, nevertheless it still forms dimers at low pH and in presence of physiological salt concentrations ( Chakraborty et al., 2020 ). Three other MaSp NT mutants have been investigated to characterize its monomeric conformation. Residues D40 and K65 located at opposite tips of the NT molecular dipole form an intermolecular salt bridge in the dimer and were proposed to mediate initial monomer association ( Schwarze et al., 2013 ; Kronqvist et al., 2014 ). The residue charge reversal in the mutant NT D40KK65D resulted in disturbed dipolar interactions between NT monomers and abolished its pH sensitivity ( Kronqvist et al., 2017 ). This variant shows excellent solubility-enhancing properties and has been applied as solubility tag for recombinant production of aggregation-prone proteins and peptides ( Kronqvist et al., 2017 , 2022 ; Sarr et al., 2018 ; Abelein et al., 2020 ). Another pH independent NT monomer was prepared by replacing an alanine residue in the middle of the dimer subunit interface with an arginine, thereby preventing self-association through charge repulsion ( Jaudzems et al., 2012 ). In the third mutant, all six methionine residues in the protein core were replaced by leucines, which significantly reduced protein plasticity and abolished the movement of α-helices necessary for NT dimerization ( Heiby et al., 2019 ). The structures of all TuSp domains from Nephila antipodiana were the first published silk protein structures ( Lin et al., 2009 ). The TuSp NT structure was determined by solution NMR in presence of 100 mM dodecylphosphocholine (DPC) to avoid protein aggregation and using a protein construct, which lacked the first 36 amino acids. The structure shows a four-helix bundle, which differs from the more recent structures of other NTs comprising five helices ( Figure 1 ). The amino acid sequence of TuSp NT also displays some notable differences in comparison to the consensus sequence of other NTs. The E119 of MaSp NT involved in the protonation during dimerization is replaced by an arginine. Furthermore, the sequence contains no methionines, which were found to enable the helical reorganization during MaSp NT dimerization. Additionally, in TuSp NT (similarly to FlSp NT), W10 is replaced by a phenylalanine. FIGURE 1 \n (A) \n Argiope argentata TuSp domain architecture ( Chaw et al., 2018 ). The number of residues for NT is different from the construct used in this study because it includes ∼20 non-conserved residues linking it to the Rep region. (B) Amino acid sequence alignment of spidroin NTs with known tertiary structures. Included are NTs from Argiope argentata TuSp (PDB ID 6TV5), Nephila antipodiana TuSp (PDB ID 2K3Q), Nephila clavipes FlSp (PDB ID 7A0O), Nephila antipodiana AcSp (PDB ID 7BUT), Euprosthenops australis MaSp (PDB ID 2LTH) and Araneus ventricosus MiSp (PDB ID 2MX9). Locations of α-helices are indicated by rectangles and amino acids excluded from the expression vector are in small caps. Residues protonated upon dimer formation (E79, E84, E119 in MaSp and in MiSp E73, E79 and E119) and W10/F10 are marked with an asterisk below the sequence. Arginines are blue, lysines are cyan, glutamic acids are red, aspartic acids are orange, methionines are green and cysteines are yellow. ClustalW alignment score to E. australis MaSp NT is shown at the end of each sequence. Considering the additional mechanistic insights gained from the recent studies of MaSp, MiSp and FlSp NT and the contrasting structural data of TuSp NT, we decided to re-investigate its structure and possible dimerization mechanism using a protein construct that comprises the complete conserved domain sequence from Argiope argentata . We use nuclear magnetic resonance (NMR) to characterize the TuSp NT conformation at pH > 7 and to determine its structure at pH 5.5. Next, several methods including NMR, size exclusion chromatography (SEC), and circular dichroism (CD) are used on the wild type (wt) protein as well as site-directed mutants to investigate the dimerization mechanism of TuSp NT.",
"discussion": "Discussion We performed structural studies and dimerization analysis of TuSp NT from A. argentata using NMR, SEC, and CD experiments, which showed that the pH-dependent stable dimer formation is conserved in tubuliform silk. However, TuSp NT could not be stabilized in the monomer conformation for NMR structure determination regardless of the environmental conditions. Furthermore, introduction of mutations in TuSp NT K40DD63R and TuSp NT A70E , corresponding to monomeric and pH insensitive MaSp and FlSp NT variants ( Jaudzems et al., 2012 ; Kronqvist et al., 2017 ; Sarr et al., 2022 ), failed to fully stabilize TuSp NT in the monomeric conformation. In order to identify residues that could potentially participate in the pH dependent dimerization mechanism, we determined the solution structure of TuSp NT dimer at pH 5.5. Guided by the insights from the structure, we designed the mutants TuSp NT E37QE82Q , TuSp NT E37QE82QE85Q , and TuSp E37QE82QE85QD130N , in which several clustered and surface-exposed acidic residues were exchanged for their amide counterparts. However, none of them behaved as a constitutive dimer and retained pH sensitivity. SEC and HSQC NMR data showed that the variants assume an intermediate conformation (i.e., with at least one of the titrating carboxylates neutralized) at high pH and are converted into dimeric conformation at pH 5.5. CD experiments showed that thermal stability of the mutants is higher at pH 8, however, not approaching the stability of TuSp NT at pH 5.5. The spidroin NT domains from most spider silk types and species have been shown to adopt a monomeric or dimeric five-helix bundle structure depending on environmental conditions. However, the previously determined TuSp NT structure from N. antipodiana shows an atypical four-helix arrangement that is distinct from the five-helix NT structures, raising questions about the fold conservation. Despite high sequence similarity (66% identity), our herein determined TuSp NT structure from A. argentata reveals a five-helix composition, which is highly similar to the structures of MaSp, MiSp, AcSp, and FlSp NTs. Superposition of the two TuSp NT structures shows a different organization of the first three helices ( Figure 6 ). The helix α1 in A. argentata TuSp NT is replaced by α3 in the N. antipodiana structure, α3 is substituted by α2, while helix α2 of the A. argentata protein is missing in the N. antipodiana structure. Knowing that the first 36 residues of the N. antipodiana protein were truncated for the structure determination, the differences are apparently due to a helical reorganization aimed to preserve its hydrophobic core, which is mainly formed by the helices α1 and α3-α5. Besides, the solution conditions used for each structure determination are vastly different—the N. antipodiana TuSp NT was studied at neutral pH (50 mM Tris-HCl, pH 7) in presence of 100 mM DPC, whereas we employed acidic conditions (20 mM sodium acetate, 20 mM NaCl, pH 5.5). Accordingly, the N. antipodiana structure at pH 7 shows a monomer, whereas the A. argentata TuSp NT at pH 5.5 is a dimer. The use of a shorter construct and presence of 100 mM DPC, which according to the authors was necessary to avoid aggregation, raises concerns about the relevance of the N. antipodiana structure. Our study shows that the structured part of TuSp NT begins already at residue 10, both for the wt protein at pH 5.5 ( Figure 3 ) as well as the A70E mutant at pH 8.0 ( Supplementary Figure S2 ). Hence, the complete first α-helix (residues S12-I27) and beginning of the second α-helix (residues P31-Q36) is lacking in the N. antipodiana TuSp NT structure. Several hydrophobic residues as I80 and I98 showing surface localization in the N. antipodiana structure, which have been suggested to play a role in the intermolecular association of TuSp NT ( Wang et al., 2021 ) are buried in the hydrophobic core in our structure ( Figure 6 ). Additionally, replacement of the α1-α4 helical interface by α3-α4 as in the N. antipodiana structure would not be possible in a full-length construct, because C22 from α1 forms a disulfide bond with C102 from α4, placing these helices next to each other. Altogether, the newly obtained results indicate that, under physiological conditions, full-length TuSp NT adopts a five-helix bundle structure as the other NTs. FIGURE 6 Superposition of the structures of A. argentata TuSp NT dimer (PDB ID 6TV5, cyan and gray) and N. antipodiana TuSp NT monomer (PDB ID 2K3Q, green). Helices of A. argentata TuSp NT are identified with black font, whereas helices of N. antipodiana TuSp NT are identified with red font. Hydrophobic residues mutated by Wang et al. (2021) are shown with stick representation in both structures in red ( A. argentata TuSp NT) and in orange ( A. antipodiana TuSp NT). The pH dependent dimerization of MaSp and MiSp NT was found to be mediated by sequential protonation of three specific glutamic acid residues (E79, E84, E119 in MaSp and E73, E76, E115 in MiSp NT). In the A. argentata TuSp NT, only E84 (E82 in TuSp NT) is structurally preserved, whereas E79 and E119 are replaced by D77 and R117, respectively, and form an intermolecular salt bridge in the dimer structure. These substitutions alter the electrostatic surface potential of each monomer subunit, making the positively charged pole more extensive (i.e. spread out) than in MaSp NT ( Figure 7 ), which could facilitate intermolecular association. Additionally, the charge repulsion between two glutamate residues at the dimer interface is abolished potentially allowing the subunits to associate before the protonation events. This could explain why we were unable to stabilize the monomeric conformation of TuSp NT for NMR structure determination. Besides these charge interactions, pre-arrangement of α-helices has been suggested to be important for the NT dimerization. In MaSp NT the helical reorganization is facilitated by swinging out of the wedged W10 residue and subsequent repacking of the hydrophobic core. In TuSp NT W10 is replaced by F10, which is buried in the hydrophobic core of the dimer structure. This is similar to FlSp NT, for which the monomer and dimer structures showed the same buried F10 side chain orientation ( Sarr et al., 2022 ). Thus, the structural reorganization during TuSp NT dimerization does not seem to require relocation of F10 side chain. Heiby et al. (2019) reported that the unusually high content of methionines within the core region of E. australis MaSp NT plays an important role in the monomer-dimer structural transition. Exchange of the core methionines with the bulkier leucines made the monomeric MaSp NT more rigid, which abolished the movement of its helices and ability to form a dimer. The lack of the tryptophan and methionine residues in TuSp NT sequence ( Figure 1 ) could abolish its hydrophobic core plasticity, locking the core in a dimer-like conformation also at neutral pH, albeit with different dynamics. Instead, the monomer-dimer structural rearrangement likely involves slight movement of helices due to neutralization (protonation) of repulsing charges across intramolecular helical interfaces. FIGURE 7 \n (A) Electrostatic potential of A. argentata TuSp NT dimer subunit. (B) Electrostatic potential of E. australis MaSp NT (PDB ID 2LTH) dimer subunit. Red color indicates negative charges, and blue color indicates positive charges. The surface-exposed charged residues are labeled. (C) and (D) shows contact surfaces of the residues R117 and E119 from the opposite subunit at the TuSp NT and MaSp NT dimer interfaces, respectively. Since E82 shows a similar side chain orientation to E84 in MaSp NT and additionally has E37 located in close proximity, we hypothesized that neutralization of these two residues may be important for establishing the characteristic handshake interaction between E82 and D40 as seen in the MaSp NT dimer. Although E37 is not conserved in most other spidroins, it is highly conserved between TuSp NTs from different species ( Supplementary Figure S6 ). However, analysis of the double mutant TuSp NT E37QE82Q showed that some other residue is additionally protonated at low pH. The only other well conserved glutamate residue among TuSp NTs is E85, which in some conformers of our structure is spatially close to D130 of the same subunit and the charge clash could prevent it from assuming a dimer-compatible conformation. However, preparation and characterization of the triple and quadruple mutants TuSp NT E37QE82QE85Q and TuSp E37QE82QE85QD130N gave similar results as for TuSp NT E37QE82Q . This result does not rule out the involvement of at least some of the mutated residues in the observed pH sensitivity, because the studied combinations may bear disruptive mutations affecting the pK a of other residues that participate in the protonation events of the wt protein. In summary, our obtained results clearly indicate that the mechanism of pH-dependent dimerization is different for TuSp NT than it is for MaSp and MiSp NT. We show that in contrast to previous findings TuSp NT has the same five-helix fold of other NTs and forms a stable dimer at low pH. However, its unique amino acid sequence does not allow full stabilization of the monomer conformation at the conditions, where MaSp NT forms stable monomers (pH 7.2 in presence of 300 mM NaCl). This may be linked to the lack of a well-defined spidroin storage sac as well as different physiological conditions (especially, pH gradient) in the tubuliform glands, which remain to be characterized. Furthermore, the pH dependent stable dimer formation that takes place in the silk duct upon fiber formation involves a very different set of amino acid residues as compared to MaSp and MiSp NT. Further research is needed to identify the exact amino acid residues that become protonated in the TuSp NT dimer at low pH."
} | 5,033 |
40307843 | PMC12044819 | pmc | 6,993 | {
"abstract": "Background Interactions between fungi and bacteria have the potential to substantially influence soil carbon dynamics in soil, but we have yet to fully identify these interactions and partners in their natural environment. In this study, we stacked two powerful methods, 13 C quantitative stable isotope probing (qSIP) and cross-domain co-occurrence network, to identify interacting fungi and bacteria in a California grassland soil. We used in-field whole plant 13 CO 2 labeling along with sand-filled ingrowth bags (that trap fungi and hyphae-associated bacteria) to amplify the signal of fungal-bacterial interactions, separate from the bulk soil background. Results We found a total of 54 bacterial ASVs and 9 fungal OTUs that were significantly 13 C-enriched. These were saprotrophic and biotrophic fungi, and motile, sometimes predatory bacteria. Among these, 70% of all 13 C-enriched bacteria identified were motile. Notably, we detected fungal-bacterial network links between a fungal OTU of the genus Alternaria and several bacterial ASVs of the genera Bacteriovorax, Mucilaginibacter , and Flavobacterium , providing empirical evidence of their direct interactions through C exchange. We observed a strong positive co-occurrence pattern between predatory bacteria of the phylum Bdellovibrionota and fungal OTUs, suggesting the transfer of C across the soil food web. Conclusions To date, our ability to associate microbial co-occurrence network patterns with biological interactions is limited, but the incorporation of qSIP allowed us to more precisely detect interacting partners by narrowing in on the taxa that were actively incorporating plant-fixed, fungal-transported labeled substrates. Together, these approaches can help build a mechanistic understanding of the complex nature of fungal-bacterial interactions in soil.\n Supplementary Information The online version contains supplementary material available at 10.1186/s40168-025-02100-2.",
"introduction": "Introduction In soil, fungi and bacteria share the same physical space and assemble into dynamic communities that consume, process and translocate plant-derived organic matter that contribute to nutrient cycling [ 1 ]. In this space, the “hyphosphere” (i.e., the area of fungal influence surrounding hyphae) impacts the assembly and activity of bacterial communities [ 2 ], and potentially leads to a selection of a core microbiome [ 3 , 4 ]. Functionally, these two groups of organisms complement each other and occupy distinct physical soil habitats [ 5 ], where fungi are better at metabolizing complex organic substrates and bacteria are more efficient in using simpler organic compounds [ 6 , 7 ]. Given that fungi and bacteria have some overlap in spatial niches and nutrient resources, the interactions between them, whether competitive or synergistic, are expected to have a substantial impact on global soil nutrient cycling [ 8 ]. However, despite the ubiquity of their distribution across all soil ecosystems, deciphering the dynamic and complex interactions within the hyphosphere and linking them to soil processes still remains a challenge. This is especially true for quantitative assessments of these interactions. By having a more quantitative understanding of fungal-bacterial interactions in the soil, especially in relation to a specific nutrient, we could better characterize how these microorganisms contribute to nutrient cycling, organic matter decomposition, and their role in the soil food web.\n As a first step to understanding fungal-bacterial interactions and their functional contributions to soil processes, some studies have used quantitative stable isotope probing (qSIP) [ 9 ] to track the fate of C fixed by plants, transferred to fungi such as arbuscular mycorrhizal fungi (AMF), and eventually to hyphosphere bacteria [ 10 , 11 ]. This transfer can occur through the assimilation of fungal-released compounds by bacteria associated with the hyphae [ 12 ] or by bacteria predating directly on the hyphae [ 13 ]. Some of the carbon from microbial necromass may come to persist as part of mineral associated organic matter [ 11 , 14 ]. The use of qSIP in these studies established the connection between microbial identity and their functional activity in their natural environment. More specifically, these studies showed cross-domain interactions among arbuscular mycorrhizal fungi (AMF), ammonia-oxidizing archaea, and many types of bacteria [ 10 , 11 ]. qSIP has also been used to provide quantitative measurements of potential antagonistic relationships between predatory and non-predatory bacteria; a recent qSIP meta-analysis suggests predatory bacteria grew 36% times faster and assimilated C at a 211% higher rate than non-predatory bacteria [ 15 ]. While these studies showed the pathways of C flow through the soil, the characterization of soil bacteria actively consuming fungal C through interactions with fungal hyphae in situ remains to be identified. Recent developments in network analysis techniques have offered more comprehensive characterization of community composition and abundance. Co-occurrence patterns identified through networks have increased our capacity to identify potential interactions among community members. While network analyses of individual groups of organisms are frequently used to infer interactions, a small number of studies have explored cross-domain co-occurrence networks of both bacteria and fungi. These studies show broad connections between network topology and the links among network members. For example, fungal-bacterial co-occurrence patterns can differ between arbuscular mycorrhizal fungi and nonmycorrhizal fungi across soil niches [ 16 ], they may be affected by different forms of organic nutrients [ 17 ], and may be connected to soil aggregate structure [ 18 ]. The network links among fungi and bacteria within a substrate suggest potential indicators of bacteria consumption of fungal biomass [ 16 ]. This has been experimentally observed by identifying isotopically labeled bacteria and fungi that fed on 13 C and 15 N labeled fungal necromass as a substrate [ 19 ]. However, a key limitation of ecological network analysis is its reliance on pairwise correlations, which may not accurately represent the biological interactions in natural environments. Moreover, this approach typically provides only a static snapshot at a single time point, neglecting the dynamic nature of ecosystems as well as potentially capturing links between taxa that are inactive during that time. By focusing solely on relative abundance patterns between pairs, we miss the broader community context, overlooking other organisms that likely influence these relationships. Additionally, spatial variation and ecological niche differences, particularly between fungi and bacteria [ 20 ], are often not adequately considered, which may lead to an incomplete understanding of ecosystem complexity when constructing a network. Despite their limitations, cross-domain networks provide ways to condense the high-dimensionality of microbiome data and are increasingly used to hypothesize the nature of fungal-bacterial interactions. While qSIP alone can be used to measure growth rates or indicate which fungi and bacteria are part of the plant-derived soil C cycle, when combined with co-occurrence network analysis, can more precisely hypothesize the abundance patterns between the microorganisms that exist in an environment in relation to their nutrient dynamics. We propose that combining these two complementary methods ( 13 C qSIP with isotope-enabled cross-domain network analysis), will allow us to better target interacting fungal-bacterial partners and generate more directed hypotheses. Here we demonstrate the utility of this approach in a grassland soil system. We followed the path of 13 C labeled isotopes fixed by plants into the hyphosphere and found strong links among the active bacteria and fungi involved in C cycling that are more precisely identified compared to total microbial community networks calculated from unlabeled data. Together, these approaches can significantly contribute towards building a mechanistic understanding of the complex nature of fungal-bacterial interactions in soil.",
"discussion": "Discussion Ingrowth bags focus fungal-bacterial interactions The ingrowth bags were effective in trapping soil fungi [ 42 , 43 ], and although challenging, we were able to detect and identify 13 C enriched hyphosphere bacteria with this method. Different from plant-associated soil compartments (e.g., rhizosphere), in which microbes directly incorporate plant metabolites, the ingrowth bag system experiences a dilution of isotopic signals where 13 C from a pulse label is released by plant roots, consumed by rhizosphere fungi that extend beyond the rhizosphere into the ingrowth bags, and finally into the hyphosphere where they are assimilated by associated bacteria. This resulted in a system with low biomass and low overall 13 C-enrichment compared to the rhizosphere. Although the difference between the average DNA density in 12 C and 13 C samples was small (indicated by the small separation between DNA density curves in Supplemental Fig. S3), the EAF values are within previously reported ranges in the hyphosphere [ 10 , 11 ]. This shift in density was enough for us to detect the taxon-specific enrichment of 13 C signal in both fungi (39 OTUs) and bacteria (214 ASVs; Fig. 1 ). However, the number of bacteria that significantly incorporated 13 C (54 ASVs with lower confidence intervals greater than 0% 13 C EAF) was several times lower than previous studies on soil incubated with glucose [ 9 ], incorporated maize residues [ 44 ], and hyphosphere grassland soils [ 10 , 11 ]. Notably, the majority of the highly 13 C-enriched bacterial genera that entered the ingrowth bags are motile (Fig. 2 ), consistent with the idea that they followed the “fungal highway” into the ingrowth bags. Since the bags were filled with nutrient-depleted quartz sand, the bacteria likely obtained nutrients through assimilation of hyphal exudates [ 12 ] or predate directly on the hyphae [ 13 ], leading to 13 C enrichment in their DNA. We cannot rule out the possibility that the bacteria could have also assimilated mobile nutrients that diffused passively into the bags, which is a limitation of the ingrowth bag method, but also reflects the difficulty of studying microbial interactions in the soil. Together, we found that although challenging, ingrowth bags are an environment that minimizes the complexity within the soil and thus are an advantageous system to study soil fungi, the bacteria that associate with them, as well as C dynamics in fungal-bacterial interactions in situ. Within the isolated ingrowth bag system, we found saprotrophic fungi to be highly 13 C-enriched. Saprotrophic fungi are pivotal decomposers in terrestrial ecosystems, catalyzing the conversion of complex organic compounds to more simple organic compounds, which are then available as either mycelium or hyphal exudates that can be consumed by bacteria. Indeed, it has been shown that fungal-bacterial interactions related to C acquisition occur by the utilization of fungal-released compounds by bacteria associated with hyphae [ 12 ] or through competitive interactions with the molecules liberated by fungi at the hyphal tips [ 45 , 46 ]. Here, we found a small subset of fungi such as Aspergillus and Coprinus , that had significantly incorporated 13 C into their cells (Fig. 2 ). In contrast, we also found saprotrophic fungi that were not enriched in 13 C. These were litter and wood saprotrophs of the genera Mycoarthris, Chaetomium, Bjerkandera, Xylodon, among others . The ability of some fungi to assimilate 13 C suggests their ability, and perhaps preference, to consume root exudates rather than the more difficult to decompose plant polymers or stabilized soil organic matter. While this experiment cannot directly show the hyphosphere effect of litter or wood decomposing fungi, it is possible that they can effectively distribute and make available plant-derived C across soil compartments [ 47 , 48 ]. As with saprotrophic fungi, we found biotrophic fungi, including plant pathogenic fungi like Fusarium and Aureobasidium, and animal parasite like Ijuhya highly 13 C-enriched (Fig. 2 ). Their direct connection to plant roots (or animals that graze directly on plant roots) provides a direct pathway for 13 C to travel into the soil. It should be noted that these fungi can also have a saprotrophic lifestyle, therefore they may also distribute C through the saprotrophic pathway. Similarly, arbuscular mycorrhizal fungi (AMF) also have direct connections to plant C. While we found that 194 AMF OTUs did enter the ingrowth bags with low relative abundance (Supplemental Fig. S5), they were not detected as labeled taxa, perhaps due to their low GC DNA content [ 49 , 50 ] that did not separate out well during the isotope fractionation. Therefore, we cannot rule out the fact that AMF, which are often highly labeled according to NanoSIMS measurements, can transfer a sizable pool of 13 C into the hyphosphere (Kakouridis et al. 2024), and that the 13 C enriched bacteria here may have also consumed 13 C from mycorrhizal fungi. While it has been a significant challenge to associate fungal-bacterial occurrence patterns to soil ecosystem functions [ 51 ], the patterns of saprotrophic and biotrophic fungi supporting associated bacteria reinforce the concept of bacterial nutrients exchange with fungi in adverse and nutrient-poor environments [ 52 ]. We propose that although these different fungal trophic guilds initially acquire plant C through different pathways, ultimately, they distribute and release plant-derived C beyond the rhizosphere through interactions with the hyphosphere microbiome. We compared the results from this study with previous studies that used soils from an adjacent field site and found some insightful connections of fungi to the hyphosphere microbiome. Each of the study design allowed for an increased focus on the bacteria associated with hyphae. These ranged from soils with detritus [ 48 ], soils influenced by hyphae [ 11 ], soils with abundant and visible hyphae [ 10 ], and ingrowth bags that separated hyphae from surrounding soil (this study). While the phyla Actinobacteriota, Bacteriodota, Proteobacteria, and Verrucomicrobia (Fig. 1 ) are consistently represented across these studies, low representation of Acidobacteriota, Bacillodota, Chloroflexi, Plactomycetota, and ammonium oxidizing archaea in this study suggests that certain phyla are more connected to the complexity of the soil whereas others may be more dependent on the direct connection with fungal hyphae. For example, relative abundance, as well as 13 C enrichment of the bacteria phylum Proteobacteria changes in the presence of arbuscular-mycorrhizal fungi [ 11 ]. In particular, with the presence of hyphae, there is a higher abundance of predatory bacteria in the phyla Bdellovibrionota (this study) and Myxococcota (Nuccio et al. 2022). While the general biology of the Verrucomicrobia is largely unknown, their presence connected to fungal hyphae suggests that at least some of the members of this group have some associations with fungi. These responses could be related to fungal-bacterial interactions in the hyphosphere. In contrast, we only detected a single ASV of the phylum Acidobacteria, which was commonly found in other studies and are often highly abundant in soils [ 53 ]. Their low enrichment here suggests that these common soil bacteria may not actively interact with labile fungal carbon. We observed an absence of labeled ammonium oxidizing archaea that was found in all other studies. Since saprotrophic and biotrophic fungi can make ammonium available through decomposition, we would expect a higher level of ammonium availability. However, within a quartz sand substrate that has low ammonium adsorption and availability [ 54 , 55 ], low microbial biomass and therefore low CO 2 from respiration, there is little electron donor nor acceptors to support ammonium oxidizing archaea. These patterns of presence and absence gleamed through multiple studies of the same soil provided good insights into the occurrence between fungi and associated bacteria taxa, but additional approaches are required to build direct connections of their relationships and nature of interactions. Isotope-enabled networks more precisely predict fungal-bacterial interactions Understanding the mechanism of microbial interactions in soil has been a daunting challenge, and while isotopic in combination with ‘omics methods allow us to detect trophic interactions where some form of nutrient exchange had occurred, they do not tell us which organisms might be interacting with each other. To make more concise predictions on specific microbes and their potential interactions, we built cross-domain networks with the organisms that have incorporated 13 C into their cells. This approach allowed us to narrow down and identify potentially interacting fungal-bacterial partners. We found that compared to the total community ( 12 C) network, the enriched community ( 13 C) network had a much higher complexity, with fewer nodes and more edges, and more fungal-bacterial links (Fig. 3 ). The majority of the highly 13 C-enriched fungi (6 of 9) that links to highly 13 C-enriched bacteria (46 of 54) in this nutrient-poor system suggests that the majority of the active bacteria were primarily associated with fungi in C exchange. This 13 C exchange was elegantly illustrated by [ 56 ], who showed that fungi can transfer nutrients to bacteria from nutrient-rich to nutrient-deprived habitats, highlighting the importance of fungal-bacteria interactions in nutrient-poor environments. While co-occurrence patterns based on amplicon sequencing data are a useful tool to create hypothesis for further testing cross-domain interactions [ 57 ], the information regarding a specific nutrient dynamic might not be fully captured. Even though the higher number of fungal-bacterial interactions we observed in the enriched community ( 13 C) network may be an attribute of the lower RMT cut-off or the reduced number of features used to construct the network (as shown by the comparison with the mock network), enriched community ( 13 C) networks allowed us to detect more precise fungal-bacterial associations by narrowing down only taxa that have actively incorporated a labeled substrate. In complex environments such as the soil, the combination of isotopic and network analysis methods brings us a step closer to understanding the mechanisms of fungal-bacterial interactions in the context of nutrient dynamics. The type of network links (positive or negative) allows us to infer the mode of interactions, such as the relationship among certain fungi and bacteria that are highly 13 C-enriched (Fig. 4 ). While the strong positive co-occurrence between Alternaria with Mucilaginobacter or Flavobacterium , and Podospora with Pedobacter , Peredibacter , or Oligoflexus suggest that both fungi and bacteria either increase or decrease in abundance, the mechanism of interactions is not clear. Notably, bacteria of the phylum Bdellovibrionota, such as Peredibacter , Bacteriovorax , and Oligoflexus , are prey-selective predatory bacteria, predating on gram-negative bacteria, are capable of incorporating 13 C labeled prey biomass [ 58 ], and are highly 13 C-enriched in this study. This amplification of isotopic signals occur in higher rates in predators, allowing them to achieve higher levels of substrate incorporation in a short period of time [ 10 , 15 ]. These predatory taxa had strong positive co-occurrence with fungi but negative co-occurrence with certain gram-negative bacteria including Perlucidibaca and Mucilaginibacter (Supplemental Fig. S6). We would therefore expect that as the abundance of fungi increases, the abundance of bacteria that feed on fungal carbon also increases, followed by an increase in the abundance of predatory bacteria. Indeed, we found these patterns in this study. There was an increase in abundance of fungi ( Alternaria and Podospora ) with an increase in the abundance of bacteria ( Flavobacterium , Mucilaginobacter , and Pedobacter ) that consumed fungal carbon; an increase in abundance of predatory bacteria ( Peredibacter , Bacteriovorax , and Oligoflexus ); and a decrease in the abundance of certain gram-negative bacterial taxa ( Perlucidibaca and Mucilaginibacter ) as the abundance of predatory bacteria ( Bacteriovorax and Peredibacter) increases. While these patterns require longitudinal experiments to confirm, the links of the enriched community ( 13 C) network narrowed the complexity of soil fungal-bacterial interactions and provided an intriguing set of hypotheses to test inter-domain food webs in soil. Study limitations and future considerations The use of ingrowth bags, qSIP, and isotope-enabled networks gave us a powerful set of tools to narrow in on fungal-bacterial interactions in the hyphosphere, but we recognize that there are limitations that should be considered for future studies. We acknowledge that the small number of replications is a limitation of our study and therefore not representative across larger environments. Since more replicates and DNA density fractions are needed in order to identify taxon-specific enrichment in samples with lower isotopic enrichment [ 59 ], we combined fractions within similar density gradients into new binned fraction groups to yield enough DNA to sequence the maximum possible fractions (7 total) in our system. For studies considering using the same ingrowth bag system, we suggest deploying larger volume bags to increase biomass recovery, adding more replicates to potentially increase the number of significantly labeled taxa detected, and incorporating other ‘omics techniques to better understand the functional contributions of the partners involved. We focused this study on fungal-bacteria interactions, therefore we did not discuss the bacterial-bacterial or fungal-fungal interactions, except for the case of predatory bacteria. Similarly, we did not study the many others organism that are part of the soil food web that are important drivers of soil C dynamics. As with any correlation methods, the inferences of 13 C fungal-bacterial interactions drawn from qSIP and co-occurrence network data should be synthesized critically, taking into consideration both ecological factors such as previously knowledge of the food-web (e.g., predatory bacteria), as well as methodological constrains (e.g., prevalence filtering methods and fractions within replicates not being not fully independent observations) associated with the study design. Nevertheless, the combination of the two methods allows us to detect the bacteria that have consumed presumably fungal 13 C and hypothesize their direct interactions that may be tested in future experiments. This study demonstrated the utility of ingrowth bags, SIP, combined with analytical qSIP and co-occurrence network analysis to predict potential fungal-bacterial interactions in dynamic natural systems processes such as C cycling. Given that qSIP can quantify active taxa involved in a specific nutrient cycling process and co-occurrence networks have the ability to correlate their abundance patterns, in combination they provide means to identify potential trophic interactions such as predation and facilitative links as demonstrated in this study. While there are drawbacks with these methods, they enabled us to focus on potential interactions that can be experimentally tested to further elucidate the mechanisms that underlie C dynamics. Ultimately, this novel approach provides significant progress towards building a mechanistic understanding of the complex nature of fungal-bacterial interactions in soil."
} | 6,014 |
23194418 | PMC4053740 | pmc | 6,994 | {
"abstract": "Genome-scale metabolic network reconstructions are considered a key step in\nquantifying the genotype-phenotype relationship. We present a novel gap-filling\napproach, MetabolIc Reconstruction via functionAl GEnomics (MIRAGE), which identifies\nmissing network reactions by integrating metabolic flux analysis and functional\ngenomics data. MIRAGE's performance is demonstrated on the reconstruction of\nmetabolic network models of E. coli and Synechocystis sp. and\nvalidated via existing networks for these species. Then, it is applied to reconstruct\ngenome-scale metabolic network models for 36 sequenced cyanobacteria amenable for\nconstraint-based modeling analysis and specifically for metabolic engineering. The\nreconstructed network models are supplied via standard SBML files.",
"conclusion": "Conclusions Our paper presents a novel method, MIRAGE, for reconstructing metabolic network models\nby integrating metabolic flux analysis and functional genomics data to resolve network\ngaps. MIRAGE was validated based on a comparison of its predictions with manually\ncurated metabolic networks for E. coli [ 46 ] and Synechocystis sp. PCC 6803 [ 50 - 52 ]. Then it was applied to reconstruct metabolic network models for an ensemble\nof cyanobacteria, with the resulting networks shown to be amenable for metabolic\nengineering applications of astaxanthin secretion. Our results show that functional genomics data enable the marked improvement of\ngap-filling in metabolic networks. Furthermore, we show that the integration of more\nthan one type of functional genomics data can further improve the performance of MIRAGE.\nNaturally, MIRAGE can be extended to account for additional functional genomics data,\nincluding protein-protein interactions and genomic context data, which were previously\nused for the identification of missing gene annotations in metabolic networks [ 38 ]. Metabolomics data can also be integrated within MIRAGE, to enable the\ndefinition of a metabolite core, consisting of metabolites that are known to be\nsynthesized, and hence corresponding pathways that connect them to the rest of the\nnetwork must be identified [ 26 ]. Several existing gap-filling methods work by searching for a minimal set of missing\nreactions that would enable the network to perform a certain task [ 22 ]. MIRAGE extends upon these methods by enabling the identification of pathways\nthat are not necessarily minimal in size, if supported by functional genomics data.\nHowever, MIRAGE is still limited in being unable to predict the presence of alternative\npathways, in case either one is sufficient to fulfill its defined objectives. This may\nexplain the relatively low recall levels achieved by MIRAGE and the other tested\napproaches. For example, this was demonstrated in Figure 1 , where\nreaction R6 will not be predicted for gap-filling, as an alternative pathway that\nfulfills the required metabolic objectives was chosen. The identification of alternative\npathways based on more complex integration of functional genomics data with metabolic\nflux analysis is currently an open challenge for all known gap-filling algorithms. An\nadditional limitation of MIRAGE is that it does not explicitly account for thermodynamic\nconsiderations as part of the network reconstruction process. Future implementation of\nthis approach may formulate additional thermodynamic constraints as part of the model\nconsistency check, as suggested in Thermodynamic Metabolic Flux Analysis (TMFA) [ 58 ] (which would require further speedups to obtain reasonable running\ntimes). Metabolic models generated by automated methods such as MIRAGE should be regarded as\nfirst draft models, requiring further manual curation to bring them up to comparable\nlevel with standard manually curated models. The growing interest in reconstructing\nmetabolic network models for hundreds of species raises the challenge of developing\nimproved such gap-filling approaches that could speed up the reconstruction process,\nwhile the approach presented here shows a marked improvement in this direction over the\nstate-of-the-art, supporting the advantage of integrating functional genomic data as\npart of model reconstruction. We expect MIRAGE to be used for automatic reconstructions\nof many other species, leading to a significant boost in the understanding of their\nmetabolism.",
"discussion": "Results and discussion MIRAGE MIRAGE is a functional genomics-based model reconstruction approach that aims to\ngenerate a genome-scale metabolic network model for an organism of interest, given a\ncore set of reactions that are known to exist in its network, and optionally, a\ndefinition of a biomass reaction. The core set of reactions can be automatically\nderived strictly from genomic data, based on strong sequence similarity with known\nenzyme-coding genes in other species. The method then aims to find missing reactions\n(from a universal database of candidate gap-filling reactions such as the Kyoto\nEncyclopedia of Genes and Genomes (KEGG)), supported by functional genomics data,\nwhose addition to the network would lead to a functional model. The method follows a\ntwo-step procedure, starting with the utilization of functional genomics data to\nestimate the probability of including each reaction from the universal database in\nthe reconstructed network, and then, metabolic flux analysis that selects the most\nlikely set of reactions whose addition to the network would satisfy the above\ndescribed objectives. For the first step, we utilize two functional-genomics data sources to estimate the\nlikelihood that a reaction from a universal reactions database should be included in\nthe target metabolic network: (i) enzymes' phylogenetic profiles, and (ii) gene\nexpression. Specifically, we define a weight for each reaction in the universal\ndatabase (that is not already included in the reconstruction's core reactions set),\nbased on the functional similarity between neighboring enzymes, in terms of\nresemblance of phylogenetic profiles, and correlation in gene expression of the\nenzyme-coding genes (Materials and methods). Enzyme phylogenetic profiles were extracted from KEGG, representing a pattern of\nenzyme presence or absence across an available collection of species. For each\nreaction in KEGG, we computed a phylogenetic weight, representing the likelihood for\nits inclusion in the network reconstruction. Specifically, the phylogenetic weight of\na certain reaction is calculated based on the maximal Jaccard coefficient between its\nphylogenetic profile and the corresponding profiles of its neighboring core reactions\nin the network (Materials and methods). Similarly, an expression weight for each\nreaction was calculated by evaluating gene expression profiles (measured in the\ntarget organism) of potential enzyme-coding genes (considering all non-annotated\ngenes in the genome), compared with the expression profiles of known genes associated\nwith neighboring core reactions. The sum of the phylogenetic and expression weights\nafter proper normalization was used as input for the second reconstruction step\n(Materials and methods). The second reconstruction step aims to find a set of high weight gap-filling\nreactions that satisfy the objectives described above. Towards this goal, we employed\nthe following reaction pruning procedure. Starting from a metabolic network model\nconsisting of all reactions in the universal reaction database, we iteratively remove\npotential gap-filling reactions, as long as the removal does not affect the\nconsistency of the model. In each iteration, the probability of choosing a certain\nreaction for removal is inversely proportional to its weight - that is, low weight\nreactions have a higher probability to be chosen first for removal. The model\nconsistency check procedure involves the usage of constraint-based modeling to verify\nthat the remaining network (i) enables each core reaction to carry non-zero metabolic\nflux within a stoichiometrically balanced flux distribution, accounting for reaction\ndirectionality constraints, (ii) enables the production of all essential biomass\nconstituents, and (iii) accounts for the growth-associated dilution of all network\nmetabolites (that is, guaranteeing that the network consists of complete pathways for\neither the transport or de novo synthesis of all metabolites that exist in\nthe network) [ 45 ]. Since the reactions' scanning order may affect the resulting model, the\nalgorithm is executed several times with different, random pruning orders (Materials\nand methods). The fraction of obtained models that contains a certain reaction\nreflects the confidence that it should be included in the final model. Hence, to\nconstruct the final metabolic network model, we run the reactions removal procedure\nagain, based on an ordering defined by the received confidence values (Materials and\nmethods). Notably, the presented method extends upon the Model Building Algorithm (MBA) of\nJerby et al. [ 28 ] that was recently used to reconstruct a model of human liver metabolism.\nThe MBA method addresses only the first objective from the above list, while not\naccounting for biomass production and growth-associated metabolite dilution, which\nare of less importance for the modeling of human tissue metabolism. Furthermore, it\naccounts for functional genomics data in a more limited manner, by using them only to\ndefine two core sets of reactions with either a moderate or high probability to be\nretained in a specific tissue model. In contrast, MIRAGE assigns a continuous score\nper each reaction that reflects its probability to be retained in a specific species\nmodel, allowing us to make better use of these data. The described method is computationally demanding since each trial of the random\nreaction pruning procedure (out of the 500 trials performed to gather sufficient\nconfidence statistics), requires eliminating each reaction from the universal\nreactions set in turn, and checking the consistency of the resulting model.\nImplementing the speedup heuristic suggested by Jerby et al. [ 28 ], which aims to minimize the number of linear optimizations required in\neach model consistency check, provided some improvement in running time. However,\neach random pruning trial still took around 35 hours, which made the entire method\ncomputationally intractable. The significant increase in running time in comparison\nto the method of Jerby et al. resulted from the markedly large size and\ncomplexity of the universal reaction database in comparison to the human network\nmodel used by Jerby et al ., and the additional reconstruction objectives\npreviously not accounted for. To overcome this, we implemented the following additional speed-up techniques\n(Materials and methods). First, the model consistency check procedure is based on\nidentifying a set of flux distributions in which all core reactions are activated\n(that is, have non-zero flux), and is applied following the removal of each reaction\nin the reaction pruning procedure. The first speed-up involved the utilization of\nflux distributions computed in one call to the model consistency check procedure in\nsubsequent calls to this procedure (testing the potential removal of subsequent\nreactions in the pruning order) to avoid time-consuming linear programming\noptimizations. Second, to further minimize the number of performed linear\noptimizations, the latter are now formulated with the objective of minimizing flux\nthrough subsequent gap-filling reactions in the pruning order. These two speed-up\ntechniques, significantly elaborated upon in Additional file 1, provide a 100-fold\nimprovement in running time. Figure 1 illustrates the working of MIRAGE on a toy model.\nReactions E1, E8, E9 and E10 are core reactions, while all the other reactions are\ncandidates for gap-filling. MIRAGE predicts the addition of reactions E2, E3, E4 and\nE7 to enable flux activation of all core reactions, biomass production, and\naccounting for growth dilution of all metabolites in the core. The inclusion of\nreactions leading from M3 to M5 is required to enable flux activation of core\nreactions E8 and E9. In this case, the choice of including both reactions E3 and E4\nfor gap-filling, instead of the single reaction E5, is based on higher support for\nthe former reactions in the functional-genomic data. Reaction E2 is predicted for\ngap-filling to compensate for growth-associated dilution of metabolites M6 and M9 [ 45 ]. Figure 1 The application of MIRAGE on a toy model . Core reactions (E1, E8, E9 and\nE10) are marked with straight lines, while gap-filling reactions are marked\nwith dashed lines. A weight for each reaction is computed based on the\ncorrelation of its phylogenetic and expression profiles with those of\nneighboring core reactions in the network. Reactions predicted for gap-filling\nby MIRAGE are in red. Specifically, E2 is chosen to enable the\ngrowth-associated dilution of metabolites M6 and M9. E3 and E4 are chosen\n(instead of E5, which has a significantly lower weight) to enable the flux\nactivation of E8 and E9. E7 is chosen to enable flux activation of E8 and E9\nunder steady-state. Reaction E6 is not chosen for gap-filling as it is\nredundant given the above-mentioned chosen essential reactions. Validation of MIRAGE in the reconstruction of a metabolic network for E.\ncoli To evaluate the performance of MIRAGE, we applied it to reconstruct a metabolic\nnetwork model for E. coli , for which a comprehensively curated model\n(iAF1260) is already available for validation [ 46 ]. Towards this end, we extracted a cross-species reactions dataset from\nKEGG having 7,211 reactions (referred to as the universal reactions set). To define a\ncore set of known E. coli reactions to be used by MIRAGE, we considered KEGG\nreactions annotated as existing in E. coli and also belonging to iAF1260,\nplus the known biomass and all exchange reactions from iAF1260. Then we removed\ndead-end reactions that cannot be activated within a feasible flux distribution when\nconsidering the entire universal reactions set, yielding a core set of 812 reactions.\nPerforming standard flux variability analysis [ 47 ] when focusing only on this set of 812 core reactions revealed that 45%\n(365/812) of these reactions are on dead-ends. MIRAGE's task is hence to identify\ngap-filling reactions that would resolve these dead-ends, aiming to identify a\nremaining set of 109 reactions from iAF1260. Notably, our analysis did not account\nfor subcellular localization of metabolic processes, and hence duplicated reactions\nin iAF1260 that correspond to multiple compartments were removed. Comparison of MIRAGE's reconstructed network model for E. coli with iAF1260\nshows a predictive precision of 41.9% and recall of 24.3%, which is significantly\nbetter than random sampling of gap-filling reactions (hyper-geometric\n P -value <10 -16 ; Figure 2 ; Additional\nfile 1, part 6, and Supp. Table 1 in Additional file 1). As controls, we assessed the\npredictive performance of using only the functional genomics data based on the\ncomputed reaction weights (by ordering potential gap-filling reactions based on their\ncomputed weights), and the predictive performance of MIRAGE without utilizing\nfunctional genomics data (by assigning reactions with random weights; as done in the\nMBA algorithm). Using only the functional genomics data, the resulting predictive\nperformance was significantly lower than that of MIRAGE (Figure 2 ), reaching a precision of 6.1%, under a recall level of 19.6%\n( P -value = 2 × 10 -9 ). Without utilizing functional\ngenomics data, the predictive performance was also markedly lower, with a precision\nof 27.5% and recall of 20.6% ( P -value <10 -16 ). Using only gene\nexpression [ 48 ] or phylogenetic weights (based on all species in KEGG) provided lower\nprecision of 31.8% and 36.9%, respectively, with slightly lower recall levels (19.6%\nand 22.4%, respectively) to those achieved when utilizing both (Figure 2 ), demonstrating the importance of integrating multiple\nfunctional-genomics data sources. As a further control, we applied MIRAGE to\nreconstruct a metabolic network model for E. coli , without prior knowledge\nof exchange reactions (which in the above analysis were taken from the model of\niAF1260), finding an overall similar predictive performance, showing an improvement\nof MIRAGE compared to other approaches (Supp. Table 2 in Additional file 1). Figure 2 MIRAGE's predictive performance on reconstructing a known metabolic network\nof E. coli . The precision and recall of MIRAGE is marked with a\nstar symbol. The precision and recall of several controls, including variants\nof MIRAGE that utilize only phylogenetic data, only expression data, or no\nfunctional-genomics data, are marked with a triangle, bar, and circle,\nrespectively. The predictive performance of the functional genomic data (that\nis, by ordering potential gap-filling reactions based on their computed\nfunctional genomic weights, without utilizing metabolic flux analysis) is shown\nby the straight lines: the performance of the phylogenetic data, gene\nexpression, and both data sources are colored green, yellow, and purple,\nrespectively. The performance of random predictions of gap-filling reactions is\ncolored blue. Table 1 Predicted knockout strategies for Astaxanthin over-production in\ncyanobacteria Organism Growth rate (mutant/wt) [h -1 ] Astaxanthin production [μmol gDW -1 h -1 ]\n(min/max) Knockout reaction names KEGG reaction ID/EC number P. marinus as9601 0.026/0.061 2.18/2.18 Dimethylallyl-diphosphate: isopentenyl-diphosphate dimethylallyl\ntranstransferase R01658/EC 2.5.1.1 5-O-(1-Carboxyvinyl)-3-phosphoshikimate phosphate-lyase R01714/EC 4.2.3.5 T. elongatus 0.016/0.048 2.08/2.08 Dimethylallyl-diphosphate: isopentenyl-diphosphate dimethylallyl\ntranstransferase R01658/EC 2.5.1.1 Phosphoenolpyruvate: D-erythrose-4-phosphate\nC-(1-carboxyvinyl)transferase R01826/EC 2.5.1.54 G. violaceus 0.033/0.061 1.64/1.64 Succinate:(acceptor) oxidoreductase R00408/EC 1.3.99.1 Dimethylallyl-diphosphate: isopentenyl-diphosphate dimethylallyl\ntranstransferase R01658/EC 2.5.1.1 Synechococcus pcc7942 0.033/0.061 1.64/1.64 Succinate:(acceptor) oxidoreductase R00408/EC 1.3.99.1 Dimethylallyl-diphosphate: isopentenyl-diphosphate dimethylallyl\ntranstransferase R01658/EC 2.5.1.1 Cyanobacteria cyb 0.032/0.058 1.57/1.57 Succinate:(acceptor) oxidoreductase R00408/EC 1.3.99.1 Dimethylallyl-diphosphate: isopentenyl-diphosphate dimethylallyl\ntranstransferase R01658/EC 2.5.1.1 T. erythraeum 0.051/0.061 0.24/0.24 Hydrogen-carbonate: L-glutamineamido-ligase R00575/EC 6.3.5.5 Dimethylallyl-diphosphate: isopentenyl-diphosphate dimethylallyl\ntranstransferase R01658/EC 2.5.1.1 A. variabilis 0.051/0.061 0/0.96 4-Methyl-2-oxopentanoate: NAD+ oxidoreductase R01651/EC 1.2.1.25 1-Deoxy-D-xylulose-5-phosphate pyruvate-lyase R05636/EC 2.2.1.7 Anabaena 0.051/0.061 0/0.96 3-(4-Methylpent-3-en-1-yl)-pent-2-enedioyl-CoA hydrolyase R03493/EC 4.2.1.57 1-Deoxy-D-xylulose-5-phosphate pyruvate-lyase R05636/EC 2.2.1.7 For each species the table shows the predicted reactions whose knockout is\nexpected to provide maximal astaxanthin production rate, the expected\nastaxanthin production rate, and the expected decline in growth rate. Comparing the predictive performance of MIRAGE on reconstructing the metabolic\nnetwork of E. coli with that of Model SEED [ 22 ] has shown a marked advantage to the former. While the number of core\nreactions considered by MIRAGE and the SEED algorithm in the reconstruction of a\nmetabolic network model of E. coli is close (812 and 826 reactions for\nMIRAGE and SEED, respectively), the number of predicted gap-filling reactions by\nMIRAGE was 62, in comparison to only 10 by SEED. This results from MIRAGE's aim to\nresolve all gap-filling problems instead of just enabling biomass production as\nperformed by SEED. The precision of MIRAGE's predictions was significantly higher\nthan that of SEED, reaching 41.9% for MIRAGE versus 10% for SEED. Re-running MIRAGE\ngiven the very same definition of a biomass reaction used in the SEED reconstruction\nof E. coli 's model (rather than the biomass definition taken from iAF1260)\nstill resulted in a higher number of 76 predicted gap-filling reactions, with a\nsignificantly higher precision of 34.2% than that achieved by SEED. Applying MIRAGE to reconstruct metabolic network models for cyanobacteria To demonstrate the utility of MIRAGE, we applied it to reconstruct genome-scale\nmetabolic network models for 36 cyanobacteria for which genomic data are available to\ndefine core reactions sets. Our analysis spans all cyanobacteria for which enzyme\nannotations are available in KEGG, including Synechocystis, Synechococcus,\nCyanobacteria, Prochlorococcus, Anabaena , and so on [ 49 ]. For all species, we considered the same biomass function, obtained from a\npreviously reconstructed model of Synechocystis sp. PCC 6803 [ 50 ], assuming that CO 2 is the sole carbon source. Due to lack of\ncomprehensive gene expression for most cyanobacteria species, we utilized here only\nphylogenetic data (considering all species in KEGG) to define reaction weights. The average size of a core reactions set for a cyanobacteria network is 570 reactions\n(Figure 3a ), out of which, 331 reactions belong to all of the\n36 network cores (Figure 3b ). The high degree of similarity\nbetween the reaction cores of the various cyanobacteria species reflects the current\nknowledge on common metabolic processes across these species, obtained mostly from\nsequence comparisons. These shared core reactions belong to highly conserved\nmetabolic pathways, such as glycolysis, gluconeogenesis, and the TCA cycle among\nothers. MIRAGE's predictions extend these networks in a species-specific manner, with\nmany reactions predicted to belong to a small number of species (Figure 3b ). These species-specific reactions belong to more peripheral\npathways, for example, diterpenoid biosynthesis, fluorene degradation and others. Figure 3 Statistics on MIRAGE's reconstructed cyanobacteria metabolic networks .\n (a) The number of core reactions (blue) and predicted gap-filling\nreactions (red) in the various reconstructed cyanobacteria models. (b) A\nhistogram of core reactions (blue) and predicted gap-filling reactions (red)\nthat participate in different numbers of reconstructed cyanobacteria models. As\nshown, cyanobacteria network cores consist of many reactions that are known to\nexist in all 36 species, while many of the predicted gap-filling reactions are\nspecies-specific. To evaluate the performance of MIRAGE in reconstructing cyanobacteria models, we\ncompared a reconstructed network model for Synechocystis sp. PCC 6803 with\nthe manually curated models of Knoop et al. [ 50 ] and iSyn811 [ 51 , 52 ]. In this case, MIRAGE was applied to reconstruct a Synechocystis\n model by further utilizing gene expression data obtained from Tu et al. [ 53 ] as part of the reconstruction process (Materials and methods). The\ncomparison shows a predictive precision of 70% and recall of 24.6% for the Knoop\n et al. model [ 50 ] and precision of 37.5% and recall of 45% for iSyn811 [ 51 , 52 ]. These results are significantly better than random sampling\n(hyper-geometric P -values are 2.99 × 10 -27 and 3.59 ×\n10 -31 for Knoop et al .'s model and iSyn811, respectively).\nAgain, we find that the predictive performance of either the functional genomics data\nor the flux analysis alone is far worse (Figure 4 ). A\ncomparison with the predictive performance of Model SEED was not possible in this\ncase, as the SEED algorithm was not applied to reconstruct cyanobacteria models\n(focusing only on well-studied and annotated genomes). As a further evaluation\ncriterion, we performed a BLAST [ 54 ] search of the known enzyme sequences catalyzing the predicted gap-filling\nreactions in other species against the genomes of the corresponding cyanobacteria.\nReassuringly, we found that the resulting BLAST E- scores show significantly\nhigher sequence similarity for the set of predicted reactions in comparison to a\nrandom set of reactions ( t- test of 1.04 × 10 -74 ). Moreover,\n20.3% of predicted reactions showed E-values below 10 -100 , compared to\n9.7% of randomly sampled reactions, testifying the overall correctness of the\npredicted set of reactions. Figure 4 MIRAGE's predictive performance on reconstructing a known metabolic network\nof the cyanobacteria Synechocystis sp . PCC6803. (a)\n Metabolic network after Knoop et al. [ 50 ]; (b) metabolic network after Montagud et al. [ 51 , 52 ]. The precision and recall of MIRAGE is marked with a star symbol.\nThe precision and recall of a variant of MIRAGE that does not utilize\nfunctional-genomics data is marked with a circle. The predictive performance of\nthe functional genomic data (without metabolic flux analysis) is shown by the\npurple line. The performance of random predictions of gap-filling reactions is\ncolored blue. As a further evaluation of our reconstructed Synechocystis model, we applied\nit to predict gene knockout lethality data provided by [ 50 ]. We find that the prediction performance of our model is comparable with\nthat of Knoop et al. (Supp. Table 3 in Additional file 1): out of 39 genes\nknown to be non-essential, Knoop et al. correctly predicted 35, while our\nmodel correctly predicts 38. Out of 11 known essential genes, Knoop et al.\n correctly predicted 7, while our model correctly predicts 6. The fact that our\nautomatically generated model reaches a similar level of prediction performance to\nthat of a manually curated model demonstrates the applicability and importance of our\nmodel reconstruction approach. Utilizing the reconstructed cyanobacteria networks for metabolic engineering To demonstrate the applicability of the reconstructed cyanobacteria networks, we\napplied a computational metabolic engineering approach called Optknock [ 55 ] on these networks to rationally design genetic modifications that would\nincrease the production of astaxanthin, which is a powerful antioxidant belonging to\nthe carotenoid family. These metabolites are known to be produced by various\ncyanobacteria [ 56 , 57 ]. Optknock works by searching for gene knockouts that would couple the\nmaximal production and secretion of a molecule of interest with a naturally selected\ntrait of maximizing growth rate. Notably, 24 of the original core networks extracted\nfrom KEGG include an astaxanthin production reaction, though only 8 of these are not\ndead-end. In contrast, 25 of the network models reconstructed by MIRAGE have a\nfunctional astaxanthin production pathway, amenable for Optknock analysis. The application of Optknock for astaxanthin production identified double gene\nknockouts in 15 species that are expected to lead to astaxanthin secretion (Table\n 1 ). For 12 out of these, Optknock predicts the knockout of\ndimethylallyl-diphosphate: isopentenyl-diphosphate dimethylallyl transtransferase\n(EC: 2.5.1.1), which consumes an essential precursor for astaxanthin biosynthesis\n(1-hydroxy-2-methyl-2-butenyl4-diphosphate). The maximal achievable astaxanthin\nproduction rate reaches 2.18 μmol gDW -1 h -1 in\nProchlorococcus marinus 9601, representing a carbon utilization of 40% for\nastaxanthin production (considering a CO 2 uptake rate of 0.22 mmol\ngDW -1 h -1 [ 50 ]). This utilization of CO 2 to produce astaxanthin is predicted\nto reduce growth rate by 57% relative to the wild-type Prochlorococcus strain (Table\n 1 )."
} | 6,870 |
36425908 | PMC9679027 | pmc | 6,995 | {
"abstract": "Abstract A spatially explicit eco‐evolutionary model assembles simulated meta‐communities which are subjected to species and community perturbation experiments to determine factors affecting the stability of the global ecosystem. Spatial structure and resource variety increase the persistence of the ensembles against the removal of an individual species, yet they remain vulnerable to re‐invasion by an existing member of the meta‐community if it is introduced to all patches with minimal population. Optimal reserve placement strategies are identified for maximally preserving global biodiversity from the effects of sequences of patch disruption, and targeted reserve placement that shields the most or the rarest biodiversity is usually effective. However, if disturbed populations are permitted to re‐settle in neighboring patches, then reserves should also be situated remotely to isolate their residents from invasion.",
"conclusion": "8 CONCLUSION A set of diverse trophic meta‐communities have been assembled on a spatial lattice using an eco‐evolutionary model, with the local ecosystems corroborating the community‐level network and structural patterns previously observed in similar models. These complex co‐evolved meta‐communities are utilized as the basis for perturbation experiments at both the species and community level, with all results applicable only to arrangements of low‐range species in spatially heterogeneous environments. In agreement with existing theory, the combined processes of dispersal and the resulting availability of multiple resources enhance the meta‐community's persistence against the artificial loss of any single species with relatively few secondary extinctions. However, invasion of the entire spatial network can be destructive (up to 12.8% of the global biodiversity) even if it is by a species that formerly existed within the meta‐community and it is re‐introduced to all patches with minimal population. In this model, large bodysize species demonstrated a greater probability of successful patch invasion. However, this advantage vanished if the same species were introduced to a meta‐network consisting only of the resources. In either case, co‐evolved species showed an advantage when re‐invading the meta‐communities over a species whose traits were randomly generated. However, this could be compensated for by increased bodysize when invading mature communities. Habitat disruption due to human activity and development can be simulated by patch‐level disruption in a spatial meta‐network. Whether it is more damaging to displace the local populations of a patch than to eliminate them can be determined by measures of the ratio of at‐risk species within the patch and within neighboring patches where affected populations would be re‐settled. To reduce the ecological impact of anthropogenic habitat loss, optimal reserve placement is dependent on the mode of habitat disruption. Generally, reserves should dynamically target the highest‐diversity patches but if re‐colonization is possible then reserves that preserve the lowest average range species are also beneficial. If the populations of perturbed patches are displaced to neighboring patches, reserves should be situated in a remote region as a single large block to maximize the isolation of the interior from the invasion of displaced species. To investigate this effect further, future work would require a systematic study of reserves consisting of at least nine patches in larger spatial networks, so that the effect of isolating an interior patch in a 3 × 3 reserve block can be untangled from the effect of placing the reserve at the edge of accessible space. Other challenges remain to assemble model trophic meta‐communities of sufficient complexity, and in particular, work on eco‐evolutionary meta‐community models should employ a habitat mosaic with a small, fixed number of habitat types for the patches. This would facilitate a more realistic variable range of the species, allow for the modeling of real‐world geographies and testing the impact of habitat autocorrelation in space, systematic changes in habitat type, corridor placement, and for direct comparison with existing studies on thresholds of habitat loss (Pillai et al., 2011 ; Yin et al., 2017 ) for community collapse. We hope that future efforts will be able to address these and provide stronger recommendations for the principles of conservation of biodiversity and, in particular, the practical question of ideal nature reserve placement."
} | 1,130 |
25424444 | PMC4601276 | pmc | 6,997 | {
"abstract": "F usarium oxysporum has been reported as being able to both produce the enzymes necessary to degrade lignocellulosic biomass to sugars and also ferment the monosaccharides to ethanol under anaerobic or microaerobic conditions. However, in order to become an economically feasible alternative to other ethanol-producing microorganisms, a better understanding of its physiology, metabolic pathways, and bottlenecks is required, together with an improvement in its efficiency and robustness. In this report, we describe the challenges for the future and give additional justification for our recent publication."
} | 152 |
25914603 | null | s2 | 6,998 | {
"abstract": "Previous syntheses on the effects of environmental conditions on the outcome of plant-plant interactions summarize results from pairwise studies. However, the upscaling to the community-level of such studies is problematic because of the existence of multiple species assemblages and species-specific responses to both the environmental conditions and the presence of neighbors. We conducted the first global synthesis of community-level studies from harsh environments, which included data from 71 alpine and 137 dryland communities. Here we: i) test how important are facilitative interactions as a driver of community structure, ii) evaluate whether the frequency of positive plant-plant interactions across differing environmental conditions and habitats is predictable, and iii) assess whether thresholds in the response of plant-plant interactions to environmental gradients exists between \"moderate\" and \"extreme\" stress levels. We also used those community-level studies performed across gradients of at least three points to evaluate how the average environmental conditions, the length of the gradient studied, and the number of points sampled across such gradient affect the form and strength of the facilitation-environment relationship. Over 25% of the species present were more spatially associated to nurse plants than expected by chance in both alpine and dryland areas, illustrating the high importance of positive plant-plant interactions for the maintenance of plant diversity. Facilitative interactions were more frequent, and more related to environmental conditions, in alpine than in dryland areas, perhaps because drylands are generally characterized by a larger variety of environmental stress factors and plant functional traits. The frequency of facilitative interactions in alpine communities peaked at 1000 mm of annual rainfall, and globally decreased with elevation. The frequency of positive interactions in dryland communities decreased globally with water scarcity or temperature annual range. Positive facilitation-drought stress relationships are more likely in shorter regional gradients, but these relationships are obscured in regions with a greater species turnover or with complex environmental gradients. By showing the different climatic drivers and behaviors of plant-plant interactions in dryland and alpine areas, our results will improve predictions regarding the effect of facilitation on the assembly of plant communities and their response to changes in environmental conditions."
} | 632 |
34792119 | PMC8684450 | pmc | 6,999 | {
"abstract": "ABSTRACT A European transect was established, ranging from Sweden to the Azores, to determine the relative influence of geographic factors and agricultural small-scale management on the grassland soil microbiome. Within each of five countries (factor ‘Country’), which maximized a range of geographic factors, two differing growth condition regions (factor ‘GCR’) were selected: a favorable region with conditions allowing for high plant biomass production and a contrasting less favorable region with a markedly lower potential. Within each region, grasslands of contrasting management intensities (factor ‘MI’) were defined: intensive and extensive, from which soil samples were collected. Across the transect, ‘MI’ was a strong differentiator of fungal community structure, having a comparable effect to continental scale geographic factors (‘Country’). ‘MI’ was also a highly significant driver of bacterial community structure, but ‘Country’ was clearly the stronger driver. For both, ‘GCR’ was the weakest driver. Also at the regional level, strong effects of MI occurred on various measures of the soil microbiome (i.e. OTU richness, management-associated indicator OTUs), though the effects were largely regional-specific. Our results illustrate the decisive influence of grassland MI on soil microbial community structure, over both regional and continental scales, and, thus, highlight the importance of preserving rare extensive grasslands.",
"conclusion": "CONCLUSION Our results demonstrate the surprisingly strong effect that small-scale grassland MI has on microbial community structure. For fungi, it even was a comparable driver to geographic factors at the continental scale. While geographic factors had a similar effect size on both fungal and bacterial community structure, the effect of ‘MI’ was far stronger on the former. Importantly, these results establish that EXT grasslands harbor distinct fungal and bacterial communities in all examined regions, across a divergent climatic gradient. Despite the MI effects on individual microbial taxa being strongly regionally specific, some microbial indicators for grassland MI were detected at the transect level (e.g. the fungal genera Clavaria and Leohumicola ). This study demonstrates the ecological importance of the rare EXT grasslands as a distinct habitat for the soil microbiota, providing additional evidence of their capacity for biodiversity promotion in agroecological programs.",
"introduction": "INTRODUCTION Relatively, more studies in the existing literature have examined the effect of environmental factors (e.g. annual temperature and precipitation, soil pH) in driving the structure of the grassland soil microbiome across large spatial/ continental scales (e.g. Chen et al . 2015 ; Bahram et al . 2018 ; Xue et al . 2018 ; Plassart et al . 2019 ). In contrast, studies looking into the effects of grassland management intensity over comparable scales are currently missing. This is a pertinent research gap, particularly in light of the sharp decline in permanent, extensively managed grasslands over the past century. Before the green revolution, traditional grassland farming with low levels of farmyard manure application and low numbers of annual utilizations created such grasslands and represent some the most biodiverse habitats in temperate Europe (Wesche et al . 2012 ). They can contain a species-rich, compositionally complex plant community (Peter et al . 2009 ; van Dobben et al . 2017 ). Land use intensification represents one of the principle drivers of global change (Erb et al . 2017 ), and extensively managed grasslands are threatened at both ends of the management intensity spectrum, through intensifcation and abandonment. Over the course of the 20th century, the intensification of European grasslands has resulted in their plant communities becoming comparatively simple, with a greatly reduced species richness (Wesche et al . 2012 ). This was achieved through high nutrient fertilizer inputs, greatly increased number of annual utilizations (i.e. cutting and/or grazing events) and even through the sowing of grass monocultures or two to four-species mixtures. This has resulted in a multitrophic homogenization of grasslands communities, including soil microbial communities (Gossner et al . 2016 ). On the other hand, abandonment of extensive grasslands threatens what remains of this grassland type (through reforestation) and its associated ecological value (Le Clec'h et al . 2019 ; Zehnder et al . 2020 ). This has now resulted in extensive, species rich grasslands being one of the most endangered habitats in Europe, with the need of their conservation representing an urgent societal challenge (Habel et al . 2013 ). Promoting and preserving this grassland type, and the traditional farming practices which preserve them, within the agricultural landscape is now a major aim of the common agricultural policy (CAP) and habitats directive (2014–2020) of the European Union (EU). The primary aim of extensive grasslands under such schemes is to reduce nutrient fertilizer application, while concomitantly enhancing their capacity for biodiversity promotion (Kampmann et al . 2012 ). The decline of extensively managed grasslands has important implications for soil microbial biodiversity patterns, as grassland management intensity has been shown to be a strong determinant of soil microbial communities (de Vries et al . 2012b ; Meyer et al . 2013 ; Sayer et al . 2013 ; Szukics et al . 2019 ). Due to very limited nutrient fertilizer inputs, extensively managed grasslands are typically considered nutrient-poor, with such environments being favorable to soil fungi (van der Heijden, Bardgett and van Straalen 2008 ). Indeed, management practices such as reduced fertilizer application, followed by reduced grazing and cutting events, have been shown to promote the abundance of fungal biomass (de Vries et al . 2007 ). By contrast, increased grassland management intensity has been shown to promote bacterially dominated soil microbial communities (Xun et al . 2018 ). This difference may be evoked by distinct mechanisms, such as differences in the quantity and quality of plant-derived substrates (i.e. plants residues and root exudates) entering the soil matrix due to contrasting plant community structures (Fox, Lüscher and Widmer 2020 ). Additionally, plant traits associated with intensively managed grasslands, e.g. high growth rate and competitive ability for nutrients, have been shown to promote bacterial abundance in grasslands (Orwin et al . 2010 ). The majority of reported studies into the effects of grassland management intensity on the soil microbiome are at the experimental field or plot scale, while few have examined the influence of grassland management intensity at the regional or continental scale (de Vries et al . 2012b ; Szukics et al . 2019 ; Felipe-Lucia et al . 2020 ; Andrade-Linares et al . 2021 ). Thus, whether management intensity or geographic pattern (and the associated differences in pedo-climatic conditions which determine growth) is the stronger driver of grassland soil microbial community structure remains unclear. Microbial diversity in soil has been shown to strongly exhibit geographic patterns (Martiny et al . 2006 ), with environmental factors being shown to be the strongest driver of bacterial community structure at the landscape scale (Mayerhofer et al . 2021 ). In contrast, soil bacterial diversity has been shown to be largely independent from geography, but highly influenced by plant diversity, a major differentiator between intensively and extensively managed grasslands (Fierer and Jackson 2006 ). While, in vineyard sites, it was shown that land use had a strong influence on fungal community structure, while geography was the stronger driver of the bacterial community (Coller et al . 2019 ). Additionally, a more comprehensive understanding of the effect of grassland management intensity on the taxonomic and functional composition of the soil microbiome is also needed. Specific fungal taxa, which are known to be associated with ‘unimproved’ European grasslands include the Hygrocybe , Entoloma , Clavaria and Geoglossum (McHugh et al . 2001 ). More broadly, high soil nitrogen (N) content has been shown to suppress soil fungi (Fox, Lüscher and Widmer 2020 ). The abundance of bacterial and archeal genes involved in key steps in the soil N cycle, fixation, nitrification and denitrification have been shown to be altered by grassland management intensity (Meyer et al . 2013 ). Additionally, temperate European intensive grasslands typically receive fertilizer inputs in the form of slurry and manure, potentially introducing exogenous microorganisms into the soil matrix, such as bacteria of the phyla Firmicutes and Bacteriodetes (Abubaker et al . 2013 ). Previous studies have examined various aspects of grassland soil biology at the European scale, especially comparing it with both arable and forest soils, as well as the associated environmental drivers (Bouffaud et al . 2016 ; Plassart et al . 2019 ), while studies comparing different grassland management types are missing. This study aimed to compare the effect size of grassland management intensity (a composite term comprising many factors) against the benchmark of environmental factors (also a composite term). To achieve this, our sampling strategy attempted to maximize the environmental conditions and management practices employed. The study did not aim to disentangle the variation explained by individual factors. The employed study design is not ideal for partitioning the effects of individual factors, given how regionally specific grassland management is in Europe (Gilhaus et al . 2017 ). Our study specifically focused on the influence of grassland management intensity on both fungal and bacterial community structure across both a continental and regional scale. Specifically, the study had two research hypotheses: Geographic factors (i.e. location and environmental variables), and not management intensity, are stronger determinants of both soil fungal and bacterial community structures in grasslands at the continental scale. As differing management regimes will exist across the different transect regions, we hypothesize that there will be inconsistent effects of grassland management intensity across many regions of the continental transect.",
"discussion": "DISCUSSION Management had a strong influence on microbial community structure at the European and local scale Contrary to our hypothesis 1, ‘MI’ had a similar effect size as ‘Country’ on fungal community structure across the European transect. Remarkably, this result indicates that small-scale, regionally localized management practices, which have previously been shown to effect soil fungal communities (de Vries et al . 2007 , 2012b ), have a comparable effect size on fungal community structure (mean centroid distance = 0.415) as biogeographically diverse, continental-scale geographic factors (mean centroid distance = 0.429). An important point to consider is that the INT and EXT grasslands in each region were sampled comparatively close to each other, while the sites in each of the regions and countries were sampled far, and extremely far, from each other, respectively. If such short distances (resulting in comparable environmental growth conditions) would influence soil fungal community structures, this would strongly reduce the magnitude of the management effect observed, compared to both region and country. Thus, the effect size of management on soil fungal community structure seen in this study is quite impressive. There was also a significant effect of ‘MI’ on soil bacterial community structure at the transect level, though this effect was not as strong a driver as ‘Country’. Effects of individual management variables/ practices have been studied under controlled conditions. For example, N and P mineral fertilizer addition has been shown to influence fungal and bacterial community composition over large spatial scales under experimental field conditions (Leff et al . 2015 ). Our study shows that effects of grassland management on soil microbial community structure are also observed in ‘on-farm’ conditions at the European scale. These changes in the soil microbial community structure as a result of MI are most likely the result of dissimilarities in the soil environment, specifically nutrient availability, and differing plant community structure and diversity in contrastingly managed grasslands (Orwin et al . 2010 ; de Vries et al . 2012a ). Under field conditions, it is not possible to separate these compounding influences as they always act concertedly, though both, and their interactions, may contribute to the observed effects. The strong effect size of MI on the soil microbiome was even more impressive given the fact we did sample the management intensity range of agricultural land and excluded nature conservation areas. This highlights the ecological value of the rare EXT grasslands in the agricultural landscape. In each country, the factor ‘MI’ had a stronger impact than ‘GCR’ on fungal community structure, which indicates that as differences among geographic patterns decrease, MI becomes the dominant influencing factor, confirming the continental scale finding at the regional scale. This is especially impressive in Sweden and Switzerland, given the large differences in environmental variables between their FR (e.g. mean annual temperature = 7.48°C and 9.84°C, respectively) and LFR (3.15°C and 5.82°C, respectively). The strong effect of management was also confirmed by the fact that ‘MI’ had a significant effect on fungal and bacterial community structure in nine and eight of the 10 regions, respectively (Table 4 and Fig. 3 ). Here, the effect of ‘MI’ was tested separately in these regions, thus the environmental conditions of the other regions could not have had an effect. Indeed, the effect of ‘MI’ was so strong at the regional level, that significance was reached even with the greatly reduced number of sites. The causes of the MI effect on soil microbial community structure, e.g. differences in soil nutrient availability and consequentially plant community structure may be especially relevant in Switzerland, as it likely reflects the implementation of an agroecological scheme called ‘ecological compensation areas’, of which all sampled EXT sites in Switzerland were part of. Introduced already in 1993, the scheme aims to promote and protect biodiversity in agricultural areas, of which EXT grasslands are the dominant type (BLW 1998 ). The scheme has been shown to promote and protect plant biodiversity within the agricultural landscape, particularly in mountain regions (Peter et al . 2008 ; Kampmann et al . 2012 ; Ravetto Enri et al . 2020 ), and to have positive effects on insects (Albrecht et al . 2010 ). The results presented here demonstrate that they additionally harbor distinct fungal and bacterial community structures. This is most likely on account of them never having been intensified, as even short-term additions of soil nutrients to grassland can influence both the plant and soil microbial community structure for decades afterwards (Spiegelberger et al . 2006 ). Even though the sampling design was optimized to sample INT and EXT under comparable environmental conditions by sampling nearby plots within small regions, the effect of MI was accompanied by significant differences in environmental variables between INT and EXT in five regions of the transect, the LFR region of Germany, Switzerland and Portugal, as well as both GCR of Azores. This is a consequence of farmers choosing a management scheme dependent on the differing abiotic conditions within their holding (Peter et al . 2008 ; Gilhaus et al . 2017 ). For example, in mountain regions (i.e. LFR Germany and Switzerland), the sharp difference in the site inclination between management types is reflective of EXT typically being on valley slopes, while INT are located on more easily accessible plots where mechanization is easier (Kampmann et al . 2008 ). Even these small, indirect MI effects could influence soil microbial community structure in grasslands, though we suspect that the direct effects of MI (e.g. fertilizer application) are more important. All the above-mentioned regions had higher P tot in INT, suggesting that MI did directly influence the observed shift in microbial community structure in these regions, as P fertilization has been shown to do (Ikoyi, Fowler and Schmalenberger 2018 ; Ikoyi et al . 2020 ). Moreover, in four of the five remaining regions (i.e. FR of Sweden, Switzerland and Portugal, LFR of Sweden) there was a significant effect of MI on both fungal and bacterial community structure, even without the aforementioned indirect effects. Geographic factors comparably influence both fungal and bacterial community structure The effect of the factor ‘Country’ was comparable for both fungal and bacterial community structure (i.e. mean centroid distance = 0.429 and 0.388; √CV = 0.247 and 0.249, respectively). The clear contrast was the effect size of ‘MI’ (as discussed above), which was much greater on fungal than bacterial community structure (i.e. mean centroid distance = 0.415 and 0.289; √CV = 0.233 and 0.174, respectively). This would indicate that soil fungal communities are more strongly influenced by the availability of soil nutrient resources (Bahram et al . 2018 ; Fox, Lüscher and Widmer 2020 ), and that the validity of hypothesis 1 depends on the component of the soil microbiome considered. Large differences in precipitation, as present across our transect (590–1690 mm), has previously been shown to be a strong determinant of both fungal and bacterial community structure in a large scale study of Mongolian grasslands (where precipitation ranged 104–412 mm, Chen et al . 2015 ). Other climatic factors, particularly temperature, has been shown to have a strong influence on soil fungal and bacterial diversity of other land use types, such as forests (Zhou et al . 2016 ). Particularly notable in our transect was the strong separation of the microbiome structures of the southern countries, Portugal and Azores, both from each other and from that of the other countries. Both Portugal and Azores represent quite different biogeographic regions, Mediterranean and Macaronesian respectively, from the other countries of the transect (Alpine, Boreal and Continental). Additionally, the isolation of Azores from mainland Europe (∼1190 km from the coast of Portugal), could be contributing to the Azores–Portugal difference. This continental scale study not only generated a range of climatic conditions, but also of soil physicochemical properties (i.e. % soil sand in Portugal) that may further differentiate the microbial community structure between countries. Soil physicochemical properties have been shown to be strong drivers of soil fungal and bacterial communities in large scale studies (Chen et al . 2015 ; Bahram et al . 2018 ). Along a European transect, which included most of the countries studied here, it was shown that spatial variations in bacterial community structure and diversity were principally driven by soil physicochemical properties (Plassart et al . 2019 ). We included GCR within countries as part of our transect to further expand the range of sampled environmental conditions and geographic patterns. The GCR within Sweden (centroid distance 1.95) and Switzerland (centroid distance 2.46) were especially biogeographically divergent, with the factor ‘GCR’ significantly determining microbial community structure in these two countries. The geographic distance between the FR (Continental) and LFR (Boreal) in Sweden was the largest within country distance of the transect at ∼1200 km. This distance is comparable to, or greater than, the distance between different countries of the transect, e.g. Portugal–Azores. This latitudinal distinction in Sweden was previously shown to influence the fungal community structure in Scots pine needles (Millberg, Boberg and Stenlid 2015 ). In Switzerland, the FR (Continental, ∼487 m) and LFR (mountain, ∼1266 m) differ sharply in altitude, and consequently their climate, and have previously been shown to differ in their grassland plant communities (Güsewell, Peter and Birrer 2012 ). There were much smaller and statistically not significant differences in the climatic variables in the GCR regions of the southern countries of Portugal (centroid distance 0.76) and Azores (centroid distance 0.18), which may not have been large enough to shape the soil microbiome among GCR. Strong regional specificity in differences of the soil microbial taxa to management intensity Strong regional specificity was seen in the particular response of the microbiome (i.e. OTU richness, abundance of fungal and bacterial phyla, management-associated indicator OTUs) to grassland MI, which is contrary to the second hypothesis of this study. This is in line with the studies second hypothesis and reflects on-farm conditions and the fact that optimal grassland management is adapted to suit regional conditions (Gilhaus et al . 2017 ). Thus the INT management can consequently address completely different plant limiting conditions, as was the case between Switzerland (N fertilization) and Portugal (irrigation), for example. Furthermore, there could be different regional interactions of grassland MI with soil type and physicochemical characteristics (Xue et al . 2018 ), climate (Chen et al . 2015 ; Bahram et al . 2018 ) and plant community diversity and composition (Gilhaus et al . 2017 ), as all differ among countries. In four regions of the transect fungal OTU richness was significantly higher in EXT. This finding supports previous studies, which indicated that EXT grasslands promote fungal dominated soil microbial communities (de Vries et al . 2007 , 2012b ). However, the opposite trend was observed in three other regions in the southern part of the transect (i.e. FR of Portugal and Azores, LFR of Azores) and may represent a different interaction effect between the southern regions and MI to that observed in Sweden, Germany and Switzerland. This, along with the effect of MI on soil fungal biomass, warrants further research. There was also a lack of consistency in the differences in the abundance of fungal and bacterial phyla, with no single phyla having a uniform response to MI in all regions. Though a small number did respond in multiple regions, such as the Firmicutes, whose tendency for higher relative abundance in INT grasslands may reflect slurry or manure application (Abubaker et al . 2013 ). There were many (in most cases hundreds) of indicator fOTUs and bOTUs for MI detected in each region. The extent of the lack of sharing of these MI-associated fOTU and bOTU was surprising, even being low between regions within the same country. This would indicate that abiotic conditions are strong determinants for the specific taxonomic differences of the soil microbiome to MI (Ma et al . 2018 ). This finding has important implications for soil biodiversity monitoring. Due to the strong regional-specificity of microbial indicators to grassland MI, monitoring efforts should focus on identifying and protecting EXT grasslands harboring regionally unique/rare microbial taxa. In addition, future studies should consider the temporal development of soil microbial diversity in differently managed grasslands, both within a growing season and across years, as this is a rarely considered aspect in soil monitoring schemes (van Leeuwen et al . 2017 ). Despite the strong regionally specific response, a number of microbial indicators for grassland MI were identified over the whole European transect. Indicator fungal genera for EXT grasslands included the Clavaria (fOTU_647 and fOTU_636, found in 74 and 66 of the 108 EXT sites, respectively) and Leohumicola (fOTU_9761 and fOTU_395, found in 76 and 64 of the 108 EXT sites, respectively). Clavaria is a saprotrophic genus associated with ‘unimproved’ European grasslands with low available sulfur and phosphate, and is highly sensitive to the application of nutrients (McHugh et al . 2001 ). The Leohumicola genus has been identified as an ericoid mycorrhiza. The symbiosis of ericoid mycorrhiza with the Ericaceae family of plants has been shown to help them adapt to acidic and/or nutrient-poor conditions (Cairney and Meharg 2003 ; Hambleton, Nickerson and Seifert 2005 ). Fungal indicator genera for INT included Thelebolus (fOTU_13, found in 87 of the 115 INT sites) and Podospora (fOTU_106, found in 78 of the 115 INT sites), both have been reported to develop on dung samples (Richardson 2001 ), and likely benefits from dung application. All of the identified bacterial taxa were indicators for INT. One indicator (bOTU_18978, found on 78 of the 115 INT sites) was assigned to the genus Candidatus Udaeobacter within the Verrucomicrobia phylum. A representative of this genus, Ca. Udaeobacter copiosus was identified as a highly ubiquitous taxa, particularly in grasslands (Brewer et al . 2016 ). It was reported as having a relatively small genome (∼40% smaller than the mean bacterial genome size) and may be sacrificing metabolic versatility as a competitive strategy. This may be a particular advantage in INT grasslands, as they have greater soil nutrient availability and lability, compared to EXT grasslands."
} | 6,416 |
35745425 | PMC9230068 | pmc | 7,003 | {
"abstract": "This study develops the nanostructured superhydrophobic titanium-based materials using a combined preparation method of laser marking step and the subsequent anodizing step. The structural properties were determined using an X-ray diffractometer (XRD) and scanning electron microscope (SEM), while the performance was explored by wear and corrosion tests. The laser marking caused a rough surface with paralleled grooves and protrusions, revealing surface superhydrophobicity after organic modification. The anodizing process further created a titanium oxide (TiO 2 ) nanotube film. Both phase constituent characterization and surface elemental analysis prove the uniform nanofilm. The inert nanosized oxide film offers improved stability and superhydrophobicity. Compared to those samples only with the laser marking process, the TiO 2 nanotube film enhances the corrosion resistance and mechanical stability of surface superhydrophobicity. The proposed preparation pathway serves as a novel surface engineering technique to attain a nanostructured superhydrophobic surface with other desirable performance on titanium alloys, contributing to their scale-up applications in diverse fields.",
"conclusion": "4. Conclusions This study manufactured a superhydrophobic surface with enhanced properties on titanium alloys. A combined preparation method of laser marking and anodizing is developed to construct favored surface nanostructures. After the anodizing process, a thin, inert nanotube film grows on the laser-marked surface and provides additional superhydrophobicity, revealing a high WCA at 158°. This anodizing step also gives the sample better corrosion resistance due to its better superhydrophobicity and increased surface uniformity, showing a stable WCA of ~155° throughout the corrosion tests. Besides, the well-designed wear test shows the superior abrasion resistance of the sample after laser marking and anodizing. The inert nanotube film protects the surface features well, and the sample presents good hydrophobicity with a WCA of 147° after the wear tests.",
"introduction": "1. Introduction Superhydrophobic surfaces extensively exist in nature, including lotus leaf, butterfly wing, and sharkskin [ 1 , 2 ]. Superhydrophobicity generally refers to a water-repellent surface with a water contact angle (WCA) greater than 150°. Such surface features correlate to some unique properties such as self-cleaning [ 3 , 4 ], oil-water separation [ 5 , 6 ], and anti-icing and anti-fog properties [ 7 , 8 ]. Surface roughness and surface free energy are two vital factors determining surface superhydrophobicity [ 9 ]. The construction of superhydrophobic surfaces on alloy surfaces generally requires two steps: generating a rough surface structure, and the following chemical modification to reduce the surface energy. So far, many previous reports have investigated the superhydrophobic surfaces on metals and their alloys [ 10 ]. For instance, Liu et al. [ 11 ] prepared a porous microstructure on AZ31 magnesium alloy using the micro-arc oxidation method, and then modified the samples using a stearic acid ethanol solution to attain superhydrophobicity. Yin et al. [ 12 ] combined electrochemical deposition and hydrothermal treatment to construct a nanorod-like superhydrophobic coating on stainless steel. The prepared coating possessed good self-cleaning properties and could degrade organic molecules under ultraviolet radiation, thereby preventing the damage from organic pollutants in practical conditions [ 12 ]. Titanium alloy shows high specific strength, good heat resistance, great mechanical stability, and corrosion resistance [ 13 , 14 , 15 ]. In general, constructing a superhydrophobic surface on the light alloys augments the coating corrosion resistance and, more importantly, provides unique functions that expand their application fields. There are some developed methods to prepare superhydrophobic surfaces on titanium alloys, such as electrochemical deposition [ 16 , 17 ], spraying [ 18 ], hydrothermal processing [ 19 , 20 ], anodic oxidation [ 21 ], and micro-arc oxidation [ 22 ]. For example, He et al. [ 17 ] deposited a dendritic Zn/ZnO/TiO 2 film on titanium alloy by electrochemical deposition, whereas the surface superhydrophobicity can be transformed into superhydrophilicity by ultraviolet irradiation. In another report, Sun et al. [ 23 ] used dual chemical treatments to prepare superhydrophobic surfaces on different light alloy materials. However, most superhydrophobic surfaces prepared on titanium alloy suffer from their weak mechanical performance, which gradually degrades the superhydrophobicity in a long-term service life [ 24 ]. Laser treatment is a facile, fast, and ecologically friendly technology [ 25 ]. Laser marking prepares the superhydrophobic surface in recent reports of titanium alloy modification [ 26 , 27 ]. Pu et al. [ 28 ] used laser marking to build a superhydrophobic surface on the Ti-6Al-4V (TC4) titanium alloy surface, showing a good abrasion resistance during friction tests. Laser marking is easy to operate, fabricating the controllable micro/nano structure. However, the laser marking process could damage the coating uniformity and decrease surface features’ stability. In another report, micro/nano-structures transition on TC4 alloy surface was studied by sandblasting and anodic oxidation. The authors reported hydrophobic surface of the compact TiO 2 film, yet more tailoring work is necessary to improve the corrosion resistance and hydrophobicity of the nanotubular surface [ 29 ]. Therefore, the study of constructing a stable and superhydrophobic surface in titanium alloys is of great importance. The titanium products with desirable superhydrophobic properties promote their applications in multiple fields, especially in aerospace engineering where the aircraft and spacecraft search for lightweight alloy materials with favorable hydrophobic performance [ 30 , 31 ]. This study develops the nanostructured superhydrophobic titanium-based materials using a preparation method combining laser marking and anodizing. The anodizing process produces a thin, uniform nanotube film on the laser-marked surface, thereby delivering enhanced superhydrophobicity, corrosion resistance, and abrasion resistance. We propose an easy and low-cost method which allows for hydrophobic modification of the surface of titanium alloy products.",
"discussion": "3. Results and Discussion 3.1. Microstructure and Phase Constituent Figure 2 presents the surface microstructures observed on the laser-marked sample. After the laser treatment on TC4 alloys, the uniform surface structure consisting of grooves and protrusions is generated as depicted in Figure 2 a,b. During the laser marking process, the titanium alloy undergoes the melting/solidification process due to laser heating. The melted materials on the laser scanning path are solidified and piled up on both sides of the laser scanning route. Such rough morphology results from the grid-like laser routes during laser marking processing. The anodizing treatment constructs nano-structural features on the laser-marked surface. The anodizing treatment barely changes the microstructures of the laser-marked sample ( Figure 2 c), whereas numerous nano-sized pores are generated on the original surface as shown in Figure 2 d,e. The porous nanotube film is fabricated during the anodizing processing on the rough laser-marked surface [ 32 ]. The balance between the anodic oxidation and chemical etching grows the TiO 2 nanotube and generates the nano-pitted morphologies on the sample surface. Figure 2 f,g further reveals the 3D surface morphology and roughness recorded on these samples with or without anodizing step. Similar 3D surface images are observed on all samples. The distance between the surface peaks is ~50 µm, which is the same as the interval instance of the laser scanning route. The anodized sample shows a surface peak height at 40–60 µm, which is lower than the laser-marked sample’s peak height at 40–70 µm. In the anodizing step, the chemical etching process consumes a certain extent of surface materials, giving rise to these relatively smooth surface peaks on the anodized samples. To examine the phase variation in the laser marking and anodizing treatments, the phase constituents detected by XRD tests are shown in Figure 3 . All these samples display the same diffraction peaks at 35.09°, 38.42°, 40.17°, 53.004°, 62.94°, 70.66°, 76.21°, corresponding to the (100), (002), (101), (102), (110), (103), (112) crystal planes of Ti (PDF#44-1294) that originates from the TC4 alloy substrates. Compared to the untreated substrate, the XRD profile of the treated samples shows several peaks appearing at 31.46°, 45.40°, 52.29°, 62.72°, 71.21°, corresponding to the (111), (112), (220), (113), (041) planes of TiO 2 (PDF#21-1236), whereas the additional anodizing process barely changes the XRD profiles. This indicates that the surface oxidation reactions occur in both laser marking and anodizing treatments. In addition to TiO 2 , a small content of Al 2 O 3 is produced at the same time due to the oxidation of aluminum in the TC4 alloy. We should note the following FAS modification step causes no change on the phase constituents of all samples. X-ray photoelectron spectroscopy (XPS) was used to determine the sample surface before and after FAS modification, as depicted in Figure 4 . The enlarged Ti spectrum embedded in Figure 4 a proves the generation of TiO 2 in the prepared samples. It can be clearly seen that the additional F1s peaks appear on the sample after FAS modification, Figure 4 b. Additionally, the functional groups such as -CF 3 and -CF 2 appear in the C1s high-resolution spectrum in Figure 4 b, which indicates that the low free energy molecular chains containing fluorine have been attached to the sample surface. The surface superhydrophobicity depends on the low surface free energy. The fluorine-containing molecules is composed of polar end-groups and a long hydrophobic chain. After the FAS modification, the self-assembled low-energy groups of FAS on the surfaces attain surface superhydrophobicity for these samples [ 33 ]. 3.2. Wettability and Corrosion Behavior The water contact angle (WCA) is detected for these sample surfaces to characterize the wettability, as presented in Figure 5 . Generally, the surface is hydrophobic when WCA is larger than 90°, whereas a superhydrophobic surface corresponds to a WCA value greater than 150°. The WCA on the samples at different processing stages is compared in Figure 5 a. The WCA on the untreated TC4 alloy surface is 72°, corresponding to a hydrophilic surface. After laser marking treatment, the surface WCA decreases to 0°, revealing a superhydrophilic surface. Additionally, the WCA of the sample after the further anodizing step remains to be 0°. The Wenzel theory explains the above phenomena: the hydrophilicity of the hydrophilic surface increases along with the increases in surface roughness [ 34 ]. The laser marking treatment constructs a rough microstructure on the TC4 alloy surface, greatly increasing the surface roughness and reducing the WCA value. The use of FAS modification reduces the surface free energy and provides superhydrophobicity. After FAS modification, the WCA is 103° for the untreated sample, transforming the original hydrophilic surface into a hydrophobic surface. The sample after laser marking attains the superhydrophobic surfaces with the WCA of 154°, and the further anodizing step increases the WCA to be 158°. We should note the fabricated titanium-based surface reveals a higher WCA than the laser-treated sample and anodized sample in earlier reports, indicating its improved superhydrophobicity awarded from the combined process [ 28 , 29 ]. Figure 5 b shows the state of a water droplet on the modified surface after laser marking and anodizing. The water droplet remains spherical since the sample surface is water-repellent. At the same time, TC4 alloy substrate becomes grey after laser marking and anodizing, due to the surface oxidation reactions. The superhydrophobic sample is also immersed in water, as depicted in Figure 5 c. Even though the sample surface is lower than the water surface, water cannot diffuse over the superhydrophobic surface. This further proves the strong water repellency of the prepared superhydrophobic surface. The WCA value largely depends on the wetting state on the surface. The most accepted Cassie theory describes the surface status using the equation below [ 34 ]: cosθ c = f 1 (cosθ 1 + 1) − 1 (1) \nwhere θ c is the apparent contact angle of the droplet, θ 1 is the intrinsic contact angle of the liquid on the solid surface, and f 1 is the ratio of the liquid-solid contact area to the total contact area. In this case, θ 1 is a constant value since the intrinsic contact angle refers to the contact angle of the liquid drop on an ideal smooth solid surface. After laser marking, the surface generates microstructures where numerous protrusions could support the droplets and prevent the droplets from entering the surface grooves. This produces a large number of air pockets between the droplets and the grooves on the surface. Therefore, the value of f 1 decreases and the contact angle θ c increases significantly to be 154°. When the anodizing treatment further processes the laser-marked surface, the rough microstructure on the surface will be covered with a thin layer of TiO 2 nanotubes. This increases the number of air pockets in each nanotube structure and further reduces the value of f 1 . Therefore, the additional anodizing treatment after laser marking can further improve the surface superhydrophobicity. The corrosion behavior poses great significance in industrial applications [ 35 , 36 ]. Figure 6 shows the Tafel polarization curves recorded on the samples at different processing stages, and the calculated parameters are listed accordingly in Table 1 . The corrosion potential (E corr ) of TC4 substrate is −0.243 V, and the corrosion current density is the largest among the tested samples. After laser marking treatment, the corrosion current density decreases due to the generation of an oxide film. The sample corrosion resistance is greatly enhanced when further applying anodizing treatment. An inert TiO 2 nanotube film is uniformly generated on the laser marked sample, protecting the underlying substrate from corrosion attack. FAS modification is used to modify the samples chemically, and the two modified superhydrophobic surfaces reveal a lower corrosion current density. Due to the awarded water repellency, the modified superhydrophobic surface could generate a lot of air pockets between the solution and the sample surface, thereby significantly reducing the contact area in the corrosion environment and increasing the corrosion resistance. 3.3. Superhydrophobic Surface Stability The superhydrophobic surface needs good stability to suit various fields in practical applications. The chemical stability of surface superhydrophobicity was tested by immersing the sample in 3.5 wt% NaCl solution for 12 h and recording the WCA every 2 h. Figure 7 shows that the WCA of the two superhydrophobic samples hardly decreases after 12 h, which is attributed to the excellent water repellency of the superhydrophobic surface. This work uses a linear abrasion test method to determine the mechanical stability of superhydrophobic surfaces, as shown in Figure 8 . The testing scheme is illustrated in Figure 8 a: the sample with 100 g weight load is pushed or pulled on the 600# SiC sandpaper at a constant speed. The superhydrophobic surface and the rough sandpaper are rubbed uniformly, aiming to determine the mechanical stability of sample superhydrophobicity. Figure 8 b presents the recorded WCA values for the superhydrophobic surfaces after different mechanical testing cycles. After five cycles, the laser-marked superhydrophobic surface maintains a contact angle of more than 150°. The WCA decreases to 141° when the testing distance reaches 2000 mm (i.e., 10 cycles), yet the surface still has good hydrophobic properties. The laser marking alters the TC4 surface into a morphology consisting of peaks and valleys. This uneven structure can accommodate the generated wear debris and reduce the abrasive wear, thereby improving the mechanical stability of surface features [ 27 ]. In addition, such a microstructure with evenly distributed peaks provides better mechanical performance, which helps maintain the superhydrophobic surface [ 10 ]. The anodizing step further improves the mechanical stability of surface superhydrophobicity. The surface maintains superhydrophobicity after 6 testing cycles, and the contact angle of the sample becomes 147° when the testing distance reaches 2000 mm. The enhanced mechanical performance can be attributed to the formation of the inert TiO 2 nanotube layer on the laser-marked surface. This layer is well-generated and adequately covered on the sample surface. Such a layer can better store debris under wearing tests, and the inert nanotube structure is tough to peel off [ 37 ]. Furthermore, the unworn nanotube structure could provide extra air cushion effects for maintaining the surface superhydrophobicity."
} | 4,330 |
22298502 | null | s2 | 7,010 | {
"abstract": "The development of nanotechnology has been largely inspired by the biological world. The complex, but well-organized, living system hosts an array of molecular-sized machines responsible for information processing, structure building and, sometimes, movement. We present here a novel light-powered DNA mechanical device, which is reminiscent of cellular protein motors in nature, especially those of green plants. This walking device, which is based on pyrene- assisted photolysis of disulfide bonds, is capable of autonomous locomotion, with light control of initiation, termination and velocity. Based on DNA sequence design and such physical conditions as temperature and ionic strength, this photon-fueled DNA walker exhibits the type of operational freedom and mechanical speed that may rival protein motors in the future."
} | 206 |
21966446 | PMC3179486 | pmc | 7,011 | {
"abstract": "Environmental parameters drive phenotypic and genotypic frequency variations in microbial communities and thus control the extent and structure of microbial diversity. We tested the extent to which microbial community composition changes are controlled by shifting physiochemical properties within a hypersaline lagoon. We sequenced four sediment metagenomes from the Coorong, South Australia from samples which varied in salinity by 99 Practical Salinity Units (PSU), an order of magnitude in ammonia concentration and two orders of magnitude in microbial abundance. Despite the marked divergence in environmental parameters observed between samples, hierarchical clustering of taxonomic and metabolic profiles of these metagenomes showed striking similarity between the samples (>89%). Comparison of these profiles to those derived from a wide variety of publically available datasets demonstrated that the Coorong sediment metagenomes were similar to other sediment, soil, biofilm and microbial mat samples regardless of salinity (>85% similarity). Overall, clustering of solid substrate and water metagenomes into discrete similarity groups based on functional potential indicated that the dichotomy between water and solid matrices is a fundamental determinant of community microbial metabolism that is not masked by salinity, nutrient concentration or microbial abundance.",
"introduction": "Introduction Microbes numerically dominate the biosphere and play crucial roles in maintaining ecosystem function by driving chemical cycles and primary productivity [1] , [2] . They represent the largest reservoir of genetic diversity on Earth, with the number of microbial species inhabiting terrestrial and aquatic environments estimated to be at least in the millions [3] . However, the factors determining the spatiotemporal distributions of microbial species and genes in the environment are only vaguely understood, but are likely to include micro-scale to global-scale phenomena with different controlling elements. Microbial community structure is determined on varying scales by a complex combination of historical factors (e.g. dispersal limitation and past environmental conditions) [4] , the overall habitat characteristics [5] , the physical structure of the habitat (e.g. fluid or sediment) and by changes in current environmental parameters (e.g. salinity and pH) [6] – [9] . Understanding the relative importance of these different effectors is central to understanding the role of microbes in ecosystem function, and therefore to predicting how resident microbial communities will adapt to, for example, increasing salinity levels due to localized climate driven evaporation and reduced rainfall [10] . Physicochemical gradients provide natural model systems for investigating the influence of environmental variables on microbial community structure. In aquatic systems, salinity is a core factor influencing microbial distribution [6] , [11] and has been identified as the primary factor influencing the global spatial distribution of microbial taxa [6] . Salinity gradients occur in estuaries, solar salterns and ocean depth profiles. Evidence exists for increases in abundance and decreases in the diversity of microbial communities spanning salinity gradients [9] , [11] – [14] . This change is wrought by variance in the halo-tolerance of different taxa and the influence of salinity on nutrient concentrations [15] . We examined the resident microbial communities inhabiting sediment at four points along a continuous natural salinity gradient in the Coorong, a temperate coastal lagoon located at the mouth of the Murray River, South Australia. To determine the relative importance of salinity, nutrient status and microbial abundance in structuring microbial community composition and function, we used shotgun metagenomics to compare the taxonomic and metabolic profiles of our samples to representative metagenomes in public databases. Our results demonstrate that the taxonomic composition and metabolic potential of our metagenomes show a conserved signature, despite the microbes existing in disparate chemical environments. Comparison to other metagenomes indicates that this signature is determined by the substrate type (i.e. sediment) of the samples.",
"discussion": "Discussion Despite the strong environmental heterogeneity along the gradient studied here ( Table 1 ), taxonomic and metabolic profiles were conserved at the phyla and SEED hierarchy 1 level ( Figs. 1 & 2 ). This similarity was even more striking at finer levels of resolution. Coorong metagenome profiles were >89% and 89.5% similar in taxonomic and metabolic composition at the genome and subsystem level respectively ( Figs. 3 & 4 ). This indicates that the four microbial communities had similar structure, despite the intense environmental variability that occurred along the gradient. While the strong similarity between these samples, relative to other samples of comparable salinity, may to some extent be attributable to identical DNA extraction and sequencing procedures, biogeography and a shared environmental history between the samples, the clustering of our metagenomes with other solid substrate metagenomes for both taxonomic and metabolic profiles at >82% and >85% respectively, indicates that the signature of our profiles is largely determined by the substrate type of the samples (i.e. sediment). The metagenomes which show a high degree of similarity to our profiles are derived from a wide range of salinities, indicating that salinity is not the major structuring factor. Particularly evident is the close metabolic clustering of the four Coorong sediment metagenomes with other examples of marine sediment ( Fig. 4 ) despite these samples coming from a lower salinity than the Coorong sediment samples. This principle is highlighted by the observation that Coorong water samples of a similar salinity and identical geographic location ( Table S3 ) do not cluster with Coorong sediment samples in terms of taxonomy or metabolic potential, but rather cluster with other water samples. We interpret this as an indication that the substrate type (e.g. water vs solid substrate) is an important determinant of microbial functional composition that supersedes bulk environmental parameters (e.g. salinity) as the dominant structuring factor. This is further supported by the observation that the majority of metagenomes analyzed for metabolic potential cluster into two groups: a water group and a solid substrate group ( Fig. 4 ), regardless of salinity or geographic location. Whilst it has been shown that metagenomic profiles cluster into defined biome groups [5] , [22] , this is the first observation of such a clear dichotomy between water and solid substrate habitats which is not masked by salinity. Salinity has previously been identified as the primary factor governing the global distribution of prokaryotic 16S rRNA sequences [6] , [23] , [24] , [25] . Whilst Lozupone & Knight [6] identified substrate type (water vs sediment) as the second most important factor structuring microbial diversity after salinity, Tamames et al \n [24] concluded that salinity is more relevant than substrate type as sediment/soil and water from similar salinities clustered together in their analysis. These findings contradict the patterns apparent in our metabolic profile clustering ( Fig. 4 ) and indicate that the phylogenetic and metabolic aspects of microbial community diversity may be driven by different dominant factors. This also implies that accessing genetic information from the entire length of the genome as opposed to a specific taxonomic marker gene can yield different interpretations. This is potentially due to the influence of lateral gene transfer and a wider representation of taxa in 16S rDNA databases as opposed to genomic databases [26] , [27] . Whilst Coorong metagenomes clustered taxonomically with other solid substrate metagenomes ( Fig. 3 ), there was not a clear dichotomy between samples from water and solid substrate types as was observed for the metabolic profiles. This indicates that the substrate type may not be as important a controlling factor for taxonomy as it is for metabolism. That substrate type is a more important determinant of metabolic composition indicates that some genes, important for living in different substrate types, are shared by varying taxa adapted to different salinities. The samples that did not metabolically cluster within the two larger branches of ‘solid substrate’ and water ( Fig. 4 ) were typically derived from more extreme hypersaline environments, such as solar salterns [28] and a hypersaline mat [29] . This indicates that in some cases, salinity can be the major factor driving the metabolic profile grouping, probably in instances where salinity reaches a critical level, whereby it selects for less diversity and more dominant taxa. This is consistent with the salinity driven clustering of the saltern metagenomes when ordinated using di-nucleotide signatures [22] . The characteristics of particular substrate types that can select the metabolic content of the microbial community could be related to the differing degree of chemical heterogeneity in fluid and solid substrate habitats. Water is mixed to a higher degree than soil/sediment thus resulting in less physiochemical heterogeneity. Soil, sediment and biofilms are extremely heterogeneous resulting in the high degree of diversity commonly observed in these habitats compared to water substrates [3] , [6] . This differing division of resources and niches likely explains the dichotomous clustering of water and solid substrate metagenomes observed in our data. Additionally, in aquatic systems, sediment and benthic habitats are generally more anoxic than the overlying water suggesting that reduction and oxidation (REDOX) status is also a potentially important factor driving this split. Indeed, initial investigations indicate that a prevalence of virulence, motility and anaerobic respiration genes in solid substrate habitats drive the water versus solid substrate split (Jeffries et al , in prep). Our interpretation that the matrix from which the sample is derived is more important in determining the functional community structure than bulk physicochemical conditions has important implications for how we predict changes in microbial community function in the context of climate change driven increases in salinity levels or eutrophication associated with anthropogenic inputs. For example, the Coorong is currently undergoing a period of increasing salinity levels and eutrophication [30] , reflected in the gradient examined here. Our results suggest that, whilst small scale changes in gene abundance occur across this salinity gradient (for example regulation/signaling and metabolism of aromatic compounds; Fig. 2 ), the overall functional potential of the microbial community remains similar between salinities and demonstrates a high degree of similarity to lower salinity marine sediment at the subsystem level ( Fig. 4 ). This indicates that while shifts in the composition of the microbial community may occur following further shifts in salinity, the overall biogeochemical potential of the community may remain relatively unchanged. Of course, extreme increases in salinity will potentially result in the emergence of dominant specialist species, decreasing diversity and potentially influencing function. There is the potential that the discrete clustering of our samples may be related to technical bias, because of the different strategies for sample collection, sequencing and analysis of metagenomes from other locations. However, when we compared our data with metagenomes generated using different DNA extraction techniques and sequencing platforms, no discernible pattern emerged that can link the relatedness of metagenomes to elements of methodology ( Figs. 3 & 4 ). DNA extraction and sequencing techniques have also been shown not to significantly influence metagenomic profile discrimination by habitat [31] . Additionally, marine sediment samples extracted in the same lab using identical techniques did not cluster taxonomically with the Coorong samples ( Fig. 3 ) and Coorong water samples extracted using the same lab and techniques did not cluster with the Coorong sediment samples ( Figs. 3 & 4 ), indicating methodology is not obscuring environmental clustering. One caveat that should be considered when interpreting our data is the use of annotated data to compare metagenomes. Our data is reflective of the genomes and metabolic subsystems present in the MG-RAST database [20] and should be interpreted as patterns observed in the context of this diversity. Metagenomic databases are composed of taxa for which whole genome sequences exist, which represent a biased subsection of microbial diversity heavily skewed towards cultured organisms chosen because of ease of growth or interesting phenotypes [26] , [27] . Thus the databases tend to be skewed towards the phyla Proteobacteria , Firmicutes , Actinobacteria and Bacteroidetes \n [26] . Whilst genome based databases represent a valid reference point for relative comparison of the taxonomic affiliation of subsystems observed in the data, which has been routinely applied for metagenomes [20] a much broader view of the taxonomic variability can be provided by the 16S rDNA gene [26] . Further analysis using clustering algorithms [32] and di-nucleotide frequencies [22] will shed light on how our un-annotated data is similar to other metagenomes. This study focused on the balance between taxonomic and metabolic identifiers to determine the dominant controlling environmental factor. We found substrate type is the dominant controller of gene abundance. To date, the majority of community scale microbial biogeography studies have considered the presence or absence of particular taxonomic units. In many cases however, microbial biogeography is not binary, with most taxa being present but at a low abundance in the so called ‘rare biosphere’ [33] . Additionally, functional genes may be passed between different taxa via lateral gene transfer [34] , [35] indicating that taxonomy alone is not a determinant of community function. More sophisticated approaches which consider complex patterns in the metagenomic structure of communities and the complex interactions between different drivers acting on different scales are necessary to understand the spatial distribution of microbial diversity. High throughput sequencing allows profiling of both taxonomic and metabolic diversity and when coupled to statistical techniques [5] , [36] – [39] and standardized records of metadata [40] patterns in the composition of microbial metagenomes begin to emerge. One such pattern in our data is the high degree of taxonomic and functional similarity between metagenomes derived across a strong salinity, nutrient and abundance gradient and between metagenomes derived from sediment/soil/mat metagenomes regardless of salinity. Another pattern is the dichotomous clustering of solid substrate metagenomes and water metagenomes into discrete similarity groups which are not masked by differences in salinity. Overall our results suggest that substrate type (water or solid substrate) plays a fundamental role in determining the composition of the metagenome and that, in addition to extant physiochemical parameters, needs to be considered when interpreting patterns in microbial community diversity."
} | 3,897 |
21629774 | PMC3100327 | pmc | 7,012 | {
"abstract": "Biodiesel is a renewable alternative to petroleum diesel fuel that can contribute to carbon dioxide emission reduction and energy supply. Biodiesel is composed of fatty acid alkyl esters, including fatty acid methyl esters (FAMEs) and fatty acid ethyl esters (FAEEs), and is currently produced through the transesterification reaction of methanol (or ethanol) and triacylglycerols (TAGs). TAGs are mainly obtained from oilseed plants and microalgae. A sustainable supply of TAGs is a major bottleneck for current biodiesel production. Here we report the de novo biosynthesis of FAEEs from glucose, which can be derived from lignocellulosic biomass, in genetically engineered Escherichia coli by introduction of the ethanol-producing pathway from Zymomonas mobilis , genetic manipulation to increase the pool of fatty acyl-CoA, and heterologous expression of acyl-coenzyme A: diacylglycerol acyltransferase from Acinetobacter baylyi . An optimized fed-batch microbial fermentation of the modified E. coli strain yielded a titer of 922 mg L −1 FAEEs that consisted primarily of ethyl palmitate, -oleate, -myristate and -palmitoleate.",
"conclusion": "Conclusions In this study, a de novo biosynthetic pathway yielding fatty acid ethyl esters was constructed by genetically engineering E. coli . The fed-batch microbial fermentation was optimized with a maximum production of 922 mg L −1 FAEEs. Although the titer of FAEEs is low for further scaling-up, this work shows the feasibility and potential to utilize lignocellulosic biomass-derived sugars instead of oily biomass-derived TAGs to produce biodiesel. FAEE production could be significantly improved by increasing the fatty acid biosynthetic flux, balancing ethanol production and fatty acid synthesis, and engineering the WS/DGAT enzyme toward higher substrate specificity to ethanol and higher catalytic efficiency.",
"introduction": "Introduction In order to meet the rapidly growing demand for transportation fuel and to achieve reduction of carbon dioxide emissions, development of renewable energy sources has become more and more urgent. Biodiesel, as one type of renewable energy, is an ideal substitute for petroleum-based diesel fuel and is usually made from plant oils or animal fats (triacylglycerides) by transesterification with methanol or ethanol resulting in fatty acid methyl esters (FAMEs) and fatty acid ethyl esters (FAEEs). However, the limited supply of bioresources to obtain triacylglycerides (TAGs) is becoming a major bottleneck for biodiesel production. The main reason is that vegetable oil feedstocks are also food sources and their planting is geographically limited. Microalgae are currently viewed as one of the most promising TAG feedstocks for biodiesel production. Although the productivity of these photosynthetic microorganisms greatly exceeds that of agricultural oleaginous crops, current microalgae production using the best available strains and cultivation methods has not yet become economically feasible for biodiesel production [1] . Lignocellulosic biomass is a relatively sustainable bioresource to make biofuels. In the past few decades, tremendous research and development efforts on cellulosic ethanol have resulted in significant advances and many technical problems have been solved [2] , [3] . However, due to its low energy density, high vapor pressure and corrosiveness, bioethanol is not an ideal alternative to petroleum-derived fuels. Moreover, the high solubility of ethanol also results in toxicity to the microbes used to produce it [4] . Recently, technical routes have been developed to produce novel biofuel products with higher energy density and hydrophobic properties from lignocellulose-derived sugars. Liquid alkanes chemically identical to petroleum fuels can be made by both biosynthesis with genetically modified E. coli \n [5] and synthetic catalytic conversion [6] . Biosynthetic production of C4–C8 alcohols with straight or branched chains through manipulation of amino acid metabolic pathways in E. coli has been reported [7] . Thus, the production of novel biofuels from lignocellulosic biomass-derived sugars is a promising alternative to bioethanol. Bioresource technologies for biodiesel production from lignocellulosic biomass, distinct from approaches using the oily fraction of biomass, can be developed through constructing non-native biosynthetic pathways of biodiesel molecules in microbial hosts. Direct microbial production of FAEEs in engineered E. coli was first reported by co-expressing genes coding enzymes for ethanol production from Zymomonas mobilis and the WS/DGAT gene encoding acyl-coenzyme A: diacylglycerol acyltransferase from Acinetobacter baylyi strain ADP1 [8] . However, biosynthesis of FAEEs in that work relied on the supplementation of exogenous fatty acids. In last couple of years, research focusing on overproduction of fatty acids by genetically engineering the fatty acid metabolic network towards providing precursors of fatty acid-based biofuels such as FAMEs, FAEEs, fatty alcohols, and fatty alkanes, has been extensively investigated in E. coli . The titer of 2.5 g L −1 day −1 free fatty acids was produced by an E. coli mutant strain in fed-batch fermentation with overexpression of acetyl-CoA carboxylase (ACC) and acyl-ACP thioesterase (TE) from E. coli as well as plant TE from Cinnamomum camphorum , and with deletion of fadD , which codes for fatty acyl-CoA synthase, the first step of fatty acid degradation [9] . An improved yield of fatty acids to a titer of 4.5 g L −1 day −1 has been demonstrated by the same group [10] . A very recent research publication in Nature has demonstrated de novo biosynthesis of FAEEs in E. coli by co-overproduction of ethanol and fatty acids in genetically engineered E. coli . In this work, several genetic engineering strategies were developed to improve fatty acid production of the ethanol-producing mutant strain to increase the yield of FAEEs, including overexpression of thioesterases, heterologous introduction of fadD gene from Saccharomyces cerevisiae , and the deletion of the fadE gene the encodes acyl-CoA dehydrogenase, leading to a yield of FAEEs of up to 674 mg L −1 . Researchers have also demonstrated the feasibility of in vivo production of FAEEs from hemicellulose achieved via genetic engineering of the endoxylanase catalytic domain from Clostridium stercorarium and the xylanase from Bacteroides ovatus \n [11] . In our study, six distinct genetic alterations have been introduced into an E. coli BL21 (DE3) host strain ( Fig. 1 ): (1) heterologous expression of the genes pdc and adhB from the ethanol-producing pathway in Zymomonas mobilis , coding pyruvate decarboxylase and alcohol dehydrogenase, respectively; (2) overexpression of accBACD genes coding acetyl-CoA carboxylase, which converts acetyl-CoA to malonyl-CoA, the first step of fatty acid biosynthesis that is proposed to be the rate-limiting step; (3) overexpression of the modified tesA' gene from E. coli , coding a leaderless version of thioesterase; (4) knockout of the fadE gene, coding acyl-CoA dehydrogenase that dehydrogenates fatty acyl-CoA, the second step of fatty acid degradation; (5) overexpression of the fadD gene from E. coli , coding a fatty acyl-CoA ligase that catalyzes the conversion of free fatty acids to fatty acyl-CoA; and (6) heterologous expression of the atfA gene from A. baylyi , coding the wax ester synthase/acyl-coenzyme A: diacylglycerol acyltransferase (WS/DGAT). Here, we also evaluated FAEE production in a scaled-up fed-batch fermentation, and optimized the nutritional and environmental conditions to improve the yield of FAEEs. 10.1371/journal.pone.0020265.g001 Figure 1 Constructed de novo biosynthetic pathway of fatty acid ethyl esters in E. coli . PDH: pyruvate dehydrogenase; ACC: acetyl-CoA carboxylase; BCCP: biotin carboxyl carrier protein; BC: biotin carboxylase; CT: carboxyltransferase; PDC: pyruvate decarboxylase; ADH: alcohol dehydrogenase; TE: thioesterase; FadD: fatty acyl-CoA synthase; FadE: acyl-CoA dehydrogenase; WS/DGAT: wax synthase/acyl-coenzyme A: diacylglycerol acyltransferase. Compared to Steen et al. 's work on FAEE production reported in Nature [11] , three significant differences were made in the current study. First, the E. coli mutant strain producing FAEE with over-expression of acetyl-CoA carboxylase that has been proved to be able to increase the rate of fatty acid biosynthesis in E. coli \n [12] was constructed in our study. Second, the effect of fadE deletion on FAEE production in both shake flask and scale-up fed-batch fermentation experiments was specially examined and analyzed in the report here. Third, we also evaluated FAEE production in a scale-up fed-batch fermentation, and optimized the nutritional conditions to improve the yield of FAEEs.",
"discussion": "Results and Discussion Shake flask experiment for E. coli strains With the aim to achieve de novo biosynthesis of FAEEs in E. coli , plasmid pXT11, which contains pdc and adhB from Zymomonas mobilis ZM4, fadD from E. coli , and atfA from Acinetobacter baylyi ADP1, was constructed. When the E. coli strain harboring pXT11 was induced by 0.5 mM IPTG, no trace of FAEEs was detected by GC-MS. This is consistent with the previously reported results that FAEEs biosynthesis was dependent on the presence of a certain amount of free fatty acids and that the native level of fatty acyl-CoA was not sufficient to make FAEEs [8] . pMSD8 and pMSD15, containing E. coli accBCDA and tesA' expressing cassettes respectively, were proven to be effective in overproduction of free fatty acids in E. coli \n [9] , [12] . In this study, pXT11, pMSD8, and pMSD15 were co-transformed into E. coli BL21 (DE3), and FAEEs were produced with the concentration of 34.6 mg L −1 by culturing the cells in shaking flasks with glucose as sole carbon source. To further increase the yield of FAEEs, the fadE gene, which encodes acyl-CoA dehydrogenase, was deleted to block degradation of fatty acyl-CoA, and it was observed that more than double the concentration of FAEEs, 77.5 mg L −1 , was produced in the recombinant strain under the same culture conditions. This result indicates that overproduced fatty acyl-CoA, which is the substrate of acyl-coenzyme A: diacylglycerol acyltransferase that catalyzes the production of FAEEs, is essential for de novo biosynthesis of FAEEs in E. coli , and that the metabolic engineering strategy applied here is effective to achieve accumulation of fatty acyl-CoA. The effect of initial culture medium on FAEE production \n Fig. S1a shows the production of different FAEEs using E. coli mutant strain BL21 (ΔfadE)/pXT11/pMSD8/pMSD15 fermented with three initial culture media. The maximum concentrations of FAEEs were 735, 922, and 328 mg L −1 ( Table 1 ) when the cells were fermented in LB, 2LB, or 2LB+Phosphates medium, respectively. Thus, 2LB medium was selected as the initial culture medium for further experiments. It suggests that higher levels of nutrients in the initial culture medium have a positive effect for FAEE production and that excessive phosphates have negative effect. 10.1371/journal.pone.0020265.t001 Table 1 FAEE production of E. coli mutant strain BL21 (ΔfadE)/pXT11/pMSD8/pMSD15 under varied fed-batch fermentation conditions described in Materials and Methods . \n Categories of fermentation conditions \n \n Varing conditions \n Maximum production of FAEE (mg L −1 ) Initial culture medium LB 735 2LB 922 2LB+phosphates 328 Feeding conditions 75 g/0.22 ml min −1 \n 588 100 g/0.22 ml min −1 \n 922 150 g/0.22 ml min −1 \n 581 200 g/0.11 ml min −1 \n 464 Culture temperature 30°C 922 25°C 652 Time for starting induction 4 hr 333 11 hr 922 16 hr 682 The effect of feeding conditions on FAEEs production The effects of glucose concentration (75, 100 and 150 g/750 ml) in the fed-batch fermentation of strain BL21 (ΔfadE)/pMSD8/pMSD15/pXT11 with the feed rate of 0.22 ml min −1 on the FAEEs production were evaluated. FAEEs were generated at 922 mg L −1 with 100 g glucose/750 ml feeding culture, while 581 and 588 mg L −1 FAEEs were produced in 75 g and 150 g glucose/750 ml feeding culture, respectively ( Table 1 , Fig. S1b ). Moreover, the total FAEE concentration was only 464 mg L −1 when the glucose concentration was increased to 200 g glucose/750 ml feeding culture and the feed rate was decreased to 0.11 ml min −1 , as shown in Table 1 . This suggests that the concentration of glucose in the feeding culture and the feeding rate are highly related to FAEE production. The effect of culture temperature on FAEE production The best FAEEs-producing strain in this work, E. coli BL21 (ΔfadE)/pMSD8/pMSD15/pXT11, contains three plasmids with three different origins. Temperature could affect the stability of these plasmids and also affect the production of FAEEs. When the strain was cultured at 37°C, no detectable FAEEs were found by GC-MS (data not shown). However, FAEEs were produced at 922 mg L −1 by the same strain when cultured at 30°C, and 652 mg L −1 of FAEEs were produced when the culturing temperature was decreased to 25°C ( Table 1 , Fig. S1c )). The effect of induction time point on FAEE production In order to test the impact of induction time on FAEE production, three different time points in the early exponential stage were tested for induction of gene expression. As shown in Fig. S1d , when the culture was induced at an OD 600 of 4, of 312 mg L −1 of FAEEs were produced at 18 h post-induction, and the total FAEE concentration did not change significantly during the next 20 h. When induced at an OD 600 of 16, the concentration of FAEEs reached 682 mg L −1 at 32 h after the induction. The maximum FAEE yield of 922 mg L −1 was achieved 45 h after induction when the culture was induced at an OD 600 of 11 ( Table 1 ). Production of FAEEs in a fed-batch fermentation under the optimized conditions To evaluate the performance of the fatty acid ethyl ester-overproducing strain in a large scale, a 5-L fed-batch fermentation was performed with the optimized cultivation conditions described above, in which fed-batch fermentations were carried out at 30°C with 2LB as the initial culture medium and 100 g glucose/750 ml as the feeding culture at a feed rate of 0.22 ml min −1 , and the culture was induced at an OD 600 of 11. Three recombinant E. coli strains, BL21 (DE3)/pXT11, BL21 (DE3)/pMSD8/pMSD15/pXT11, and BL21 (ΔfadE)/pMSD8/pMSD15/pXT11, were grown, and the concentrations of cells, ethanol, and FAEEs were measured. As shown in Fig. 2a , the growth of all three E. coli mutant strains was consistent with the logistic growth model. In the fed-batch fermentation, there was nearly no lag phase and cells grew directly into the exponential period, followed by a steady period after around 65–70 h. The maximum optical densities of the three strains, BL21 (DE3)/pXT11, BL21 (DE3)/pMSD8/pMSD15/pXT11, and BL21 (ΔfadE)/pMSD8/pMSD15/pXT11, reached around 24, 30 and 28 respectively. There is no significant difference among three cell growth profiles. 10.1371/journal.pone.0020265.g002 Figure 2 Analysis on fed-batch fermentations under the optimized conditions described in Materials and Methods . (a) Cell growth, (b) Ethanol production, (c) FAEE production of three E. coli mutant strains, BL21 (DE3)/pXT11 and BL21 (DE3)/pMSD8/pMSD15/pXT11 and BL21 (ΔfadE)/pMSD8/pMSD15/pXT11. (d) Three paralleled experiments for FAEE production of the strain BL21 (ΔfadE)/pMSD8/pMSD15/pXT11. Ethanol production of the mutant strain BL21 (DE3)/pXT11, not producing FAEE, was clearly different than that in the two FAEE-producing strains ( Fig. 2b ). After induction, the concentration of ethanol increased slightly to around 1.3 g L −1 and then remained unchanged until the end of fermentation ( Fig. 2b ). It is reasonable that there is no ethanol consumption since there is no WS/DGAT enzyme in this strain to convert ethanol and no FAEE production in the strain. The ethanol production profile of the strain BL21 (DE3)/pMSD8/pMSD15/pXT11 was similar to that of the strain BL21 (ΔfadE)/pMSD8/pMSD15/pXT11 ( Fig. 2b ), and the maximum ethanol accumulations of both FAEE-producing strains were also similar (about 3 g L −1 ). The decrease of ethanol accumulations of the strain BL21 (ΔfadE)/pMSD8/pMSD15/pXT11 occurred earlier than that from the strain BL21 (DE3)/pMSD8/pMSD15/pXT11, and it is consistent with the FAEE production shown in Fig. 2c . Based on the results obtained from the shake flask experiments, it was expected that no detectable FAEE would be obtained after induction during the fed-batch fermentation of BL21 (DE3)/pXT11 ( Fig. 2c ). FAEE production during fermentation of BL21 (DE3)/pMSD8/pMSD15/pXT11 and BL21 (ΔfadE)/pMSD8/pMSD15/pXT11 increased along with the decrease in ethanol production, and reached a maximum of 477 and 922 mg L −1 respectively, about 14- and 12-fold higher than the production in the corresponding shake flask experiments respectively. The fed-batch fermentation experiments also demonstrated the same doubling effect of the deletion of fadE on FAEE production as was observed in the shake flask experiments. To confirm FAEE production under the optimized fed-batch fermentation conditionsAnd thereby, two other parallelled fermentation experiments of fermentation were performed with the strain BL21 (ΔfadE)/pMSD8/pMSD15/pXT11 under the optimized culture conditions ( Fig. 2d ). The FAEE production, the glucose conversion efficiency and the specific productivity were calculated to be 818.50±94.64 mg L −1 , 24.56±2.84 mg FAEE/g glucose and 0.46±0.11 mg L −1 OD −1 h −1 respectively. Composition of FAEEs during the optimized fed-batch fermentation \n Fig. 3 illustrates the compositions of FAEEs produced in the fed-batch fermentation cultures of the three E. coli mutant strains. No detectable FAEE was found in the culture of BL21 (DE3)/pXT11. The fermentation of BL21 (DE3)/pMSD8/pMSD15/pXT11 produced 151.6 mg L −1 ethyl myristate (C14∶0; 32%) as the major constituent, with 99.6 mg L −1 ethyl oleate (C18∶1; 21%), 80.2 mg L −1 ethyl palmitate (C16∶0; 17%), 76.3 mg L −1 ethyl palmitoleate (C16∶1; 16%), 33.2 mg L −1 ethyl laurate (C12∶0; 7%), and 33.1 mg L −1 ethyl myristoleate (C14∶1; 7%) as the minor FAEEs observed. The fed-batch fermentation of BL21 (ΔfadE)/pMSD8/pMSD15/pXT11 strain under the same cultivation condition produced ethyl palmitate (16∶0; 31.3%) and ethyl oleate (18∶1; 31.4%) as the two major FAEE constituents, together with ethyl myristate (14∶0, 24.1%), ethyl palmitoleate (16∶1, 9.9%), and other two minor FAEE constituents, ethyl laurate (12∶0) and ethyl myristoleate (14∶1). 10.1371/journal.pone.0020265.g003 Figure 3 Composition of fatty acid ethyl esters with different carbon chain length and saturation degree during the optimized fed-batch fermentation of E. coli mutant strain BL21 (ΔfadE)/pMSD8/pMSD15/pXT11. By comparison of the FAEE production of the two E. coli mutant strains, BL21 (DE3)/pMSD8/pMSD15/pXT11 and BL21 (ΔfadE)/pMSD8/pMSD15/pXT11, we found that blocking the degradation of fatty acyl-CoA through the deletion of the fadE gene encoding acyl-CoA dehydrogenase caused significant changes in the FAEE composition. Both mutant strains produced fatty acid ethyl esters with carbon chain length varying from 12 to 18. Neither of mutant strains produced ethyl stearate (18∶0), and ethyl laurate and ethyl myristoleate were minor constituents for both mutants. After fadE deletion, major products were ethyl palmitate (16∶0) and ethyl oleate (18∶1) rather than ethyl myristate (14∶0) in the mutant strain BL21 (DE3)/pMSD8/pMSD15/pXT11. Conclusions In this study, a de novo biosynthetic pathway yielding fatty acid ethyl esters was constructed by genetically engineering E. coli . The fed-batch microbial fermentation was optimized with a maximum production of 922 mg L −1 FAEEs. Although the titer of FAEEs is low for further scaling-up, this work shows the feasibility and potential to utilize lignocellulosic biomass-derived sugars instead of oily biomass-derived TAGs to produce biodiesel. FAEE production could be significantly improved by increasing the fatty acid biosynthetic flux, balancing ethanol production and fatty acid synthesis, and engineering the WS/DGAT enzyme toward higher substrate specificity to ethanol and higher catalytic efficiency."
} | 5,125 |
29468117 | PMC5779716 | pmc | 7,013 | {
"abstract": "Soil microorganisms mineralize lignin-derived aromatic carbon sources using oxidative catabolic pathways, such as the β-ketoadipate pathway. Although this aromatic pathway is one of the best-studied pathways in biochemistry, the complete pathway, including its regulation by aromatic carbon sources, has not been integrated into the metabolic network. In particular, information about the in vivo operation ( e.g. , kinetics and flux capacity) of the pathway is lacking. In this contribution, we use kinetic modeling and thermodynamic analysis to evaluate the in vivo operation of this key aromatic multi-step pathway. The resulting ab initio deterministic model of benzoate degradation via the β-ketoadipate ( ortho -cleavage) pathway in Pseudomonas putida KT2440 is presented. The kinetic model includes mechanistic rate expressions for the enzymes and transport processes. The design and experimental validation of the model are driven by data generated from short-term perturbation experiments in a benzoate-limited continuous culture. The results of rigorous modeling of the in vivo dynamics provide strong support for flux regulation by the benzoate transporter and the enzymes forming and cleaving catechol. Revisiting the β-ketoadipate pathway might be valuable for applications in different fields, such as biochemistry and metabolic engineering, that use lignin monomers as a carbon source.",
"introduction": "1 Introduction In nature, the metabolic conversion of lignin to aromatic monomers occurs via a series of cometabolic and secondary metabolic events that involve both fungi and bacteria ( Bugg et al., 2011 ). Understanding the innate metabolism of microbes while they utilize complex aromatic compounds as their sole source of carbon and energy may provide clues for decoding the possible links among lignin degradation, aromatic monomers, central carbon metabolism and bio-product formation. Soil microorganisms ( e.g. , the Pseudomonads) are known to exhibit a common procession that is termed the ‘metabolic funnel’ to mineralize structurally diverse aromatic molecules into a few central aromatic intermediates through peripheral catabolic pathways. Those aromatic intermediates are further processed via the central aromatic pathways to the core metabolism of the cell ( Díaz, 2004 ). The genomic analysis of a well-characterized member of the metabolically versatile bacterial species Pseudomonas putida revealed the existence of at least six major aromatic pathways for degrading central aromatic intermediates (i.e., protocatechuate, catechol, homogentisate, phenylacetate, nicotinic acid and gallic acid) ( Jiménez et al., 2008 , Jiménez et al., 2002 , Nogales et al., 2011 ). Of these major pathways, the pathway involved in catechol degradation (i.e., the β-ketoadipate pathway) is the most prevalent and extensively studied. In addition to having a wide taxonomic distribution among both the prokaryotic and eukaryotic groups, the β-ketoadipate pathway provides a mechanism for the dissimilation of aromatic and hydroaromatic compounds ( Parke et al., 2000 ). The pathway follows a two-step strategy: a ring cleavage by ring modification reactions is followed by ring fission and subsequent reactions that lead to the generation of TCA cycle intermediates ( Harwood and Parales, 1996 ). This versatile metabolism of aromatic compounds has now sparked interest in renewable resource-based chemistry, and the possible impacts were reviewed previously ( Fuchs et al., 2011 , Wells and Ragauskas, 2012 ). One example is the microbial synthesis of cis , cis -muconic acid from benzoate. This route is explored as an alternative to chemical synthesis for the production of adipic acid, which is a commodity chemical for the production of nylon-6,6 ( Mizuno et al., 1988 , van Duuren et al., 2011 , Vardon et al., 2015 ). The potential to apply these catabolic pathways for aromatic compounds in lignin-enriched streams and lignin-derived aromatic monomers can be further explored by examining the formation of medium-chain (C 6 –C 14 ) polyhydroxyalkanoates by the versatile soil bacterium P. putida KT2440 ( Linger et al., 2014 ). In P. putida KT2440, the β-ketoadipate pathway is native, encoded on the chromosome, and lacks the meta -cleavage route that is localized on a mega plasmid ( Nelson et al., 2002 ). Previous studies have shown direct evidence for the activation of the ortho -cleavage route of the β-ketoadipate pathway during the growth of P. putida on benzoate as the sole source of carbon and energy ( Cao and Loh, 2008 , Ornston and Stanier, 1966 ). Although numerous biochemical studies of the in vitro properties of the corresponding enzymes are available, the in vivo kinetics of this pathway remains poorly understood. The central goal of this study is to quantify the in vivo metabolic response during benzoate degradation via the β-ketoadipate ( ortho -cleavage) pathway in P. putida KT2440 ( Sudarsan et al., 2014 ). Consequently, we analyze the enzyme kinetics in vivo using a mathematical model based on deterministic rate equations. The analysis is data-driven, and the response of the metabolome to dynamic system excitation ( Theobald et al., 1997 ) is used to identify the kinetics of multistep enzyme reactions and to estimate the kinetic parameters using a bottom-up approach ( Bruggeman and Westerhoff, 2007 ). The identified model is then used to predict system dynamics (i.e., changes in the intra- and extracellular concentrations of metabolites both at steady state and after a benzoate pulse). In the second part of this study, we draw on quantitative metabolome analysis and apply network-embedded thermodynamic (NET) analysis ( Zamboni et al., 2008 ) as a tool to provide novel insights about the active regulatory sites of the β-ketoadipate ( ortho -cleavage) pathway coupled to central carbon metabolism. The systematic incorporation of insights from the identified dynamic model and thermodynamic analysis will be useful in the future as a framework for formulating large-scale kinetic models.",
"discussion": "5 Discussion We investigated the well-known but poorly understood in vivo regulatory interactions of the β-ketoadipate ( ortho -cleavage) pathway in P. putida KT2440 using an iterative model building process. Critical assumptions were necessary to describe the in vivo behavior of metabolites with appropriate reaction kinetics. In addition to metabolic regulation, there must be an additional regulation phenomenon at the enzyme level; this phenomenon was observed during the entire course of our pulse response experiments. The changes in metabolic regulation is known to occur on the time scale of seconds ( Theobald et al., 1997 ). The hypothesis that an additional regulation phenomenon is superimposed on metabolic regulation was supported by the increasing degradation rate of benzoate with time ( Fig. 7 A). This observation is the main reason for restricting the metabolic response to 12 min after perturbation and not to a longer scale (12–23 min). The importance of additional regulation at the enzyme level was further illustrated in an experiment with sequential pulse experiments ( Fig. 7 B and C), in which the uptake rate of benzoate increased as the number of generations increased in a continuous culture kept at a constant dilution rate. Although the linear increase in the substrate uptake rate is a characteristic feature of continuous cultivation, whether the entire pathway or only the benzoate transport enzyme is optimized for carbon uptake is unknown and must be investigated thoroughly. A previous study showed that when the yeast S. cerevisiae is confronted with multiple glucose stimuli, the specific uptake rate of glucose is increased by a factor of seven compared with its steady-state value; this change accompanies altered expression of glucose transport proteins ( Buziol et al., 2008 ). Moreover, prolonged glucose limitation of S. cerevisiae in a chemostat causes multiple steady states with different metabolite and enzyme levels ( Wu et al., 2006 ). The observed deviations between the model predictions and the measurements ( Fig. 3 , Fig. 4 ) also indicate the presence of an additional regulation phenomenon in P. putida that cannot be compensated by parameter fitting with the existing model structure. Fig. 7 Changes in benzoate concentration and benzoate uptake rate in a benzoate-limited chemostat. (A) Change in benzoate concentration after a pulse of benzoate. The broken lines represent the time point of the benzoate pulse. (B) Specific rate of benzoate uptake during benzoate-limited cultivation of P. putida KT2440 at a dilution rate of 0.2 h −1 . The data points represent the benzoate uptake rates calculated after the steady-state conditions were disturbed with identical benzoate pulses. The data points presented with filled circles represent one prolonged benzoate-limited chemostat (filled circles), and the data points presented with open squares, open diamonds, and open triangles represent individual chemostat experiments. (C) Changes in extracellular benzoate from three identical benzoate pulses performed in one prolonged chemostat at different time points. The broken lines represent the time point of the benzoate pulse. The data points presented with open circles, open squares and crosses represent the three different pulse experiments that were performed at 10.9, 17.8 and 27.3 generations, respectively. These figures represent the presence of an additional regulation phenomenon that cannot be described by the present model. Fig. 7. Major assumptions were made with respect to the feedback inhibition of benzoate transport and benzoate cis -diol dehydrogenase. Although no evidence to support a molecular basis for this phenomenon was reported in Pseudomonas for the former case of benzoate transport inhibition by benzoate cis -idol, a study with Corynebacterium glutamicum has shown that hydroxyl-substituted benzoates inhibit benzoate uptake significantly ( Wang et al., 2011 ). In the latter case, feedback inhibition of benzoate cis -diol dehydrogenase by catechol was used purely to describe the in vivo dynamics of benzoate cis -diol and catechol concentrations. In addition, the reason for the low in vivo concentration of catechol compared with benzoate cis -diol might be the existence of two dedicated catechol dioxygenases, of which CatA2 was recently described as a safety valve that protects against toxic concentrations of catechol ( Jimenez et al., 2014 ). Consequently, all of these factors might result in significant control of the total pathway flux by the benzoate cis -diol dehydrogenase. The thermodynamic analysis indicated that the estimated Gibbs energy values of some reactions ( ∆ r G ) were largely negative and that these reactions were operating far from equilibrium ( Fig. 5 ). These reactions are catalyzed by benzoate cis -diol dehydrogenase (BZDIOLDH), catechol 1,2-dioxygenase (CATDOX), muconate cycloisomerase (MUCCYCI), β-ketoadipyl CoA thiolase (3OXCOAT), and β-ketoadipate CoA transferase (3OADPCOAT). These five reactions were classified as non-equilibrium reactions because at large − ∆ r G , the thermodynamic driving force is significant, and the rate of the reaction is regulated predominantly by changes in the enzyme capacity (for example, via transcription or the binding of effectors). These indications for sensitivity to additional regulation phenomena are partially manifested in the corresponding kinetic expressions. In our benzoate-pulse experiments, the substrates of these reactions were also found to be extracellular; this observation was true to a greater extent for some metabolites, such as β-ketoadipate, benzoate cis -diol, and cis , cis -muconate. The excretion or the overflow of the individual metabolites is most likely a result of the adaptation of the culture to the steady-state conditions of the continuous culture. Some of these excreted metabolites are extremely toxic ( e.g. , catechol), and other metabolites may be excreted due to simple metabolic overflow. The scenario can be changed by using different dilution rates, which may result in different concentrations of the enzymes. However, these phenomena were not studied in this project. Bacterial mineralization of aromatics is well known to be critical e.g. on the level of benzene ring cleavage ( Dagley, 1971 , Evans, 1963 ). We initially hypothesized that catechol accumulation in shake flasks experiments are regulated by ring cleavage enzymes. The model and especially the quantitative description of all the rates allowed us to verify this hypothesis. In addition, catechol formation and especially benzoate uptake were identified as key and previously not recognized regulation points. For benzoate uptake, this new hypothesis was experimentally verified by sequential pulses. Amazingly cells reacted via aging/adaptation to the bottleneck benzoate uptake by increasing the benzoate uptake rate ( Fig. 7 B and C). This now can be taken as starting point again for defining new objectives, e.g. substrate uptake, product formation, etc . to run the systems biology iterative process (experiment-model-hypothesis-experiment…) a second time. In addition, model building is a purpose-driven-process. In other words: first define the purpose before starting to build the model. The purpose for the presented dynamic model is the application of re-designing the enzymatic makeup for the process/product of interest. In our specific case, objectives for follow up biotechnology research might be rate improvements for benzoate uptake and catechol formation and cleavage – the major flux bottlenecks identified by in vivo dynamic modeling ( Fig. 1 ) on whatever level appropriate (metabolic engineering, reaction engineering, process engineering). In summary, a data-driven modeling approach increased our understanding of the complex interactions of the β-ketoadipate ( ortho -cleavage) pathway. The proposed model can capture the stationary and dynamic behavior of intracellular and extracellular metabolites. The systematic integration of quantitative metabolite data with thermodynamic information was pivotal for demonstrating that only a few reactions of benzoate metabolism are candidates for active regulation ( e.g. , transcriptional and/or allosteric regulation), whereas the rates of most reactions depend only on substrate concentrations that are close to equilibrium. Although the critical assumptions that were made regarding the regulation of the benzoate transport rate and benzoate cis -diol dehydrogenase activity are supported by in vivo dynamic modeling, they do invite future confirmation by in vitro investigations of these proteins. These findings highlight the necessity of quantifying metabolic rates of well known metabolic pathways in vivo . The biochemistry of pathway enzymes in the cytoplasm has to be described both theoretically in (dynamic) models as well as verified experimentally, e.g. in chemostat experiments. In this respect, direct translations of genomic sequence data/ annotations into metabolic flux networks are necessary, but not sufficient."
} | 3,807 |
26788423 | PMC4715433 | pmc | 7,014 | {
"abstract": "Coral disease literature has focused, for the most part, on the etiology of the more than 35 coral afflictions currently described. Much less understood are the factors that underpin the capacity of corals to regenerate lesions, including the role of colony health. This lack of knowledge with respect to the factors that influence tissue regeneration significantly limits our understanding of the impact of diseases at the colony, population, and community level. In this study, we experimentally compared tissue regeneration capacity of diseased versus healthy fragments of Gorgonia ventalina colonies at 5 m and 12 m of depth. We found that the initial health state of colonies (i.e., diseased or healthy) had a significant effect on tissue regeneration (healing). All healthy fragments exhibited full recovery regardless of depth treatment, while diseased fragments did not. Our results suggest that being diseased or healthy has a significant effect on the capacity of a sea fan colony to repair tissue, but that environmental factors associated with changes in depth, such as temperature and light, do not. We conclude that disease doesn’t just compromise vital functions such as growth and reproduction in corals but also compromises their capacity to regenerate tissue and heal lesions.",
"conclusion": "Conclusions Diseases of corals not just compromise vital functions such as growth and reproduction, but also compromise their recovery capacity. Arguably, resources invested against pathogens could also be the same driving the tissue repair as stated by limited budget theory proposed by Oren et al. (2001) . This raises questions regarding the sharing of resources and resource depletion. For instance, in the eventuality of two simultaneous but different immunological insults, how corals should prioritize its resources to respond to both events? How intense should a disturbance be in order to induce immune responses that affect several life history traits? It that regard, it could be possible that the environmental conditions in this study may have indeed caused stress on the sea fan fragments, but these stresses were manifested in other vital functions such as reproduction or growth which were not studied in this work. Our study also shows that sea fans are very robust corals which can tolerate variable environmental conditions; this may explain why this species thrives relatively well in many coral reefs across Puerto Rico regardless of environmental degradation.",
"introduction": "Introduction Most of the present-day coral reef habitats no longer exhibit the complex community structure that was commonly observed several decades ago. This is particularly evident in the Caribbean where the most important reef species such as the coral-building Caribbean Acropora palmata , A. cervicornis and the Orbicella complex (formerly Montastraea ), and the predatory reef fish and herbivores such as the black sea urchins and sea fan corals, have dramatically decreased in abundance ( Kim & Harvell, 2002 ). These losses have not just changed the seascape of the reefs, but have also caused important ecological alterations to coral survival, growth and reproductive schedules at local and regional scales ( Sutherland, Porter & Torres, 2004 ; Hoegh-Guldberg et al., 2007 ; Weil, Cróquer & Urreiztieta, 2009 ; Burns & Takabayashi, 2011 ; Ruiz-Diaz et al., 2013 ). Of the myriad of stressors affecting the viability of corals, disease is currently ranked at the top of the list. Coral diseases are typically diagnosed based on changes in the normal coloration of corals and by the appearance of lesions (partial tissue mortality). Under severe circumstances, such as when a pathogen is highly virulent or the coral host is immune-suppressed, disease-induced lesions can increase in size quickly, killing the colony. However, given a strong immune response, diseased-induced wounds can be contained and either persist for a prolonged period (if the colony is able to contain the disease but not regenerate new tissue) or are temporary (if the colony is able to regenerate tissue over the whole lesion) ( Ruiz-Diaz et al., 2013 ). Several studies have identified wound characteristic as a major factor affecting the rate at which a colony can regenerate new tissue and eliminate a lesion. For instance, several studies agree that regeneration rate decreases with an increase in lesion size ( Bak & Steward-Van, 1980 ; Oren et al., 2001 ; Kramrsky-Winter & Loya, 2000 ). Other studies suggest that the area/perimeter ratio of a lesion largely governs the rate of wound healing process ( Lirman, 2000 ). Further studies suggest that wound position within the colony (i.e., lesions at the edge of the colony vs. lesion at the center of the colony) determine the wound healing process ( Meesters, Bos & Gast, 1992 ). Many researchers have also linked the ongoing environmental degradation experienced by most coral reefs with the advent of coral diseases, which currently is one of the main sources of lesions on corals. For instance, in a study by Toledo-Hernández, Sabat & Zuluaga-Montero (2007) , the capacity of corals to recover from diseases (i.e., lesion recovery) was correlated with turbidity and/or sedimentation. Corals in areas with high turbidity and sedimentation had higher frequencies of disease-induced lesions and larger lesions compared to corals in less degraded habitats. Higher water temperature has been linked to the progression of lesions caused by black band disease, which affects several coral species in the Caribbean and the Great Barrier Reef ( Kuta & Richarson, 2002 ; Haapkylä et al., 2011 ). Similarly, nutrient enrichment increased the severity of aspergillosis of Gorgonia ventalina and yellow band disease on Orbicella annularis and O. franksi ( Bruno et al., 2003 ). Muller & Woesik (2009) showed that white-plague lesion significantly decreased on Corpophyllia natans shaded from solar radiation when compared to C. natans without shading. Although results from these studies have been useful in advancing our understanding of the healing process on corals, we still lack comprehensive knowledge of how other factors such as the health state of a colony baring lesion, affect the healing process. However, progress has been made. For instance, Fine, Oren & Loya (2002) (working with bleached scleractinian corals) and Ruiz-Diaz et al. (in press) (working with diseased gorgonians) have shown that diseased corals regenerate man-made lesions slower than man-made lesion inflicted on healthy-looking corals. Initiatives to mitigate the effects of coral disease lack information about factors affecting the recovery of corals from disease-induced lesions. While we do have some understanding about the factors that make a coral vulnerable to disease (i.e., abnormally high temperature and sedimentation among others) we lack understanding regarding how the health condition of the coral affects its recovery. The objective of this study is to experimentally test if the health state and variability in environmental factors correlated with depth, significantly influence lesion regeneration on the sea fan G. ventalina . To do this, we established eight nursery lines at two depths within the same reef (four nursery lines per depth, 5 m and 12 m). Each nursery line consisted of four fragments from two healthy and two diseased G. ventalina colonies. We scraped tissue from some of the healthy fragments and scraped the diseased area of the diseased fragments and followed their recovery through time. Concomitantly, we measured the temperature and light intensity at both depths (5 m and 12 m) to document differences in these factors between depths. We hypothesized that fragments from healthy colonies would regenerate new tissue at a faster rate than those from diseased colonies because, at the start of the experiment, diseased colonies are expected to have an activated immune response and thus fewer resources to allocate to tissue regeneration than healthy ones. We also reasoned that, independent of health state, tissue recovery rate at 12 m would be slower than at 5 m due to reduced light availability.",
"discussion": "Discussion Coral colonies are very vulnerable to tissue loss due to predation, pathogens, and abrasion, among other factors. Failure to regenerate lost tissue could impair their survivorship by allowing potentially harmful organisms to settle in the exposed skeleton, further infecting healthy areas of the corals. Repair failure could also affect other vital function of corals such the heterotrophic feeding and ultimately growth, in addition to reproduction, as loss of polyps will negatively affect such activities. Thus, tissue regeneration should be of utmost importance in order for coral colonies to reduce the risk of diseases, thereby improving their survivorship, competitive capacity and ultimately reproduction and somatic growth. Numerous researchers have studied the link between environmental factors, and the frequency and severity of coral diseases. In fact, some of these studies have argued that as climate change continues to exacerbate these factors, so will be the physiological stress associated with it, and that consequently the frequency and severity of coral disease, will also increase ( Kuta & Richarson, 2002 ; Haapkylä et al., 2011 ; Cróquer et al., 2006 ; Williams et al., 2014 ). In comparison, studies addressing how the health state of corals affects the coral’s capacity to repair are by far less common (however, see Mascarelli & Bunkley-William, 1999 ; Fine, Oren & Loya, 2002 ; Ruiz-Diaz et al., in press ). This study is an attempt to address this knowledge gap by documenting the relationship between the recovery dynamics of healthy and diseased coral colonies and environmental factors such as temperature, light intensity while controlling for genetic variability. Effect of the state of coral health on lesion recovery This study shows that the health state of colonies (i.e., being diseased or healthy) has a significant effect on the tissue repair capacity of sea fans. All healthy fragments, regardless of the depth where they were placed (thus regardless of temperature and light regimes), exhibited full recovery whereas diseased fragments did not. Furthermore, scraped healthy fragments healed faster than scraped diseased fragments (i.e., on average 78 days vs. 97 days, respectively). It is possible that genetic differences among colonies, which may have lead to different levels of susceptibility to disease in the first place, might have lead to the observed differences in healing rate. However, the result is that unscraped diseased fragments (DF) healed at a significantly slower rate than scraped ones (DFS) supports that tissue with lesion cannot heal as fast as tissue without a lesion even if they come from the same colony. In other words, growing tissue over a skeleton covered with fouling organisms is a slower process because it is more costly, as the coral is competing for space and also allocating resources into tissue regeneration. By contrast, scraped fragments can allocate resources into tissue regeneration. The results of the experiment agree with our initial hypothesis, which stated that the health state does affect the capacity of fragments to recover. In fact, our results show that being diseased negatively affects the capacity of fragments to recover. These results also concur with several authors who have argued that the diseased condition negatively affects the tissue regeneration capacity of corals. For instance, Mascarelli & Bunkley-William (1999) compared the rates of tissue regeneration of Orbicela annularis corals under contrasting health conditions (healthy and artificially bleached fragments) and reported that healthy ramets did not just heal completely but also recovered faster than diseased ones. By contrast, two of the bleached ramets died, and the remaining fragments did not exhibit full recovery. Likewise, Ruiz-Diaz et al. (in press) scraped naturally occurring lesions from sea fan colonies and as control scraped the equivalent of 10% of the surface area of healthy sea fan colonies, and found that tissue recovery was significantly slower in diseased fans when compared to healthy fans. A plausible explanation for these differences is that diseased colonies have fewer resources to invest into tissue repair as their resources were already compromised by the immune response prior to scraping ( Nagelkerken et al., 1997 ). Further evidence in support of this explanation of resource limitation would have been obtained by contrasting regeneration rates of healthy fragments from diseased colonies with that of diseased and healthy fragments from healthy colonies; however, we did not included healthy fragments from diseased colonies in our experimental design. Corals, like all living organisms, have finite resources to allocate into several vital functions such as growth, reproduction, immune defense or lesion regeneration. Given these resource constraints, the allocation of resources into certain vital functions, such as immune defense, means that fewer resources could be available for lesion regeneration ( Oren et al., 2001 ). Several studies conducted on a variety of corals support this hypothesis. For instance, Petes et al. (2003) working on sea fan coral G. ventalina reported reproductive suppression in diseased colonies, presumably due to a shift in resource allocation from reproduction to immunity. Similarly, Palmer, Bythell & Willis (2010) suggest that Porites sp. invests considerably more resources into immune constituents such as melanin biosynthesis than A. millepora . This investment of resources into immunity provides Porites with a higher disease and bleaching resistance. By contrast, A. millepora invests more resources into growth compared to Porites , although at a cost in reduced immunity, as acroporids are among the corals most susceptible to bleaching and disease. Effect of depth on lesion recovery One of the main concerns of the scientific community is that changes in environmental conditions could induce physiological stress on corals ( Alker et al., 2004 ). These stresses could impair vital life history traits such as grow, reproduction or even the capacity of corals to recover after a disturbance. In our study, however, environmental factors associated with changes in depth, showed no evident effects on the capacity of sea fan fragments to regenerate tissue, even though, the parameters measured were statistically different between depths. Our failure to detect depth effects could have several explanations, not necessarily mutually exclusive. For instance, it could be possible that the difference in environmental factors recorded between 5 m and 12 m were not sufficient to induce physiological stresses on the fragments, thereby not precluding their capacity to regenerate tissue. Alternatively, it could be that there was a depth effect but it manifested on other life history traits such as reproduction or somatic growth, in which case we were not able to detect it. It is also plausible to argue that sea fans are rather tolerant to changes in environmental conditions. Indeed, Ruiz-Diaz et al. (in press) found no differences in tissue recovery of in G. ventalina inhabiting reefs with contrasting water quality."
} | 3,847 |
35196097 | PMC8865766 | pmc | 7,015 | {
"abstract": "Metabolic processes that fuel the growth of heterotrophic microbial communities are initiated by specialized biopolymer degraders that decompose complex forms of organic matter. It is unclear, however, to what extent degraders structure the downstream assembly of the community that follows polymer breakdown. Investigating a model marine microbial community that degrades chitin, we show that chitinases secreted by different degraders produce oligomers of specific chain lengths that not only select for specialized consumers but also influence the metabolites secreted by these consumers into a shared resource pool. Each species participating in the breakdown cascade exhibits unique hierarchical preferences for substrates, which underlies the sequential colonization of metabolically distinct groups as resource availability changes over time. By identifying the metabolic underpinnings of microbial community assembly, we reveal a hierarchical cross-feeding structure that allows biopolymer degraders to shape the dynamics of community assembly.",
"introduction": "INTRODUCTION Microbial communities mediate a staggering number of biological processes that contribute to the health of humans, animals ( 1 ), and the planet as a whole ( 2 ). The metabolic activity of co-occurring species results in the formation and depletion of nutrients in shared resource pools, where different modes of competition and cooperation affect community composition and form a network of metabolic cross-feeding ( 3 – 23 ). Individual species often provide specific metabolic functions that influence the growth of a broader community, including use of terminal electron acceptors ( 14 , 24 , 25 ), degradation of complex carbon sources ( 26 – 29 ), removal of toxic by-products ( 30 – 32 ), or assimilation of nitrogen and sulfur ( 18 , 33 – 35 ). However, we remain with a limited molecular understanding of the extent to which individual species affect the formation and structure of cross-feeding networks and community assembly. Here, we demonstrate multiple metabolic mechanisms by which specialized biopolymer degraders influence the trajectory of community assembly, exemplified with an 18-member community that thrives on chitin, the second most abundant polysaccharide on the planet ( 36 ). Similar to other polymer degrading communities ( 26 – 28 ), chitin communities require specialized degraders to supply nutrients to nonspecialized downstream consumers. Our community consists of phylogenetically diverse seawater isolates with distinct metabolic capabilities, which become abundant at different points during assembly of chitin communities. This provides a system that allows us to demonstrate how specialized chitin degraders initiate a hierarchal food web that shapes the community during the assembly process.",
"discussion": "DISCUSSION Resolving microbial community function to the level of contributing species is hampered by the difficulty in monitoring how nutrients are dispersed between species and how individual species can affect the broader community through metabolic interaction networks. Here, we demonstrate an approach for how community interaction networks can be learned from the characterization of individual species and how specialized biopolymer degraders scaffold these networks to shape the population. Degraders secrete chitinases that produce different oligomer profiles that influence the abundance of exploiters, which thrive primarily on oligomers that they cannot generate themselves. Moreover, the composition of oligomer profiles influences which metabolite exploiters secrete into a shared resource pool that is available to the entire population. Hence, degraders initiate a hierarchical cascade of nutrient flow into a population. An open question regarding the exchange of nutrients in a community is the extent to which it is mediated by broad and nonspecific cross-feeding networks ( 6 ) or by species-specific interactions with an ordered structure ( 7 , 38 , 39 ). Our results provide evidence that the exchange of nutrients has an ordered structure that allows individual species to shape the flow of metabolites and, subsequently, that nonspecific cross-feeding networks do not capture the complexity of nutrient exchange. We show that a multitude of secreted metabolites provide metabolic niches to support a large population, including major growth supporting substrates such as acetate, ammonium, and glutamate whose exchange appears to be nonspecific between most species. However, our results further demonstrate that scavengers, which can grow on neither chitin nor its degradation products, preferentially consume metabolites that are formed at later points in community succession, demonstrating how degraders scaffold downstream colonization of nonspecialized species. With our approach of integrating individual species consumption and secretion data, we were able to infer metabolic exchange networks that hypothesize specific metabolites that support growth of different species. Collectively, these results demonstrate how a hierarchical structure is formed within microbial communities by radiating cross-feeding networks. In principle, this cross-feeding structure can be generalized to the multitude of communities that require specialized biopolymer degraders, including those involved in human health ( 1 ), biogeochemical transformations ( 2 ), and biotechnology ( 40 ). Although hierarchical structures may be convoluted in more complex environments, we outline a concept for generating hypotheses on cross-feeding networks when constructing smaller consortia that recapitulate key aspect of the entire community."
} | 1,412 |
26941935 | PMC4761763 | pmc | 7,016 | {
"abstract": "Summary Many factors affect the presence and exchange of individuals among subpopulations and influence not only the emergence, but the strength of ensuing source–sink dynamics within metapopulations. Yet their relative contributions remain largely unexplored. To help identify the characteristics of empirical systems that are likely to exhibit strong versus weak source–sink dynamics and inform their differential management, we compared the relative roles of influential factors in strengthening source–sink dynamics. In a series of controlled experiments within a spatially explicit individual‐based model framework, we varied patch quality, patch size, the dispersion of high‐ and low‐quality patches, population growth rates, dispersal distances, and environmental stochasticity in a factorial design. We then recorded source–sink dynamics that emerged from the simulated habitat and population factors. Long‐term differences in births and deaths were quantified for sources and sinks in each system and used in a statistical model to rank the influences of key factors. Our results suggest that systems with species capable of rapid growth, occupying habitat patches with more disparate qualities, with interspersed higher‐ and lower‐quality habitats, and that experience relatively stable environments (i.e., fewer negative perturbations) are more likely to exhibit strong source–sink dynamics. Strong source–sink dynamics emerged under diverse combinations of factors, suggesting that simple inferences of process from pattern will likely be inadequate to predict and assess the strength of source–sink dynamics. Our results also suggest that it may be more difficult to detect and accurately measure source–sink dynamics in slow‐growing populations, highly variable environments, and where a subtle gradient of habitat quality exists.",
"introduction": "Introduction Spatial variation in habitat quality is the basic factor that structures source–sink dynamics in heterogeneous landscapes (Pulliam 1988 ; Dias 1996 ). Demographic surpluses in higher‐quality habitats (e.g., sources) and deficits in lower‐quality habitats (e.g., sinks) commonly arise, and movement among local populations can stabilize dynamics at regional scales (Dias 1996 ). At steady state, some local populations become the net exporters of individuals (i.e., sources) where births outweigh deaths ( b > d ) and emigrants outnumber immigrants ( e > i ), and other populations become net importers (i.e., sinks) where the opposite demographic and movement conditions hold (Pulliam 1988 ). Although differences in habitat quality (see Hall et al. 1997 ) provide a basis for source–sink dynamics to emerge, other habitat and population characteristics can affect reproduction and survival through space and time, and play a role in strengthening or weakening source–sink dynamics (Dunning et al. 1992 ). Habitat characteristics including differences in patch sizes and qualities, as well as the proximity of high‐quality to low‐quality habitats, have the potential to influence the strength of sources and sinks. Similarly, species and population factors such as growth rates, dispersal abilities, and demographic responses to environmental variability can affect the severity of source–sink dynamics, driving sources, and sinks to become more or less extreme. Although these habitat and population factors have been individually found to affect source–sink dynamics, their relative importance is not well understood. Populations are increasingly conceptualized and managed based on their source–sink status or the suspected presence of source–sink dynamics within the system. Hence, there is a clear practical need to be able to distinguish among source and sink populations. Differential management of sources and sinks can be particularly important in avoiding the counterproductive actions associated with falsely assuming that an animal's realized niche in a sink habitat represents their fundamental niche (Pulliam and Danielson 1991 ; Boughton 2000 ). Further, the large continuum of source–sink strengths ranging from minor asymmetries to overwhelming directional flows of individuals suggests that an understanding of the strength of the system and the factors that augment or diminish its strength has the potential to guide effective decisions and actions. Yet to date, studies have primarily evaluated the conditions under which source–sink dynamics are incited, with the limited evaluations of the factors that strengthen dynamics once incited. Source–sink literature points to dispersal and habitat selection behavior as providing the basis for emergent source–sink dynamics in heterogeneous landscapes, directing the flow of individuals among habitats and resulting birth, death rates, and local densities. Source–sink dynamics can arise as a result of random dispersal (e.g., diffusion), as well as with passive dispersal mechanisms (e.g., exchange of a fixed or stable proportion of dispersers among asymmetrically sized populations; Boughton 2000 ). Ideal preemptive habitat selection behavior can also lead to source–sink dynamics, with animals that arrive first preempting the use of the best sites, maximizing their reproductive output and fitness. As population density increases, late arrivals are forced to settle in lower‐quality habitats, potentially leading to lower reproductive success, higher mortality, and the creation of population sinks (Pulliam and Danielson 1991 ). In empirical populations, realistic animal dispersal and habitat selection can incorporate all these elements into complex decision processes that influence ensuing source–sink outcomes. We combine these forces to incite source–sink dynamics in a range of landscapes and life history characteristics, using a novel approach that incorporates mechanistic density‐dependent habitat selection, the concepts of passive diffusion in emigrating from disparate‐sized habitat patches, diffusion in the form of quasirandom walks through matrix to find habitat, and ideal preemption in excluding latecomers from the best sites. To compare the relative influences of habitat and population factors on the strength of emergent source–sink systems, we simulated metapopulation dynamics in a range of controlled landscapes, using realistic movement and territorial animal behavior within the context of a spatially explicit individual‐based population model. We used a set of hypothetical landscapes and species and six potential drivers (population growth rates, differential patch qualities and sizes, patch quality patterns, dispersal distances, and levels of environmental variation), in a factorial design to rank the influence of the different factors. To assess the consistency of landscape and population drivers of source–sink strength, we also examined their relative influence among alternative habitat selection scenarios with varying degrees of awareness of the landscape, habitat quality, and abilities to optimize fitness. Many animals occupy sink habitats, resulting from an inability to discern adverse fitness consequences (i.e., an ecological trap; Howe et al. 1991 ; Battin 2004 ), an unwillingness to emigrate elsewhere (e.g., strong site fidelities), or an improbability of surviving to the next opportunity to relocate (e.g., high overwinter mortality for kangaroo rats; Holt and Gaines 1992 ; Heinrichs et al. 2010 ). With limited knowledge of the landscape, we expected animals that are better able to detect habitat quality to select fitness‐optimizing territories to a greater extent than less discerning animals, producing weaker source–sink systems (and the converse to have stronger dynamics). We expected the strength of source–sink systems to depend on habitat and populations factors, influencing local population densities and density‐dependent emigration, and to be driven by differences in habitat quality. As patch occupancy is a necessary precondition for local habitat quality to be important, factors increasing th size of potential colonizing populations were expected to strengthen source–sink dynamics. The ability for local populations to grow, reach carrying capacities, and induce density‐dependent emigration and fitness consequences was expected to be a key driver of source–sink strength. We also expected that systems that were not challenged by periodic population depressions would produce stronger sources and sinks than those affected by perturbations. We expected that greater accessibility and propensity of individuals in sources to diffuse to sinks would strengthen source–sink dynamics. At stable state, we expected asymmetric patch sizes to strengthen system dynamics through passive dispersal mechanisms. Lastly, Species with longer dispersal distances relative to their territory size were expected to select better territories and maximize their fitness to a greater degree than shorter dispersers, weakening source–sink dynamics. Although we do not directly derive hypotheses for empirical populations, these general expectations lend themselves to thorough testing with empirical data. If supported, these hypotheses could be used to indicate systems with a high degree of correlation and dependency among populations, the degree to which density‐dependent regulation may be influencing population outcomes, and where source–sink dynamics may require additional effort to detect and quantify.",
"discussion": "Discussion Relative influence Many factors have been found to incite source–sink dynamics, yet their relative contributions in strengthening dynamics remained relatively unexplored. In an effort to help identify the characteristics of systems that are likely to exhibit strong source–sink dynamics, we compared the roles of influential factors in strengthening source–sink dynamics in a series of controlled simulation experiments to generate hypotheses for future exploration with empirical data. Our results suggest that systems with species capable of rapid growth, occupying habitat patches with more disparate qualities among patches, and in relatively stable environments (i.e., fewer negative perturbations), are more likely to exhibit strong source–sink dynamics. The pattern of high‐ and low‐quality habitats was also influential in inciting and strengthening dynamics. Dispersal ability and differences in patch sizes had weaker impacts. Although some factors were much stronger drivers than others, strong source–sink dynamics emerged under a range of factors in multiple different scenarios, suggesting that multiple lines of data and inference are likely to be needed to predict the strength of source–sink dynamics. In both the naive and informed habitat selection scenarios, the strongest driver of source–sink dynamics was population growth (i.e., fecundity). Although the intuitive expectation is that differential habitat quality drives source–sink dynamics, our simulations suggest that heterogeneous habitat quality provides an influential stage upon which density‐regulated populations can act to further strengthen source–sink dynamics. Smaller, slow‐growing populations operate more frequently below carrying capacity, where there is less impetus for density‐dependent emigration from natal patches and less exposure to survival and reproductive limitations. This weakened sources and sinks over populations are capable of faster slower growth in our simulations. Although stronger source–sink dynamics were associated with higher population sizes, this relationship was increasingly variable in regional populations with stronger dynamics, indicating that that population size alone is unlikely to generally predict source–sink severity. The combination of population growth potential and population size relative to carrying capacities may better predict the strength of source–sink dynamics. As demographic rates are themselves a function of inherent species characteristics, as well as current and past environmental drivers (Treurnicht et al. 2016 ), future research could decompose population growth in specific systems to gain a more complete indication of the key demographic drivers influencing source–sink dynamics. Our results indicate that systems with greater disparity in habitat quality among patches are likely to incite and produce stronger source–sink dynamics, with the strongest dynamics emerging in systems with both high quality disparity and population growth. Species with inherently lower capacities for population growth (e.g., lower fecundity rates) with a naive habitat selection strategy may be particularly sensitive to the disparity of habitat quality among patches (e.g., growth 0.75 in Fig. 5 ), and the ensuing source–sink dynamics may be largely driven by habitat quality differences. Species subject to greater intensities and durations of density‐dependant population regulation (e.g., growth = 1.75 in Fig. 5 ) were still primarily influenced by habitat quality disparity, but variation, pattern, and dispersal factors weighted in more heavily than in low‐growth scenarios. Across different kinds of systems, habitat quality information is likely to be helpful in inferring the presence of source–sink dynamics; however, this information alone is unlikely to accurately predict the strength of dynamics. Habitat quality disparity may better indicate the strength of source–sink dynamics among populations of the same species or groups of species with a similar low potential for population growth. Response surfaces indicate that threshold effects may exist (e.g., low quality = ~40% of the value of high quality), above which quality disparity among patches may produce much stronger source–sink dynamics. Such thresholds may complicate source–sink strength inferences based on quality disparity data. Habitat quality disparity may also be more important than described here for species who are unable to expand their territories to compensate for a low quality or quantity of resources, or philopatric social organisms that share resources (e.g., scramble competition) with an unwillingness to relocate or break up the group. Environmental stability and the spatial pattern of sources and sinks were also important in predicting the strength of sources–sink dynamics. When used in combination with population growth and quality disparity information, environmental stability may indicate the presence and strength of source–sink dynamics, particularly for species with the informed habitat selection. Temporal fluctuations in demographic rates can influence the magnitude and duration of density‐dependent effects (e.g., alteration of dispersal patterns; Holt 1996 ; Virgl and Messier 2000 ) and influence the strength of source–sink dynamics. In our model, populations subject to weak environmental variation experienced only minor reductions in abundance and maintained higher average population sizes than populations more strongly affected by variation. Through time, higher‐quality patches were strengthened as a result of more consistent occupancy and higher local population densities. Emigration from highly occupied patches to lesser quality patches strengthened sinks and created greater disparity in productivity among sources and sinks. For species with the informed habitat selection, population stability leading to density‐dependent movement was more important than habitat quality disparity. With the recognition of lower‐quality habitat, the ability to leave in search of better habitat with higher dispersal success, informed species could seek out locations that better optimize their fitness, predictably weakening the impact of habitat quality disparity. As our examination was restricted to periodic stressors or catastrophes, future studies could invoke different kinds of environmental stochasticity affecting population dynamics including positive autocorrelation (e.g., Crone 2016 ) to further assess the sensitivity of source‐sink strength to environmental variability. The likelihood that individuals leaving sources can successfully reach a sink is generally expected to affect the strength of sources and sinks (Pulliam 1988 ; Walters 2001 ; Holland et al. 2009 ). Sinks that are proximate to sources (e.g., interspersed) are more likely to be encountered and occupied by emigrants from sources, strengthening nearby sinks and their contribution to overall dynamics. Conversely, clustered sinks (e.g., low‐quality patches along a habitat quality gradient) can result in lower population persistence (Matthews and Gonzalez 2007 ), indicating that lower population sizes and high local extinction rates may reduce the long‐term severity of sinks and diminish the overall strength of source–sink dynamics. The pattern of high‐ and low‐quality patches modified the strength of source–sink dynamics in all scenarios, with interspersed quality patches strengthening, and gradients weakening source–sink dynamics similarly in both naive and informed selection scenarios. We found limited support for dispersal ability mediating the effect of spatial patterning of habitat quality on the strength of source–sink dynamics. In gradient landscapes, naive species with unlimited dispersal had weaker source–sink dynamics compared to dispersal‐limited species. However, in interspersed landscapes, source–sink severity was similar among short and long dispersers. Similarly, Walters ( 2007 ) found that the effects of breeding patch configuration outweighed dispersal characteristics (including distance) on dispersal success, suggesting that the influence of source–sink patterning may generally outweigh that of dispersal in strengthening sources and sinks. The exchange of individuals among populations is a necessary condition for source–sink dynamics to arise, as the existence of sink populations often relies on immigration from sources (Gunderson et al. 2001 ; Schooley and Branch 2007 ). Hence, the increased ability of animals to travel across the landscape in search of optimal habitat (relative to interpatch distances) was expected and observed to weaken source–sink dynamics. In the naive habitat selection scenario, animals with short dispersal distances (relative to interpatch distances) were limited to settling in nearby patches regardless of their quality, strengthening proximate sinks and source–sink dynamics. Animals capable of longer distance dispersal (but similarly the limited perceptual ranges) were able to migrate to distant, more isolated patches, increasing their immigration and occupancy rates, and weakening the effect of source–sink pattern on the severity of dynamics. In our limited exploration of dispersal, an animal's ability to search the landscape for new habitat was not a key factor driving the strength of source–sink dynamics. However, dispersal ability might be more important in species that rely more on random explorative searches for habitat, with long dispersal abilities coupled with higher habitat selection criteria than those examined here, and in landscapes with a complex, heterogeneous matrix. All else being equal, we expected that landscapes with disparate patch sizes would have stronger sources than those with similar patch sizes. Larger patches have greater capacities, receive more immigrants via diffusion movements in the matrix (i.e., higher encounter rates; (Bowman et al. 2002 ), and emit fewer emigrants resulting from passive dispersal, which can strengthen source patches (Walters 2001 ). In turn, stronger sources would be expected to create stronger sinks and source–sink dynamics by increasing sink immigration and occupancy rates. Although more disparate patch sizes generally strengthened source–sink dynamics , patch size disparity was not a prominent driver of source–sink strength, owing in part to the effects of other factors on emigration routes and rates. Patch size disparities may be more important in species that rely more on passive dispersal and random diffusion than explored here, and in systems with strong responses to patch edges. Implications for source–sink populations Hypothetical source–sink systems provide a controlled means of gauging the relative influences of a number of general ecological conditions and can serve as a tool for generating testable hypotheses. These models lack the context‐specific details of complex natural systems, yet in an informal evaluation of modeled source–sink dynamics for Black‐capped vireos and Ord's kangaroo rats based on Heinrichs et al. ( 2015 ), we found support for population growth rates indicating the strength of source–sink dynamics across systems. Heinrichs et al. incorporated species' life history details (demography, movement, density‐dependent habitat selection) and landscape information into realistic spatially explicit individual‐based models and then computed habitat patch productivity as has been carried out here. These population growth rates successfully predicted the rank order of the overall strength of source–sink dynamics across these case study systems and scenarios. Despite this, the degree to which the strength of source–sink dynamics can be predicted in individual systems may still depend on the importance of case‐specific details and dependencies (e.g., Loehle 2012 ); hence, empirical data are required to further test and develop these hypotheses. Results from theoretical source–sink research are often criticized for being difficult to operationalize in empirical systems, and management based on demographic and source–sink concepts is often constrained by practical constraints (e.g., Kerr et al. 2016 ), including the costs of intensive data collection. Yet, uncertainty in demographic conditions, source–sink characterizations, and the strength of source–sink systems can undermine the management efforts (Barthold et al. 2016 ; Griffith et al. 2016 ). A conceptual understanding of the nature of dynamics among subpopulations could be helpful in guiding and targeting the intensive resources required to collect data to assess and evaluate cross‐system patterns of source–sink intensities. Well‐developed and tested theory predicting the expected strength of source–sink systems can provide a low effort screening tool to identify situations in which source–sink analyses should be undertaken and used to inform management strategies. Conservation and management actions may need to be different for systems with weak versus strong source–sink dynamics, and approaches and decisions made for systems with weak dynamics may not hold for those with strong dynamics. Knowledge of the strength of source–sink dynamics present within a system should be also helpful in indicating the degree of interdependency and the importance of connectivity among populations, and in identifying actions that could be used to alter the severity of source–sink dynamics, particularly for declining species. For instance, systems with particularly strong sources and/or sinks may have patches that disproportionately contribute to and drive regional population dynamics (Schlaepfer et al. 2002 ; Kawecki 2004 ; Runge et al. 2006 ). In such metapopulations, it may be particularly important to accurately identify and assess the strengths and contributions of sources and sinks prior to the selection of local habitats for protection, restoration, or monitoring population trajectories. For example, the strongest and most central source ( b – d = 1891) in Figure 3 A drives the performance of all of its neighbors and would likely be a primary target for preservation. Weaker sources or sinks may be targets for habitat restoration, and strong sinks (e.g., b – d = −585 in Fig. 3 A) may also be particularly suitable sites for monitoring changes in source reproductive output (Jonzen et al. 2005 ) or targeted for habitat removal. Lastly, generalizations about the factors influencing source–sink severity could be helpful in identifying systems wherein source–sink dynamics may be difficult to detect and where local source–sink identifications might prove difficult or require increased accuracy. Lesser differences in productivity among sources and sinks are expected in systems with weaker source–sink dynamics, making the status of local populations more difficult to identify with confidence, particularly given the difficulty in collecting demographic information and the uncertainty inherent in such data (Runge et al. 2006 ; Johnson 2007 ; Robinson and Hoover 2011 ). Our results suggest that demographic differences among subpopulations may be easier to detect and measure in populations that are not continually challenged by stochastic events, capable of rapid growth, and that inhabit heterogeneous quality landscapes with interspersed high‐ and low‐quality patches. Conversely, it should be more difficult to detect and measure source–sink dynamics in slow‐growing populations, highly variable environments, and where a subtle gradient of habitat quality exists. This suggests that if all else is equal, populations wherein reproduction is consistently (intrinsically or extrinsically) suppressed, and populations subject to significant periodic survival stressors (e.g., weather events, exposure to toxins, disease, interspecific interactions), are less likely to exhibit large differences in demographic measurements. Similarly, populations inhabiting landscapes with gradations in habitat quality (e.g., mirroring the transition of underlying vegetation or geologic conditions) are expected to have weaker sources and sinks. In weak source–sink systems, data collection may need to be more comprehensive to detect differences in productivity among local populations and determine their meaning and relevancy for habitat and population management."
} | 6,441 |
33537471 | PMC7840857 | pmc | 7,018 | {
"abstract": "Poly(3-hydroxybutyrate) (PHB) belongs to the family of polyhydroxyalkanoates, biopolymers used for agricultural, industrial, or even medical applications. However, scaling up the production is still an issue due to the myriad of parameters involved in the fermentation processes. The present work seeks, firstly, to scale up poly(3-hydroxybutyrate) (PHB) production by wild type C. necator ATCC 17697 from shaken flasks to a stirred-tank bioreactor with the optimized media and fructose as carbon source. The second purpose is to improve the production of PHB by applying both the batch and fed-batch fermentation strategies in comparison with previous works of wild type C. necator with fructose. Furthermore, thinking of biomedical applications, physicochemical, and cytotoxicity analyses of the produced biopolymer, are presented. Fed-batch fermentation with an exponential feeding strategy enabled us to achieve the highest values of PHB concentration and productivity, 25.7 g/l and 0.43 g/(l h), respectively. The PHB productivity was 3.3 and 7.2 times higher than the one in batch strategy and shaken flask cultures, respectively. DSC, FTIR, 1 H, and 13 C NMR analysis led to determine that the biopolymer produced by C. necator ATCC 17697 has a molecular structure and characteristics in agreement with the commercial PHB. Additionally, the biopolymer does not induce cytotoxic effects on the NIH/3T3 cell culture. Due to the improved fermentation strategies, PHB concentration resulted in 40 % higher of the already reported one for wild type C. necator using other fed-batch modes and fructose as a carbon source. Thus the produced PHB could be attractive for biomedical applications, which generate a rising interest in polyhydroxyalkanoates during recent years.",
"conclusion": "4 Conclusions This work presents the PHB production by fermentation of wild type C. necator ATCC 17697 in a stirred-tank bioreactor. Different fermentation strategies were tested and the experimental results were compared to our previously achieved using shaken flask cultures. Variation in fermentation conditions has been explored to increase the parameters production of PHB. The sequential presentation allows assessing the effects of improvements in the batch and fed-batch fermentation strategies. Firstly, it was necessary to increase the scale from 250 ml Erlenmeyer to a 5 l bioreactor with the culture medium previously optimized in shaken flasks. For that purpose, in batch mode fermentations several parameters were adjusted, such as aeration, agitation mode, and concentrations of carbon and nitrogen sources. In batch mode, a slight improvement on biomass, PHB productivity, and PHB concentration was achieved. Secondly, fed-batch fermentation with an exponential feeding strategy enabled us to achieve the highest values of PHB concentration and productivity. The PHB productivity obtained by this fed-batch fermentation strategy was 3.3 and 7.2 fold higher than in batch strategy and shaken flask cultures and 40% higher than reported for wild type C. necator strains. Furthermore, the purified polymer was characterized by DSC, FTIR, H 1 and C 13 NMR techniques; the structure and characteristics of the biopolymer produced by C. necator ATCC 17697 correspond to the PHB. Finally, it was confirmed that the polymer does not induce cytotoxic effects on the NIH/3T3 cell culture which is one of the features to be fulfilled for biomedical application. Thus, the following steps of this work will intend to scale up the production of this biopolymer to produce filaments for 3D printing of resorbable scaffolds tailored to the patient.",
"introduction": "1 Introduction Currently, polyhydroxyalkanoates (PHAs) are one of the most researched bioplastics because they are biodegradable, renewable, biocompatible and environmentally friendly [ 1 ]. They have similar physical characteristics (including molecular mass, brittleness, melting point, and glass transition temperature) to that of synthetic petrochemical polymers such as polypropylene [ 2 ]. These characteristics, together with the worldwide problem of the depletion of fossil fuels, place PHAs as potential substitutes for conventional plastics, the petroleum-based polymers, especially in short-lived industrial applications [ 3 ]. In addition, these polymers have a wide field of use in medicine due to their innocuousness [ 4 ]. Therefore, PHAs have immense industrial potential, as already shown by applications in tissue engineering, drug delivery, and packaging [ 5 ]. However, one drawback of PHAs is that the production cost is still not competitive with the conventional polymers. So, the target of the PHAs production focused on advanced medical applications and products for tissue engineering. In fact, applications such as polymer-based devices for controlled drug delivery and hormone release or 3D printed resorbable scaffolds for tissue regeneration [ 6 ] require biodegradable polymers, so that degradation products are not harmful to the body. Several PHAs fulfill these issues [ 7 , 8 ] since their monomeric and oligomeric in vivo degradation products have no deleterious effect on living cells or tissues [ 9 , 10 ]. Poly(3-hydroxybutyrate) (PHB) is the only homopolymer of this large family, produced on an industrial scale and also the most studied [ 11 ]. More than 300 different microorganisms including Eubacteria ( Pseudomonas sp ., Ralstonia sp ., Bacillus sp ., Vibrio sp ., Azotobacter sp. , Methylobacterium sp., Burkholderia sp.) and archaea (e.g. Haloarchaea ), intracellularly produce PHA granules [ 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 ], as storage of carbon and energy under limiting conditions of essential nutrients such as nitrogen or phosphate and in the presence of an excess of carbon source [ 13 , 15 , 20 ]. Cupriavidus necator (formerly known as Wautersia eutropha, Ralstonia eutropha, and Alcaligenes eutrophus ) is the most promising producer of PHB due a remarkable capacity for accumulating PHB up to 90 % of the cellular dry weight using a wide range of substrates in a heterotrophic or autotrophic pathway [ 21 , 22 ]. Several carbon sources have been tested as substrate for the PHB production by wild type C. necator : fructose, glucose, lactic acid, xylose, sucrose, molasses, sorbose, acetic acid, starch, sodium acetate, glycerol, lactose, propionic acid and different lignocellulosic biomass hydrolysates; fructose is the one that allows the highest PHB productivity [ 23 , 24 ]. Recently the advantage of using fructose has been demonstrated in certain bacteria: Halomonas sp . and Bacillus sp, prefers fructose over other sources; in some cases increasing its size by using this carbon source [ 14 , 25 ]. To enhance the fermentation process from Cupriavidus necator it is necessary to increase the cell concentration and the intracellular PHB accumulation. One way to improve the PHB production is by tuning the growing medium and the operating conditions [ 26 , 27 ]. Since the specific growth rate for the PHB production could be inhibited by the substrate concentration, the fed-batch fermentation is an useful approach to improve the PHB productivity [ 28 ]. Different feeding sources with a limiting nutrient were used to improve the cell growth and the PHB-productivity. Furthermore, the aerobic dynamic feeding by pulses of the carbon source enables the increase of the PHB content [ 29 , 30 , 31 ]. A successful strategy is the three-stage fermentation: The first stage is a batch culture for adaptation of the bacteria, the second is a fed-batch culture where the total biomass further increases and the third stage under nitrogen limitation in the medium allows to increase the PHB accumulation [ 32 ]. Moreover, different feeding policies were used to improve PHB production in fed-batch cultivation [ 26 , 33 , 34 ]. Most researchers use the C. necator ATCC 17699 strain for the PHAs production using various carbon substrates and fermentation strategies [ 22 , 32 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 ], however C. necator ATCC 17697 have been studied only by one research group [ 23 , 26 , 47 , 48 , 49 ], so it is important to extend its research. Regarding this goal, Khanna and Srivastava have been studied the PHB production by C. necator using different fructose feeding strategies with limited nitrogen source [ 23 , 26 , 47 , 48 , 49 , 50 ]. They produced PHB in a batch system using ATCC 17697 strain, where the dissolved oxygen concentration was maintained at 30 % saturation by manually adjusting the agitation speed and/or airflow rate [ 47 , 50 ]. Another model based on fed-batch fermentation with nitrogen and fructose feeding by applying different constant rates, at different times, was also successfully performed [ 26 , 48 , 49 ]. Through these fed-batch strategies, the maximum concentration of PHB in the culture medium was 18.6 g/l. Nevertheless, in accordance to Blunt et al. [ 51 ] it is possible to further increase the PHB concentration by applying other fermentation strategies, particularly a two-step fermentation: first in batch mode to adapt the microorganism to the medium and then a fed-batch to maximize cell density at a high rate. In our previous work, full factorial design and response surface analysis were carried out to optimize media parameters including the concentration of fructose, ammonium sulfate, potassium dihydrogen phosphate, microelement solution, and initial pH to increase PHB production by shaking cultures of wild type C. necator ATCC 17697 [ 52 , 53 ]. The present work seeks, firstly, to scale up the PHB production from shaken flasks to a stirred-tank bioreactor with the optimized media and fructose. Secondly, the aim is to improve the production of PHB both by batch and fed-batch fermentation strategies in comparison with previous works of wild type C. necator with fructose as carbon source. Furthermore, thinking of biomedical applications, physicochemical and cytotoxicity analyses of the produced biopolymer are presented.",
"discussion": "3 Results and discussion 3.1 Development of the fermentation process to enhance PHB production 3.1.1 Enhanced batch production of PHB through the fermenter Two batch fermentations of C. necator ATCC 17697 were performed to determine the time evolution of total and residual biomass, fructose and ammonium consumption and PHB production. Figure 1 A shows the parameters profile for a basal medium with 20 g/l fructose and 32 h incubation (Batch 1). Figure 1 Batch fermentation profiles of C. necator ATCC 17697 in culture medium with (A) 20 g/l fructose and 1.5 g/l ammonium sulfate for 32 h of incubation and (B) 40 g/l fructose and 3 g/l ammonium sulfate for 76 h of incubation. Figure 1 PHB accumulation begins in the initial exponential phase and continues to increase until culture reaches the stationary phase. Fructose and ammonium consumptions correlate with bacterial growth and consequent polymer accumulation. Both cell growth and PHB accumulation are maximum at 31 h, 7.93 g/l, and 3.35 g/l, respectively. At this time, the PHB volumetric productivity (P PHB ) is 0.11 g/(l h), the ammonium sulfate was already exhausted and fructose was consumed completely. In this batch fermentation, the biomass concentration and polymer productivity were significantly higher, compared to the values obtained in our previous work [ 52 ] in shaken flask cultures during 72 h: 6.5 g/l and 0.06 g/(l h), respectively ( Table 1 ). However, due to the depletion of the carbon source in the culture medium, the accumulation of PHB was stopped in the early stage of the stationary phase, without reaching higher values. The best medium chosen for shaken flask cultures is not always the best medium for bioreactor fermentations. This result has already been pointed out by Kennedy and Krouse [ 59 ]. Table 1 Summary of PHB production and yield and its comparison with previous fermentation from fructose reported for C. necator ATCC 17697. Table 1 Fermentation strategy Biomass PHB PHB P PHB Reference g/l g/l % g/(l h) Shaken flasks 6.5 4.6 71 0.06 [ 52 ] Batch 1 7.9 3.4 42 0.11 This work Batch 2 14.4 9.9 69 0.13 Fed-batch 1 35.5 17.5 48 0.25 Fed-batch 2 50.8 25.7 51 0.43 Batch 19.7–20.7 9.3–10.9 45–55 0.16–0.18 [ 23 , 47 ] Fed-batch 32–36 14–18.6 44–53 0.28–0.48 [ 26 , 48 , 49 ] By this batch fermentation (Batch 1), higher biomass level was attained in less than half time compared to results obtained in our previous work using shaken flasks culture [ 52 ]; however, the PHB accumulation process stopped due to fructose depletion. Therefore, to optimize the production process in a bioreactor, a second batch fermentation (Batch 2) was carried out in a culture medium, doubling the concentrations of carbon and nitrogen sources ( Figure 1 B). In this fermentation, the exponential cell growth phase began after a lag phase of around 8 h. At 25 h all the ammonium sulfate in the culture medium was consumed, triggering PHB synthesis in the last phase of the fermentation process. The exponential growth finished after 33 h with a production of 10 g/l of biomass, where 4.13 g/l corresponds to the amount of accumulated polymer. Then, the residual biomass remained constant, while the total biomass increased exclusively by the intracellular accumulation of PHB. As a result, after 76 h of this batch fermentation 14.4 g/l of biomass and 9.90 g/l of PHB concentration were obtained, which represents a P PHB of 0.13 g/(l h) ( Table 1 ). These results were consistent with those reported by Khanna and Srivastava [ 23 ], showing a slight improvement in PHB production. Therefore, the second batch fermentation led to an increase of 10 % the PHB productivity and two and three times the biomass and PHB concentration, respectively, comparing to our results from shaken flask cultures. Since PHB productivity and concentration were not very high, the trial of fed-batch fermentation will be considered in the next section. 3.1.2 Improved fed-batch fermentation to enhance PHB production Two fed-batch fermentations of C. necator ATCC 17697 were performed: one by fructose feeding regulated with dO 2 level ( Figure 2 A) and the other by exponential fructose feeding ( Figure 2 B). Figure 2 Fed-batch fermentation profiles of C. necator ATCC 17697 using fructose feeding regulated with dO2 level (A) and exponential fructose feeding (B). The arrows indicate the beginning of each stage. Figure 2 The first fed-batch fermentation (Fed-batch 1) was developed using a three-stage production system, as shown in Figure 2 A. The first stage was a batch culture for adaptation and biomass production. During 10 h of this stage corresponding to the lag phase of bacterial growth, the dO 2 remained with a value higher than 90 %. Then, the dO 2 began to decrease due to the exponential bacterial growth as well as the fructose concentration, which is completely depleted after 20 h. After this batch phase, total biomass and PHB concentrations were 8.0 and 1.6 g/l, respectively; this means that the carbon source was used mainly for biomass production. The second stage was a fed-batch culture with fructose feeding regulated with dO 2 level, in which the total biomass increased along with the PHB accumulation. The ammonium concentration, used to regulate pH and as a source of nitrogen remained relatively constant; thus, their supply and consumption rates were equal. Biomass and PHB concentrations increased to 30 g/l and 13.7 g/l, respectively, while the residual biomass reached a value of 16.3 g/l. The third stage consisted of fed-batch culture with fructose feeding and without nitrogen supply, where only PHB production occurred. During the first 6 h, the ammonium concentration remained almost constant and then decreased from 1.4 g/l to 0.1 g/l; this condition of excess carbon is ideal to produce PHA [ 32 , 47 ]. Throughout this stage, the residual biomass remained constant. At the end of this stage, 35.5 g/l of biomass and 17.5 g/l of PHB were achieved, with PHB productivity of 0.25 g/(l h) ( Table 1 ). These results are slightly higher than those obtained by Khanna and Srivastava with the same strain but with a constant flow feeding strategy of 100 ml/h of 360 g/l fructose [ 26 ]. Therefore, using this fed-batch fermentation strategy it was possible to increase the PHB production in more than 3.5 and 1.5 fold, in comparison with shaken flask cultures and Batch 2 fermentation, respectively ( Table 1 ). The second fed-batch fermentation strategy (Fed-batch 2) was developed using exponential feeding to allow cells to grow at a constant specific growth rate to achieve high cell density in a short period [ 60 , 61 ] and thus increase the amount of catalytic biomass. After 17 h of the first batch stage, a biomass concentration of 8.17 g/l and PHB production of 0.30 g/l have been reached. In the second stage with exponential feeding from 17 h to 28 h, the bioreactor software determined the feeding flow rate according to Eq. (1) , with V 0 = 2.2 l and X 0 = 7.5 g/l. Thus, the biomass raised to 32.2 g/l with a PHB content of 6.64 g/l ( Figure 2 B). The main difference between the two fed-batch strategies in Stage 2 is the residual biomass concentration: 25.7 g/l with exponential fructose feeding, and only 16.4 g/l for fructose feeding with dO 2 regulation. Also, due to the reduction of the second stage to less than half the time, the productivity of this phase highly improved. Therefore, the third stage of this fed-batch started earlier with more catalytic biomass for the biopolymer accumulation. In the third stage —where only PHB accumulation takes place— the residual biomass remained constant and the total biomass reached its maximum value of 50.8 g/l with a PHB content of 25.7 g/l. The maximum volumetric productivity of this fed-batch process was 0.43 g/(l h) after 60 h of fermentation ( Table 1 ). By using the fed-batch fermentation with an exponential feeding strategy, we attained a PHB level (in g/l) 40% higher than that obtained by Khanna and Srivastava, who performed fed-batch fermentations using constant fructose feeding strategies [ 26 , 48 , 49 ]. Specific growth rates were computed by linear regression of natural log biomass concentration versus time; the slope of this line estimated the specific growth rates in the exponential phase (μ 1 ) and the stationary phase (μ 2 ) ( Table 2 ). Table 2 Comparison of specific growth rates in the exponential phase (μ 1 ) and the stationary phase (μ 2 ) for the indicated fermentation strategies using C. necator ATCC 17697. Table 2 Fermentation strategy μ 1 μ 2 h −1 h −1 Shaken flasks 0.115 0.006 Batch 1 0.118 0.004 Batch 2 0.114 0.011 Fed-batch 1 0.101 0.013 Fed-batch 2 0.119 0.011 Error ≤10 % Cupriavidus necator is a model organism which has a strong ability to produce PHB in a non-growth-associated manner [ 62 ]. μ1 was determined in the growth-associated PHB production phase; its value is mainly due to the growth of biomass. μ2 was determined in non-growth-associated PHB production phase; this second rate is due exclusively to PHB production since the residual biomass remains constant. The occurrence of some growth-associated PHB production besides non growth-associated PHB production was demonstrated, although it is inhibited in the presence of nitrogen [ 63 ]. μ 1 calculated for batch and fed-batch fermentations coincide with the value measured in shaken flask cultures, 0.115 h −1 ( Table 2 ) and is similar to those obtained in fed-batch feeding of C. necator DSM 545 with glucose [ 64 ]. μ 2 values calculated for shaken flasks were 0.006 h −1 which represents about 5 % of the maximum specific growth rate of this microorganism and could increase in large-scale fermentation where PHB production is higher. This feature, confirmed by comparing the μ 2 values obtained at different fermentation scales, as shown in Table 1 . Therefore, the residual biomass remains constant but the total biomass increased with the amount of accumulated biopolymer. In the case of batch 1 fermentation, the μ 2 value was lower than that calculated in shaken flask cultures, since PHB accumulation stopped due to the depletion of the substrate in the culture medium. 3.2 Biopolymer characterization Biopolymer samples produced by fermentation of C. necator ATCC 17697 were analyzed by differential scanning calorimetry, ATR-FTIR, 13 C and 1 H NMR spectroscopy, and indirect cytotoxicity testing. Figure 3 shows the polymer physicochemical characterization. The main properties of the produced PHB and their comparison with values from the literature are presented in Table 3 . Figure 3 Polymer physicochemical characterization. DSC thermogram for PHB produced by C. necator ATCC 17697 (A). FTIR-ATR spectra of PHB produced by C. necator ATCC 17697 a) and commercially available b) (B). 13C NMR spectrum (C) and 1H NMR spectrum (D) of PHB produced by C. necator ATCC 17697. Figure 3 Table 3 NMR and DSC characterization of PHB produced by C. necator ATCC 17697 and comparison with PHB produced by other microorganisms. Table 3 This work Ref. [ 64 ] Ref. [ 66 ] Ref. [ 65 ] Ref. [ 67 ] Bacterial strain Cupriavidus necator Cupriavidus necator Cyanobacteria Bacillus cereus Cupriavidus necator Bacillus megaterium Standard PHB ATCC 17697 DSM 545 spp. SPV MTCC 8320 MTCC 453 Carbon source fructose glucose glucose glucose fructose fructose Extraction method solvent solvent solvent solvent solvent and ultrasonication solvent and ultrasonication DSC analysis Tg (°C) 3.5 6 6 2 6 6 –8 Tm (°C) 165.4 180 171 169.7 175 176 176 Tc (°C) 54.3 - 78.8 - 84 104 90 Xc (%) 56.0 64.6 56.8 57.7 44 23 - NMR spectra (chemical shift in ppm) 13 C NMR spectrum -CH 3 19.9 - - 21.2 19.95 19.95 19.95 -CH 2 - 40.1 - - 42.7 40.99 31.09 40.99 31.09 40.99 31.09 -CH- 67.7 - - 68.5 67.81 67.80 67.81 -CO- 169.3 - - 169.7 169.32 169.32 169.32 1 H NMR spectrum -CH- (m) 5.26 - 5.22–5.28 - 5.26 5.26 5.26 -CH 2 - (dq) 2.45–2.63 - 2.43–2.64 - 2.17–2.60 2.17–2.62 2.17–2.60 -CH 3 (d) 1.27 - 1.27–1.29 - 1.28–1.60 1.26 1.28 m: multiplet, dq:double quadruplet, d: doublet. The thermal properties of PHB were determined using the DSC method ( Figure 3 A). The melting temperature (T m ), glass transition temperature (T g ) and crystallinity (X C ) are key parameters to polymer processing and applications. Figure 3 A shows the curve obtained from the second heating from which T m was found at 165.4 °C and the crystallization temperature (T c ) at 54.3 °C. From these data, it was determined that the melting enthalpy and crystallinity were 81.3 J/g and 56 %, respectively. T g was determined at 3.5 °C. Results obtained for the PHB produced by C. necator ATCC 17697 are similar to those reported in the literature for this biopolymer [ 64 , 65 , 66 , 67 ]. ATR-FTIR spectra of the PHB samples from C. necator ATCC 17697 and the commercially available PHB are shown Figure 3 B. The IR spectra revealed an intense band at 1720 cm −1 associated with the C=O bond stretching, which corresponds to the characteristic ester carbonyl group of polyhydroxyalkanoates [ 22 , 68 ]. At 1181 cm −1 there is a band that is well known in the FTIR spectra of PHB due to the asymmetric stretching vibration of the C–O–C group. The C–H stretching from methyl and ethyl groups was assigned to the bands located in the spectral region around 2900 cm −1 . The obtained ATR-FTIR spectrum of the polymer produced by fermentation was in agreement with the corresponding spectrum to the commercial PHB. Figure 3 C shows the 13 C NMR spectrum of the polymer synthesized by C. necator ATCC 17697. Signals at chemical shifts (ppm): 169.3, 67.7, 40.1, 19.9 were assigned to –CO–, –CH–, –CH 2 - and –CH 3 , respectively. The signal at ca. 77 ppm is a triplet that corresponds to the solvent, CDCl 3 . 13 C RMN analysis is in agreement with the expected values for the PHB chemical structure [ 65 , 67 ]. Figure 3 D shows the 1 H-RMN of the polymer synthesized by C. necator ATCC 17697. The peaks observed correspond to those previously reported for the PHB structure [ 66 , 67 ]. The doublet resonance signal at 1.27 ppm is attributed to the methyl –CH 3 protons of the pendant chain in the PHB molecule. The doublet of quadruplet resonance signal at 2.45–2.63 ppm and the multiplet resonance signal at 5.26 ppm, corresponding to the methylene –CH 2 - and the methine –OCH– protons of the backbone chain, respectively. The results obtained employing ATR-FTIR and NMR spectroscopy confirmed that the biopolymer produced by fermentation of C. necator ATCC17697 is the homopolyester PHB. The physico-chemical characterization is in the range expected for PHB obtained with the same extraction method but produced by different strains grown with different carbon sources ( Table 3 ). Indirect cytotoxicity assessment was carried out to establish the cytotoxic effect of PHB polymer produced by C. necator ATCC 17697. Microscopic photographs obtained from the NIH/3T3 fibroblast cells in the control and PHB extract assays are presented in Figure 4 . Figure 4 Indirect cytotoxicity test of PHB polymer synthesized by C. necator ATCC 17697. “Null control” (DMEM medium without polymer), “negative control” (PTFE), “positive control” (latex rubber) and “PHB” (PHB produced by C. necator ATCC 17697). Magnification 100x. Figure 4 No cytotoxic effect of PHB polymer was observed on the NIH/3T3 fibroblast cells. There was neither alteration in cell morphology nor fibroblast monolayer detaching, both in PHB pure extract assay and diluted. The same results were observed in the negative and null control tests. Only in the positive control test, the fibroblast monolayer was not formed."
} | 6,453 |
34778649 | PMC8582040 | pmc | 7,019 | {
"abstract": "In this work, synergistic\neffects derived from surface engineering\nand dielectric property tuning were exploited to enhance the output\nperformance of a triboelectric nanogenerator (TENG) based on an inorganic/porous\nPDMS composite in a contact–separation mode. BaTiO 3 (BT)/porous PDMS films with different BT weight ratios were fabricated\nand evaluated for triboelectric nanogenerator (TENG) application.\nMaximum output signals of ca. 2500 V, 150 μA, and a power density\nof 1.2 W m –2 are achieved from the TENG containing\n7 wt % BT, which is the best compromise in terms of surface roughness,\ndielectric constant, and surface contact area as evidenced by SEM\nand AFM studies. These electrical signals are 2 times higher than\nthose observed for the TENG without BT. The 7BT/porous PDMS-based\nTENG also shows high stability without a significant loss of output\nvoltage for at least 24 000 cycles. With this optimized TENG,\nmore than 350 LEDs are lit up and a wireless transmitter is operated\nwithin 9 s. This work not only shows the promoting effects from porous\nsurfaces and an optimized dielectric constant but also offers a rapid\nand template/waste-free fabrication process for porous PDMS composite\nfilms toward large-scale production.",
"conclusion": "3 Conclusions A BT/porous PDMS composite film was successfully\nfabricated using\na fast, solventless, and template-free fabrication technique. The\nvirtually linear relationship between the dielectric constant of the\ncomposite film and BaTiO 3 content is observed, suggesting\nsuccessful incorporation between BaTiO 3 and PDMS using\nthe developed technique. The BT/porous PDMS-based TENG shows superior\noutput performance compared to the TENG with bare porous PDMS. Such\nenhanced performance could be attributed to the increase of surface\ncharge density induced by the high dielectric constant and porous\nsurface. The best output performance is accomplished from the PDMS\ncomposite-based TENG with 7 wt % BaTiO 3 , producing voltage,\ncurrent, and power density of ca. 2500 V, 150 μA, and 1.21 W\nm –2 , respectively. This optimized TENG also shows\nvery high durability without a significant loss of its voltage for\nat least 24 000 cycles. The capability of such optimized TENG\nas a power source to drive commercial green LEDs and wireless transmitter\ncircuits is successfully demonstrated. The simple, green, and scalable\nfabrication process for the porous PDMS composite film developed in\nthis work would benefit large-scale production of the porous PDMS\ncomposite-based TENG and its practicality as a sustainable energy\nsystem in small electronic devices.",
"introduction": "1 Introduction Small, portable, and wearable low-powered electronics such as smart\nwatches and health monitoring sensors have increasingly been integrated\ninto our everyday lives. These devices, however, are mostly powered\nby batteries that have caused unfavorable environmental consequences\nand limited further applications as flexible, self-sustainable, and\nbiocompatible electronic systems. 1 , 2 Therefore,\nit is desirable to integrate flexible, environmentally friendly, and\nrenewable energy harvesters into these electronics not only to alleviate\nenvironmental problems associated with batteries but also to realize\nself-powered systems with more flexibility to mechanical deformations\nunder real working conditions. For these purposes, a triboelectric\nnanogenerator (TENG) has emerged as a promising sustainable energy\nharvesting technology that transforms wasted mechanical energies from\nthe surrounding environment into electricity via the coupling mechanism\nof triboelectrification and electrostatic induction. 3 − 8 With TENG technology, real-time monitoring and self-powered electronics\nfor modern healthcare diagnostics and delicate mechanical/chemical\nsensing have successfully been demonstrated. 9 − 13 Poly(dimethylsiloxane) (PDMS) is one of the\nmost negative tribomaterials\ncommonly used in TENG because of its flexibility, high electronegativity,\nnontoxicity, and biocompatibility. To enhance PDMS-based TENG electrical\nperformance, various structures and patterns such as micro/nanoporous\nand sponge-like structures are created to increase the porosity, surface\nroughness, and hence effective contact area of the PDMS film. 14 − 17 Additionally, incorporating high dielectric inorganic materials\ninto a PDMS matrix could promote the relative permittivity and charge\ndensity of tribomaterials, which further boost the PDMS-based TENG\nelectrical output. 18 − 22 Therefore, increasing both surface porosity and dielectric constant\nof the tribomaterials is expected to substantially boost the electrical\nperformance of PDMS composite-based TENGs. 2 , 23 To\nfabricate highly porous PDMS composites, sacrificial agents or excess\norganic solvents are often required to achieve high porosity and well-dispersed\ninorganic particles. 17 , 20 , 24 , 25 Consequently, additional steps to eliminate\nsacrificial templates and solvents are inevitable. For example, microwave\nirradiation was employed for the fabrication of sponge-type PDMS,\nthe pore size of which depends on the boiling point of the sacrificial\nsolvent. 24 NaCl and Na 2 CO 3 were used as sacrificial templates for porous PDMS composite\nfilms. A porous structure was then obtained after immersing the template/PDMS\nfilm in excess water under constant stirring 17 or bath sonication. 25 A polystyrene sphere\n(PS) was used to fabricate a PDMS inverse opal-structured film, from\nwhich the sphere was later removed by soaking the PS-embedded PDMS\nfilm in acetone for at least 24 h to ensure complete removal. 16 These fabrication methods not only generate\nchemical wastes but also make the process more time-consuming and\ncause inefficient chemical usage. Therefore, the development of rapid\nand environmentally friendly fabrication processes is necessary. Encouraged by the literature above, a combined approach of surface\nengineering and an increasing dielectric constant was adopted to enhance\nthe output performance of an inorganic/PDMS composite-based TENG using\na fast and waste-free fabrication process developed in this work.\nTo the best of our knowledge, there have been no attempts to explore\nthe possibility of making porous PDMS-based composites without using\nsacrificial agents or organic solvents. By taking the advantage of\ncentrifugal and shearing forces for fabricating PDMS composites and\nsilicon molds for surface patterning, a fast, solventless, and template-free\nfabrication process was achieved. The process employed herein can\nfabricate each porous PDMS composite film within 15 min and thus is\nfavorable to large-scale production. The results revealed that open-circuit\nvoltage ( V OC ) and short-circuit current\n( I SC ) of the TENG with a composite film\nwere almost twofold compared to those of the TENG with bare porous\nPDMS. The application of the BT/porous PDMS-based TENG as an energy\nsupply for LEDs and wireless transmitters as well as its durability\nwas also well demonstrated.",
"discussion": "2 Results and Discussion BaTiO 3 (BT) was chosen as a model inorganic dielectric\ndue to its high relative permittivity. As illustrated in Figure 1 , the fabrication\nprocess of a BT/porous PDMS composite layer began by mixing a PDMS\nelastomer and a cross linker (Sylgard 184, Tow Corning) in a 10:1\nweight ratio. BT powders (an average particle size of <3 μm,\nSigma-Aldrich) in different weight percentages (0, 2.5, 5, 7, 9, 13,\n17, 20, and 23 wt %) were then added to the PDMS mixture. The samples\nwere named x BT/porous PDMS, where x is the BT weight percentage. The resulting mixture was blended using\na planetary centrifugal mixer (ARE-310, Thinky Co.) at a revolution\nspeed of 1800 rpm for 90 s and then at 1500 rpm for 40 s. A surface\npatterning technique modified by our previous work 26 was used to create a porous PDMS surface utilizing a nanograss\nsilicon (Si) mold. After the nanograss Si mold was treated with hexamethyldisilazane\n(HMDS, J.T.Baker), the blended BT/PDMS mixture was subsequently cast\non the mold using spin coating at 250 rpm for 30 s and then at 350\nrpm for another 30 s. The composite film was thermally cured at 120\n°C for 10 min and finally peeled off. The average film thickness\nwas ca. 130 μm. No solvent or sacrificial templates were required\nin the process. Figure 1 Schematic diagram for the fabrication process of the BT/porous\nPDMS composite layer. Top-view SEM images of\nbare PDMS and its composite films with 7,\n9, and 17 wt % BT are displayed in Figure 2 a–d, respectively. A rough surface\nwith a highly porous structure is visibly observed for bare PDMS ( Figure 2 a). The film can\nretain its porous structure with the increase of the BT amount from\n2.5 to 7 wt % ( Figures S1a,b and 2 b), indicating successful pattern transfer for the\ninorganic/porous PDMS composite film. By increasing the BT content\nfurther to 9 and 17 wt % ( Figure 2 c,d), protruded particles can easily be found on the\nPDMS surface. Moreover, the films also lose their porous feature,\nas shallow pore depth can be clearly observed. This is probably because\nblended mixtures with high particle contents tend to have higher viscosity,\nresulting in the lower fidelity of pattern transfer 27 , 28 and the greater probability of particles emerged from the polymer\nsurface. 17 Cross-sectional SEM images of\nbare porous PDMS and x BT/porous PDMS films ( Figures 2 e–h and S2 ) reveal well-dispersed BT particles in films\ncontaining BT up to 17 wt %; however, the agglomeration of particles\nis easily spotted on increasing the BT content further. Digital images\nof all x BT/porous PDMS films ( Figure 2 i) show that transparent PDMS changes to\nhomogeneous opaque white when BT particles were introduced. Here,\nthe blending step is important to achieve the high dispersibility\nof BT particles and complete deaeration. Spiral vortical flow within\nthe container generated by a combination of shearing and centrifugal\nforces promotes mixing and degassing at the same time. 29 , 30 Each BT/porous PDMS composite film can successfully be fabricated\nin less than 15 min without any additional solvent/template elimination\nsteps, which is favorable to large-scale fabrication. Figure 2 SEM images showing the\nsurface (a–d) and cross section (e–h)\nof the composite films: 0BT/porous PDMS (a, e), 7BT/porous PDMS (b,\nf), 9BT/porous PDMS (c, g), and 17BT/porous PDMS (d, h). Photographs\nof x BT/porous PDMS composite films (i) fabricated\nwith different BT amounts. Triboelectric charge transfer occurs during contact between the\ntriboelectric materials, and it is this transferred charge density\nthat determines the output performance of TENGs. According to previous\nreports, 31 , 32 the dielectric constant of the triboelectric\nmaterial is a critical factor boosting the maximum charge density\n(σ′) expressed as follows 32 1 where σ 0 is the triboelectric\ncharge density at the equilibrium state, x ( t ) is the gap distance (2 mm in this study), and d PDMS and ε PDMS are the thickness\nand dielectric constant of the PDMS-based film. Therefore, increasing\nthe dielectric constant would improve the maximum charge density and\nTENG output performance. The dielectric properties of triboelectric\nmaterials constituted in the TENG device have shown a strong influence\non capacitive characteristics and corresponding TENG electrical output\nperformance. 17 , 21 , 33 To measure the capacitance of the obtained x BT/porous\nPDMS composite, the film was sandwiched between two single-sided copper\nboards to form a parallel plate capacitor (the inset of Figure 3 a), and its dielectric constant\nwas determined from the following equation 2 where C , ε 0 , ε r , A , and d composite are\nthe measured capacitance, vacuum permittivity\nof free space (8.85 × 10 12 F m –1 ), relative permittivity (also called dielectric constant), area,\nand thickness of the composite material, respectively. From Figure 3 a, the dielectric\nconstant of bare porous PDMS is ca. 3.13, which is consistent with\nother studies. 34 , 35 As expected, the dielectric constant\nincreases almost linearly with the increase of the BT weight ratio\nand reaches a maximum value of ca. 4.27 for the 23BT/porous PDMS film.\nA similar trend was found in previous reports, although with different\ndielectric constant values, 17 , 21 , 36 which is probably due to diverse degrees of particle distribution\nand agglomeration, compromising the homogeneity of the composite films.\nBecause the surface potential of materials can also affect TENG output\nperformance, 37 the surface potential of\nthe obtained composite films was therefore measured by the KPFM technique.\nFrom Figure 3 b, the\naverage surface potentials for PDMS composite films with BT contents\nof 0, 7, 9, and 17 wt % are 85.77, 495.63, 710.06, and 926.42 mV,\nrespectively. Clearly, increasing the BT particles also improves the\nsurface potential of the composite films. Since the TENG can be simply\nmodeled as a parallel plate capacitor, 22 , 32 , 38 , 39 such enhanced dielectric\nconstant and increased surface potential achieved from the composite\nfilm would boost triboelectric charge density and improve the output\nperformance of the TENG in harvesting mechanical energy. Figure 3 (a) Dielectric\nconstant of the composite film as a function of\nBaTiO 3 weight percentage. The inset illustrates an experimental\nsetup for parallel plate capacitor measurement. (b) KFPM measurement\nshowing different surface charge potentials of (i) bare PDMS and composite\nfilms with (ii) 7 wt %, (iii) 9 wt %, and (iv) 17 wt % of BT. Figure 4 presents\nthe working mechanism of the x BT/porous PDMS-based\nTENG in a contact–separation mode by applying a periodic vertical\nforce. The TENG comprises an aluminum sheet and an x BT/porous PDMS film as positive and negative triboelectric layers\nwith an interlayer distance of 2 mm. As the composite-based TENGs\nare fabricated without a spacer, full contact between aluminum and\ncomposite films is expected. In the beginning, no charge is initially\ndeveloped, as the two triboelectric layers are not in contact. As\nthe aluminum film is pressed to come in contact with the composite\nfilm, triboelectrification occurs and surface charges with an opposite\nsign are generated between them. During this step, electron transfer\ndue to the electron affinity difference is generally considered as\nthe main source of triboelectrification in metal–polymer TENGs. 17 , 40 However, contributions from ion transfer and material transfer mechanisms\nto contact electrification are also possible. 41 Since our electrical characterization was performed under the same\nrelative humidity range (50–60%), the effect of ion transfer\non all samples is relatively similar. A possible contribution from\nthe material transfer mechanism is supported by the XPS results, as\nshown in Figure S3 and Table S1 , from which\nan increase in Si 2p concentration is clearly observed on the surface\nof aluminum tape upon coming in contact with the PDMS composite film.\nAs a result of these mechanisms, the composite layer by gaining electrons\nbecomes negatively charged (−σ′), whereas the\naluminum film by losing electrons becomes positively charged (+σ′).\nWhen the aluminum film is separated from the bottom structure, its\npositive charges attract electrons (−Δσ) from a\nbottom electrode (silver) through an external circuit until an equilibrium\nstate, called the open-circuit voltage ( V OC ) condition, is established ( Figure 4 c). At this moment, V OC can be determined as follows 33 3 Figure 4 (a–d) Charge generation mechanism of\nthe porous PDMS composite-based\nTENG under an external force. Equation 3 indicates\na direct relation between V OC and σ′,\nwhich is strongly influenced by the capacitance of the TENG configured\nin the contact–separation mode and the effective surface area. 42 Once the aluminum film comes again in contact\nwith the composite layer, electrons flow from the aluminum film back\nto the silver electrode. An opposite current is therefore observed.\nAt this point, x ( t ) reduces to zero,\nleading to zero voltage potential according to eq 3 . This is known as the short-circuit condition\n( I SC ), which is given by 33 4 where v ( t ) is the velocity of contact–separation movement.\nIt can be\nclearly seen from eqs 3 and 4 that both V OC and I SC are directly proportional to\nσ′, which can be enhanced with the increase of the dielectric\nconstant. The x BT/porous PDMS-based TENGs were\nelectrically\ncharacterized via a vertical contact–separation mode at different\nfrequencies (1, 2, 4, and 5 Hz). Identical force and 2 mm distance\nbetween the aluminum and composite films were applied to all fabricated\nTENGs. As shown in Figure 5 a–d, introducing BT particles into bare PDMS obviously\npromotes both V OC and I SC of the fabricated TENG. The output signals ( Figure 5 e,f) gradually enhance\nwith the increase of the BT amount and reach their maximum values\nusing the 7BT/porous PDMS composite. The composite layer with a higher\ndielectric constant possesses a higher surface charge density. 43 The intensity of the electrostatic induction,\ngenerated by the increased surface charges, is therefore stronger\non both electrodes, which leads to the enhancement of V OC and I SC . Increasing the\nBT content further to 9 wt %, however, decreases the output performance.\nAt 5 Hz ( Figure 5 f),\nthe highest V OC and I SC observed for the TENG with the 7BT/porous PDMS film\nare ∼2500 V and ∼150 μA, respectively. These values\nare almost twice those measured using the bare porous PDMS film, which\nevidently indicates a successful enhancement of porous PDMS-based\nTENG efficiency by incorporating a high permittivity ceramic into\nthe porous PDMS matrix. A similar trend, where the output performance\nfirst increases to its maximum and then decreases when a very high\namount of BaTiO 3 was added, is also observed at frequencies\nof 1 and 4 Hz ( Figure S4 ). Such behavior\ncan be understood by considering two crucial factors that are the\nrelative permittivity and surface contact area of the PDMS film. It\nis clear from Figure 3 a,b that the dielectric constant and surface potential keep increasing\nwith the increase of BT content; therefore, V OC and I SC should increase as well.\nHowever, V OC and I SC decrease with the increase of the BT amount to more than\n7 wt %. As evidenced by SEM in Figure 2 c, the composite film significantly loses its porosity,\nand the BT particles are clearly found on the PDMS surface when the\nBT content is higher than 7 wt %. Because PDMS is a superior tribomaterial\nthan inorganic BT, a decrease in the PDMS surface area can result\nin a reduction of generated surface charge density (σ′).\nIn addition, AFM investigation ( Figure S5 ) also reveals that the average surface roughness ( R a ) decreases with the increase of BT content. The R a of the 7BT/porous composite is ca. 89.8 nm,\nwhile those of composite films with higher BT contents (9–23\nwt %) are in the range of 72.7–65.7 nm. Previous studies 43 − 46 have shown that surface roughness is one of the most important factors\ninfluencing the output signals of TENGs. Lower surface roughness would\nhave a smaller effective contact area and a low deformation level\nand therefore could generate less surface charge density during contacting\nand relaxing. 47 − 49 Therefore, such decreased V OC and I SC observed for composite\nfilms with high BT amounts are likely originated from a reduced surface\ncontact area as supported by SEM and AFM analyses. The above characterization\nresults indicate that the 7BT/porous composite film is the best compromise;\ntherefore, the highest TENG output performance is observed. Our results\ncorrespond well with previous reports, although an optimum BaTiO 3 content may differ due to different TENG fabrication processes\nas well as diverse physical properties and homogeneity of BaTiO 3 particles in BT/PDMS composites. 24 , 36 , 46 , 50 , 51 Figure 5 V OC (a, b), I SC (c, d), and their peak-to-peak value (e, f) of the x BT/porous PDMS-based TENGs as a function of BT content\nat pressing frequencies of 2 Hz (a, c, e) and 5 Hz (b, d, f). It can also be noticed from the results shown in Figures 5 e,f and S4 that increasing the working frequency significantly\naffects I SC but not V OC ,\nwhich is attributed to the fact that V OC depends only on triboelectric charge density and the distance between\ntwo triboelectric materials ( eq 3 ), which are constant for a certain TENG device. Since I SC and the amount of transferred charges are\nclosely related, the relationship between frequency and these two\nparameters for each TENG was further determined. The transferred charge\nwas estimated from the measured peak current of TENGs using the equation Q = ∫ i d t , where i is the instantaneous current. The transferred charge density\n(Δσ) as a function of BT content is then plotted in Figure S6 . The result shows that the maximum\nΔσ is obtained from the 7BT/porous composite film, which\nis similar to those observed for V OC and I SC . A comparison between Figures 5 e,f and S6 indicates\nthat I SC increases with increasing frequency,\nwhile the total amount of charges transferred remains almost the same.\nThe results imply that the enhanced I SC is mainly driven by the increased acceleration rate of contact–separation\nmovement v ( t ) ( eq 4 ) rather than the number of transferred charges. Because the 7BT/porous PDMS-based TENG provides the highest output\nperformance among all TENGs studied here ( Figure 5 ), this TENG was then chosen for further\ninvestigation. The maximum output power of the 7BT/porous PDMS-based\nTENG was determined by connecting the TENG to an external load with\na resistance range of 5 kΩ–100 MΩ. As shown in Figure 6 a, the current slightly\ndecreases in a resistance range of 5 kΩ–2 MΩ since\nthe charge transfer process, within this resistance range, is relatively\nsimilar to the short-circuit condition. It then drastically decreases\nwith a resistive load of higher than 3 MΩ. The output current\nslowly approaches zero by further increasing the resistance beyond\n20 MΩ. Based on these results, the maximum instantaneous power\ndensity of 1.21 W m –2 is achieved at a resistance\nvalue of 5 MΩ. The maximum power density of our device is in\na similar range as the previous study, 24 in which a 3D porous PDMS/BaTiO 3 film was used. As the\n3D porous network film generally has a large surface area, the comparable\nresults here thus underline our porous composite-based TENG with boosting\nperformance. The ability of the 7BT/porous PDMS-based TENG in charging\ndifferent capacitors is also evaluated. As shown in Figure 6 b, the composite-based TENG\nis used to charge 10, 33, and 100 μF capacitors under a periodic\ncontact and separation mode at 5 Hz. After the TENG output is rectified,\nthe 10, 33, and 100 μF capacitors are charged to 10 V in 18,\n57, and 157 s, respectively. Figure 6 (a) Maximum output current and power as a function\nof external\nload resistance and (b) the charging ability of the 7BT/porous PDMS-based\nTENG at 5 Hz. The application of the composite-based\nTENG as a power source for\nlow-powered electronics is also demonstrated. It can be used to drive\nat least 372 green LEDs connected in series to high brightness, even\nthough the room is full of lights ( Figure 7 a and Video S1 in the Supporting Information). Furthermore, the electrical energy\nscavenged from the applied vertical force at 5 Hz can charge a 10\nμF capacitor to 7 V within 9 s. This stored energy can then\nbe successfully used to power a wireless transmitter circuit for sending\na command to a receiver to turn on/off the light ( Figure 7 b and Video S2 ). The capacitor voltage characteristics of porous PDMS-based\nTENGs with and without BT are compared in Figure 7 c. Evidently, the output performance of the\nPDMS composite-based TENG is far superior to that of the bare PDMS-based\nTENG, which needs almost 30 s to harvest enough energy to supply the\ntransmitter. To evaluate the stability and durability of the 7BT/porous\nPDMS-based TENG, its electrical output voltage was measured for more\nthan 40 min (at least 24 000 cycles). The output voltage is\nbarely reduced, as presented in Figure 7 d, implying the excellent stability and durability\nof the porous PDMS composite-based device fabricated in this work. Figure 7 (a) Photograph\nof 372 green LEDs. The inset shows a schematic structural\ncomponent of the TENG. (b) Wireless transmitter operated by the 7BT/porous\nPDMS-based TENG. (c) Comparison of capacitor voltage characteristics\ncharged by porous PDMS-based TENGs with/without BT. (d) Stability\nand durability test of the PDMS composite-based TENG under a periodic\ncompression of 5 Hz for >24 000 cycles."
} | 6,204 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.