(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



A sporulation signature protease is required for assembly of the spore surface layers, germination and host colonization in Clostridioides difficile [1]

['Eleonora Marini', 'Instituto De Tecnologia Química E Biológica António Xavier', 'Universidade Nova De Lisboa', 'Avenida Da República Ean', 'Oeiras', 'Carmen Olivença', 'Sara Ramalhete', 'Andrea Martinez Aguirre', 'Texas A M University', 'Department Of Biology']

Date: 2023-12

A genomic signature for endosporulation includes a gene coding for a protease, YabG, which in the model organism Bacillus subtilis is involved in assembly of the spore coat. We show that in the human pathogen Clostridioidesm difficile, YabG is critical for the assembly of the coat and exosporium layers of spores. YabG is produced during sporulation under the control of the mother cell-specific regulators σ E and σ K and associates with the spore surface layers. YabG shows an N-terminal SH3-like domain and a C-terminal domain that resembles single domain response regulators, such as CheY, yet is atypical in that the conserved phosphoryl-acceptor residue is absent. Instead, the CheY-like domain carries residues required for activity, including Cys207 and His161, the homologues of which form a catalytic diad in the B. subtilis protein, and also Asp162. The substitution of any of these residues by Ala, eliminates an auto-proteolytic activity as well as interdomain processing of CspBA, a reaction that releases the CspB protease, required for proper spore germination. An in-frame deletion of yabG or an allele coding for an inactive protein, yabG C207A , both cause misassemby of the coat and exosporium and the formation of spores that are more permeable to lysozyme and impaired in germination and host colonization. Furthermore, we show that YabG is required for the expression of at least two σ K -dependent genes, cotA, coding for a coat protein, and cdeM, coding for a key determinant of exosporium assembly. Thus, YabG also impinges upon the genetic program of the mother cell possibly by eliminating a transcriptional repressor. Although this activity has not been described for the B. subtilis protein and most of the YabG substrates vary among sporeformers, the general role of the protease in the assembly of the spore surface is likely to be conserved across evolutionary distance.

Clostridioides difficile, an anaerobic spore-forming bacterium, colonizes the gastro-intestinal tract when, as during antibiotic treatment, the protective effect of the microbiota is disrupted. A leading agent of nosocomial infections, causing a range of symptoms from mild diarrhea to life-threatening conditions, the organism is recognized as a global and urgent threat. Infection begins with the ingestion of spores, which will germinate in response to bile salts. Two proteinaceous spore surface layers, the coat and the exosporium, play a crucial role in infection and colonization, as they contribute to spore resistance, binding to host cells and the interaction with and the response to germinants. The yabG gene, part of a genomic signature for sporulation, codes for a cysteine protease, with residues required for catalysis embedded in a CheY-like response regulator receiver domain. YabG is required for proper morphogenesis of the spore surface layers, germination and host colonization. YabG also regulates the mother cell line of gene expression by allowing the expression of genes required for assembly of the coat and exosporium. While this latter function has not been described for other organisms, the general role of yabG in the assembly of the spore surface layers is likely to be conserved among spore-formers.

Funding: This work was supported by the European Union Marie Sklodowska Curie Innovative Training Networks (contract number 642068) to AOH and EM was the recipient of a PhD fellowship under that contract. This project was supported by awards PTDC/BIA-MIC/29293/2017 to MS from FCT (“Fundação para a Ciência e a Tecnologia") and 5R01AI116895 and 1U01AI124290 to J.A.S. from the National Institute of Allergy and Infectious Diseases. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIAID. The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication. A.M.A. was supported by the Mexican Science and Technology Council (CONACYT Mexico) under award number 625561/472087. This work was also financially supported by Project LISBOA-01-0145-FEDER-007660 (“Microbiologia Molecular, Estrutural e Celular”) funded by FEDER funds through COMPETE2020 – “Programa Operacional Competitividade e Internacionalização” (POCI), by national funds through the FCT. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIAID. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

We show that C. difficile YabG has a C-terminal domain that resembles a single domain response regulator such as CheY. Residues within the CheY-like domain are required for an auto-proteolytic activity that leads to its complete degradation and for cleavage of its substrates. These residues include C207 and H161 which occupy positions homologous to those shown to form a catalytic diad required for auto-proteolysis in B. subtilis YabG [ 52 ]. We show that YabG is recruited to the developing spore and that its assembly is temporally controlled by auto-proteolysis. We show that YabG governs attachment of the coat to the underlying cortex peptidoglycan, formation of the exosporium and is also involved in germination in line with its recently demonstrated role in regulating the sensitivity of C. difficile spores to co-germinants [ 43 ]. Importantly, a yabG mutant is impaired in host colonization. Finally, we show that YabG is also required for the expression of a late class of σ K -dependent genes involved in coat/exosporium assembly, thus contributing to the control of the mother cell line of gene expression. YabG defines a novel type of cysteine (thiol) protease dedicated to the assembly of the spore surface layers in sporeforming organisms.

In B. subtilis and in B. anthracis, a role for yabG in the assembly of the spore coat has been shown [ 36 – 41 ] and in C. difficile, yabG plays a part in spore germination ([ 42 , 43 ]; see also below). In B. subtilis YabG associates with the spore coat and is required for cleavage of at least six spore coat proteins [ 36 – 39 ]. Two of those proteins, SafA and C30, which have important morphogenetic functions in assembling of the inner coat layers, are cleaved in vitro by partially purified YabG [ 38 ]. The YabG-dependent cleavage of SafA, C30 and two other inner coat proteins, is important for their subsequent cross-linking by a coat-associated transglutaminase [ 38 ]. These substrates of B. subtilis YabG are not found in C. difficile [ 34 , 44 ]. Another likely substrate of B. subtilis YabG is SpoIVA, a conserved morphogenetic ATPase required for the formation of a basal layer upon which the coat/crust/exosporium are assembled [ 31 , 45 ], SpoIVA may also be a YabG substrate in C. difficile because the protein is found at significantly higher levels in coat extracts from spores of a yabG mutant [ 42 ]. Two other proteins, not found in Bacillus sporeformers, also accumulate in their unprocessed forms in spores of a C. difficile yabG mutant, CspBA and Pre-pro-SleC [ 42 , 43 , 44 ]. Interdomain processing of CspBA, releases CspB, whereas cleavage of Pre-pro-SleC produces pro-SleC [ 42 , 43 ]. CspB is a subtilisin-like serine protease involved in the activation, together with the germinant receptor CspC, of pro-SleC [ 46 , 47 ]. SleC, in turn, is a cortex hydrolase essential for spore germination in response to bile salts [ 47 – 51 ]. Processing of Pre-pro-SleC and interdomain processing of CspBA requires yabG [ 52 ]. YabG also resulted in processing of SleC FL to pro-SleC in E. coli extracts, and a processing site was tentatively identified in CspBA which is also a direct substrate of the protease [ 42 , 43 ]. CspA is important for recognition of co-germinants and mutations in yabG were found to render germination in response to the bile salt taurocholate independent of co-germinants such as glycine [ 43 ].

At the onset of sporulation, the rod-shaped cell divides asymmetrically to form a large mother cell and a smaller forespore. The genetic programs that are then activated in the two cells are governed by a cascade of cell type-specific RNA polymerase sigma factors, σ F and σ G in the forespore and σ E and σ K in the mother cell [ 29 – 31 ]. σ F and σ E control early stages of development, whereas σ G and σ K are active mainly when the mother cell completes engulfment of the forespore, to produce a cell within a cell. Formation of the spore surface structures is mainly a function of the mother cell and requires both σ E and σ K [ 32 , 33 ]. Using the known B. subtilis sporulation genes, a core machinery for sporulation was identified, and a genomic signature defined as those genes present in at least 95% of the genomes of organisms known to sporulate, and in less than 5% of other bacterial genomes [ 34 ]. Strikingly, other than four vegetative genes that are co-opted for sporulation and expressed from σ K -dependent promoters, the σ K -regulon contributed to the genomic signature with a single gene, yabG [ 34 ]. While this reflects the diversity of the genes coding for components of the spore surface layers among sporeformers, it also hinted at an important, phylogenetically conserved function for yabG. As its B. subtilis counterpart, the yabG gene of C. difficile is under the control of σ K [ 30 , 31 , 35 ].

In C. difficile, an exosporium tightly adherent to the underlying coat, forms the spore outermost structure [ 19 – 22 ]. Proper assembly of the coat and exosporium is important for colonization and infection. Components of the coat and exosporium have been identified that highlight the role of these structures in the interaction of spores with the colonic mucosa, colonization and virulence [ 23 – 25 ]. These include the cysteine-rich proteins CdeC and CdeM; cdeC and cdeM mutants, which have a misassembled coat and exosporium, show altered colonization and virulence [ 20 , 24 – 26 ]. Importantly, recent work has shown that the interaction of spores with E-cadherin promotes spore binding to and internalization by intestinal epithelial cells, which in turn contributes to infection recurrence [ 27 , 28 ]. Clearly, the identification and functional characterization of proteins that govern assembly of the coat and exosporium layers remains an important research goal that will inform us on the role and behaviour of spores during the initial stages of infection.

A strict anaerobe, C. difficile relies on spore formation for dissemination and environmental persistence [ 2 – 4 ]. Infection starts with the ingestion of spores that will germinate in the small intestine in response to certain bile salts; at least a fraction of the vegetative cells that outgrow from spores will produce the TcdA and TcdB toxins, and some will differentiate into spores [ 9 , 11 – 14 ]. Spores have a central compartment harbouring the chromosome, delimited by a membrane and surrounded by a thin layer of peptidoglycan that becomes the wall of the cell resulting from spore germination. This unit is enclosed in a much thicker layer of modified peptidoglycan, called the spore cortex. The cortex, essential for spore dormancy, is in turn covered by several proteinaceous layers that together form the surface of spores; the structure and composition of these layers differs greatly among sporeformers [ 15 , 16 ]. In B. subtilis the spore surface consists of a glycosylated crust tightly adherent to an underlying multi-layered coat [ 17 ]. In the pathogens B. cereus and B. anthracis, the coat is surrounded by an exosporium separated from the coat by an interspace [ 15 , 16 ]. The coat/crust afford protection against cortex-lytic enzymes, such as lysozyme, and against small molecules such as oxidizing agents. The exosporium provides physical robustness and serves as a permeability barrier that excludes enzymes and antibodies. Both the coat/crust and the exosporium also affect the interaction of spores with germinants and mediate the interactions of spores with host cells and abiotic surfaces [ 16 , 18 ].

Able to colonize the gastro-intestinal tract when the protective effect of the microbiota is disrupted, Clostridioides difiicile [ 1 ] is the leading cause of nosocomial diarrhoea linked to antibiotic therapy. Infection can, however, lead to more serious complications, including pseudomembranous colitis, toxic megacolon, bowel perforation and in the most severe cases sepsis and death [ 2 – 4 ]. Changes in the epidemiology of C. difficile are causing increased incidence in the community and the risk of zoonotic transmission is an additional threat [ 2 , 4 – 8 ]. Two large toxins, TcdA and TcdB, are the main virulence factors and the direct cause of the disease symptoms [ 9 , 10 ].

Results

Residues of YabG involved in auto-proteolysis The AlphaFold2 model suggests that the catalytic Cys207 in B is at the bottom of a cleft formed between the A domain and the top of B (Fig 1C). The distances estimated between Cys207 and His161 from the model, and between the latter and Asp162 (Fig 1D) are longer than in other Cys proteases and the side chain of His161 is not oriented towards Cys207 [58,59]. However, of the 204 YabG sequences in the MEROPS U57 family, where YabG used to be included [57], 96.1% have His, Asp and Cys residues conserved at equivalent positions. A second Cys residue at position 119 in the C. difficile protein is also invariant among YabG orthologues (Fig 1B). To test whether C. difficile YabG showed auto-proteolytic activity and if so, what residues of the putative active site were involved, we overproduced YabGWT in E. coli, using an auto-induction regime [73], as a N-terminal His 10 fusion. His 10 -YabGWT did not accumulate in whole cell extracts, as assessed by Coomassie staining (Fig 1E). In contrast, B. subtilis YabG accumulated but underwent auto-proteolytic degradation over time [52]. Why the B. subtilis appears more stable than its C. difficile counterpart is presently unknown. To test whether Cys207, His161 and Asp162 were required for YabG auto-proteolysis, YabGC207A, YabGH161A and YabGD162A were also overproduced in E. coli. As assessed by Coomassie staining, the YabGC207A, YabGH161A and YabGD162A forms of the protein accumulated in the extracts as species of about 34 kDa, consistent with the predicted size of the protein (34.7 kDa), indicating that the substitutions impaired auto-proteolysis (Fig 1E). In contrast, another single alanine substitution, D248A in domain B, did not cause the protein to accumulate in extracts (Fig 1E). Only substitutions that impair protease activity allow YabG to accumulate. Thus, YabG residues Cys207 and His161, homologous to the dyad described for the B. subtilis protein, and Asp162, are required for auto-proteolysis. We also tested whether a variant with Cys119 changed to Ala accumulated in extracts. In contrast to YabGC207A, the YabGC119A form did not accumulate in whole cell extracts (S5A Fig). Thus, Cys119 is not required for the auto-proteolytic activity of YabG. Auto-proteolysis of B. subtilis YabG leads to the accumulation of transiently stable fragments through cleavage after Arg5, Arg17, Arg49 and Arg93 [52]. Of these, Arg17 and Arg49 are conserved in C. difficile YabG (Arg9 and Arg41, respectively); the position homologous to Arg93 is occupied by a Lys in the C. difficile protein but there are two Arg´s in the vicinity of this residue (S5B Fig). This suggests that the C. difficile protein may also be cleaved at least after Arg9 and Arg41. That C. difficile YabG is likely to be specific for Arg at the P1 position is in line with cleavage of the SleC and CspBA after Arg residues [43].

C207 is required for YabG-dependent CspBA interdomain processing CspBA accumulates in spores of yabG mutants [42,43,74]. Release of the CspB domain from CspA is necessary for the activation of the cortex hydrolase SleC, required for germination completion [47,51]. YabG was found to be involved in CspBA processing [42,43]. To assess the role of Cys207 in the reaction, YabGWT or YabGC207A were co-produced in E. coli together with CspBA. Production of CspBA alone resulted in the accumulation of a species of about 125 kDa, consistent with the expected size of CspBA (MW 124.6 kDa), detected with anti-CspB antibodies (Fig 1F, black arrowhead in the bottom panel). A species of about 60 kDa was also detected (Fig 1F, green arrowhead). This species, termed CspB*, results from alternative processing that occured in a YabG-independent manner [42]. In this experiment, and as described above, only YabGC207A was detected (Fig 1F). However, even though YabGWT was not detected, its co-production with CspBA resulted in the disappearance of CspBA and the accumulation of a species of about 55 kDa, the size expected for the CspB moiety, detected with anti-CspB antibodies (Fig 1F, red arrowhead). Accumulation of this protein required the activity of YabG: the co-production of CspBA together with YabGC207A did not lead to depletion of CspBA or to the accumulation of CspB, whereas the YabG-independent CspB* still accumulated (Fig 1F, bottom panel). We conclude that Cys207 is required for CspBA interdomain processing, consistent its predicted role in catalysis and with the accumulation of the precursor protein in yabG spores [42,43].

yabG is expressed in the mother cell during late stages of spore morphogenesis Synthesis of the proteins that form the spore coat and exosporium layers is driven by σE and σK, which are mainly active before and after engulfment completion, respectively [15,75]. In B. subtilis, yabG is under the control of σK [39] and genome-wide transcriptional profiling studies of C. difficile sporulating cells have suggested that yabG is also under the control of σK in this organism [30,31,35]. Consistent with these studies, putative -10 and -35 promoter elements that match the consensus for σK recognition are present in the yabG regulatory region (S6A Fig) [30,76]. To determine the time of yabG expression relative to the stages of spore morphogenesis, we made use of a transcriptional fusion between the yabG promoter region and the SNAPCd reporter [29,30,77]. The P yabG -SNAPCd fusion was introduced into the WT strain 630Δerm as well as into congenic sigE::erm and sigK::erm mutants [29]. The resulting strains were inoculated on 70:30 agar plates [53] and imaged by phase contrast and fluorescence microscopy, following labelling with the SNAP substrate TMR-Star, 14 and 20 hours thereafter. In the WT, expression of P yabG -SNAPCd at hour 14 (S6B Fig, top panels) and at hour 20 (bottom panels) was confined to the mother cell. A fluorescence signal from P yabG -SNAPCd was undetected in the sigE::erm mutant, but still detected in the mother cell of a sigK::erm mutant (S6B Fig). This suggests dual control of yabG expression by σE and σK. The consensus for σE recognition is also included in S6A Fig; the highly conserved ATA motif for σE recognition in the -35 region is absent, but the putative -10 region conforms better to the consensus for σE binding than to the consensus for σK binding [76]. To determine the main period of yabG expression, we measured the expression of P yabG -SNAPCd during spore morphogenesis. P yabG -SNAPCd expression was detected at 20 hours of growth in 45% of the cells during asymmetric division and engulfment, in 46% of the sporangia of phase-dark forespores, in 45% of the sporangia of phase-grey forespores, and in 53% of sporangia of phase-bright forespores (S6B Fig, yellow arrowheads). The average fluorescence intensity from P yabG -SNAPCd in sporangia of phase-bright forespores at hour 20 was higher than that of sporangia of phase-dark or phase-grey spores or cells during asymmetric division and engulfment (S6C Fig). At hour 14 the percentage of sporangia with a signal was highest for sporangia of phase-dark spores and the intensity of the fluorescence signal per cell remained relatively uniform regardless of the developmental stage (S6B and S6C Fig). While no expression was detected in a sigE mutant, the intensity of the fluorescence signal for P yabG -SNAPCd in the sigK::erm mutant was decreased relative to the WT, particularly at hour 14 (S6C Fig). Nevertheless, 90 percent of the sporangia of the sigK mutant at hour 14 and 74 percent at hour 20 showed a fluorescence signal presumably because of persistent activity of σE. Together, these results indicate that the onset of yabG expression in the mother cell occurs soon after asymmetric division, under the control of σE, when the mother cell starts engulfing the forespore; it then increases towards the final stages of sporulation, with the contribution of σK, when the forespore transitions from phase-dark/grey to phase-bright (S6 Fig)

The activity of YabG is required for proper spore germination Amino acids such as alanine or glycine act as co-germinants during spore germination triggered by cholic acid derivatives as for instance taurocholate (TA) (reviewed in [78]). Previous work has shown that deletion of or point mutations in yabG allowed spore germination in response to TA alone in epidemic strain R20291 [43]. To examine the role of YabG in the germination of C. difficile 630Δerm spores, we constructed a congenic yabG in-frame deletion mutant, ΔyabG, using a CRISPR-Cas9 system (S7A and S7B Fig). We constructed two additional strains in the ΔyabG background, with either a copy of the WT yabG gene (yabGC) or yabGC207A at the non-essential pyrE locus (S7C and S7D Fig). In the background of strain 630Δerm we found that spores of the yabG mutants germinated slower than WT spores in response to TA in a rich medium (S8A, S8B and S9 Figs). Previous work has shown that the efficiency of germination for ΔyabG spores, as assessed by plating spores exposed to TA onto plates of a rich medium containing the germinant was 0.8 of the WT [42]. Using the same assay, we obtained a plating efficiency of 0.71±0.13 for ΔyabG spores, 0.89±0.13 for yabGC207A spores and 0.96±0.18 for yabGC spores (S8C Fig, top panel). Germination involves release of dipicolinic acid (DPA) from the spore core. DPA accumulated and/or was retained by the spores can also influence germination and thus the impaired germination could result from reduced extrusion of DPA. To test this point, we measured the DPA content of the various spores. We found the DPA content, normalized to the WT, of ΔyabG (125.5% ± 11) and yabGC207A (118.8% ± 10) spores to be not significantly different than that of WT (100%) and yabGC (116.7% ± 2) spores (S8C Fig, middle panel). Moreover, WT spores released 0.4 ± 0.04 DPA (expressed as a ratio of the OD 270 /OD 600 ) during germination, yabGC spores released 0.3 ± 0.02 DPA, while ΔyabG and yabGC207A spores released respectively 0.3 ± 0.01 and 0.4 ± 0.09 DPA (S8C Fig, bottom panel). Thus, 1 hour after induction of germination, the DPA released from yabG mutant spores is not significantly reduced compared to WT and yabGC spores. Only CspB (55 kDa) was detected in extracts from WT or yabGC spores; in contrast, both CspBA and CspB* (see above) were detected in ΔyabG and yabGC207A spores (S8D Fig) [42]. Pre-pro-SleC (47 kDa) was processed to its Pro form (34 kDa) in WT and yabGC spores, but only full-length SleC was detected in ΔyabG or yabGC207A spores (S8D Fig, middle panel). Finally, the levels of CspC did not differ significantly between WT and ΔyabG spores or between yabGC and yabGC207A spores (S8D Fig, bottom panel). Thus, the partial germination defect of ΔyabG and yabGC207A spores may be due, at least in part, to loss of YabG activity, which in turn impairs the release of CspB from CspBA and the production of pro-SleC. Because germination is also influenced by the status of the spore surface layers, we next wanted to investigate a possible role of YabG in the assembly of the coat and exosporium.

yabG spores are more permeable to lysozyme Since the integrity of the surface layers is important to prevent access of peptidoglycan-breaking enzymes to the spore cortex, we tested the resistance of spores to lysozyme treatment. Density-gradient purified spores were plated in a rich medium in the presence of the germinant taurocholate, before or after treatment with lysozyme. Survival was of 98.3% for WT spores, 69% for ΔyabG spores 72.3% for yabGC207A spores and 90.6% for yabGC spores. These numbers are close to the efficiency of plating of spores onto plates containing taurocholate without lysozyme treatment (see above) and were not statistically significant as assessed by one-way ANOVA and Tukey’s multiple comparison tests, suggesting that lysozyme had no effect on spore survival. Nevertheless, upon exposure to lysozyme, 46% of the ΔyabG spores and 41% of the yabGC207A spores become phase-dark (6% for WT spores and 15% of the yabGC spores) (S10 Fig). Thus, the alterations in assembly of the coat/exosporium layers appear sufficient to allow access of lysozyme to the spore cortex.

YabG is required for the expression of cotA and cdeM, involved in coat and exosporium assembly One explanation for the absence of CotA and CdeM from the coat/exosporium extracts of yabG spores, is that somehow YabG could affect production of these proteins. To test this possibility, we first analysed the accumulation of CotA, CdeM and CdeC, for reference, by immunoblotting, in whole cell extracts prepared from sporulating cells harvested after 14 and 20 hours of growth on 70:30 agar plates. CotA and CdeM were detected in the extracts prepared from the WT at 14 and 20 hours of growth but not in the extracts prepared from the yabG mutants; in contrast, CdeC was detected in the WT and the mutants at both time points (Fig 3A and 3B). Thus, the absence of CdeM and CotA from the coat/exosporium extracts of yabG and yabGC207A spores appears to be a consequence of reduced synthesis or accumulation of the two proteins. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. YabG affects the expression of genes coding for coat and exosporium components. A and B: Coomassie stained gel and immunoblotting analysis of sporulating cells of the WT, ΔyabG, yabGC and yabGC207A 14 (A) and 20 hours (B) after inoculation in 70:30 agar plates [53]. Proteins in whole cell extracts were resolved by SDS-PAGE and the gels subjected to immunoblotting with anti-CotA, anti-CdeC, and anti-CdeM antibodies. The red arrowheads point to the various forms of CotA, CdeC and CdeM. C: Quantification of the expression of the indicated genes (cdeC, cdeM and cotA) by qRT-PCR. Total RNA was extracted from C. difficile 630Δerm and ΔyabG strains grown in 70:30 agar plates for 14 and 20 hours. The graph shows the fold-change in the expression of cdeC, cdeM and cotA between the ΔyabG and the WT. Error bars correspond to the standard deviation derived from three biological replicates. Statistical analysis used a Student’s t-test: * p<0.01; **p<0.001. https://doi.org/10.1371/journal.ppat.1011741.g003 We next measured the transcript levels of cotA, cdeM and cdeC, in sporulating cells of the WT and the ΔyabG mutant using qRT-PCR. These experiments showed decreased levels of cotA and cdeM transcripts in ΔyabG cells compared to the WT both at 14 h (expression ratio ΔyabG/WT of 0.060 for cotA and 0.216 for cdeM) and at 20h of growth (expression ratio ΔyabG/WT of 0.098 and 0.126, respectively) (Fig 3C). In contrast, cdeC expression increased from a ΔyabG/WT ratio of 1.873 at 14 hours to 7.139 at hour 20 (Fig 3C). While not excluding a direct role for YabG in the assembly of CotA and CdeM (but see section below), these results show that yabG is required for the expression of cotA and cdeM. Moreover, since both the ΔyabG and yabGC207A mutations strongly reduce the levels of the two proteins in spores, we infer that the proteolytic activity of YabG is required for the expression of cotA and cdeM. One possibility is that the activity of YabG is required for the removal of a negative regulator of cotA and cdeM expression; the increased expression rate of cdeC observed at 20 h of growth might be an indirect effect of the absence of YabG activity at late stages of sporulation (see also the Discussion).

Bypass of YabG for expression of cotA and cdeM and assembly of CdeM and CotA That both the ΔyabG and yabGC207A alleles reduced the levels of the cotA and cdeM transcripts raised the possibility that the activity of YabG is somehow required to antagonize a transcriptional repressor or to activate a factor required for transcription of both genes. If so, then replacing the cotA and cdeM promoters by a yabG-independent promoter, should bypass the need for yabG for CotA and CdeM production. Both genes were placed under the control of the cotE promotor; cotE also codes for a late coat protein, produced under the control of σK; the cotE promoter, however, does not appear to be YabG-dependent, since CotEFL accumulates in yabG spores (Fig 2A). Strains bearing the P cotE -cotA and P cotE -cdeM fusions in the WT, yabG and yabGC207A backgrounds were grown under sporulation conditions, spores were purified and coat/exosporium extracts prepared and analysed. CdeM is was detected in Coomassie-stained gels [25], whereas the presence/absence of CotA in spores extracts required verification by immunoblotting. Expression of cdeM or cotA from the cotE promotor restored the presence of CdeM (Fig 4A) and CotA (Fig 4B, bottom panel) in extracts from ΔyabG or yabGC207A spores. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. YabG-independent expression of cotA and cdeM restores assembly of CotA and CdeM to ΔyabG and yabGC207A spores. Coomassie stained SDS-PAGE gel of the proteins extracted from purified spores of the WT, ΔyabG and yabGC207A mutants and derivatives expressing P cotE - cdeM (A) or P cotE - cotA (B). In A, the bottom panel shows the immunoblot analysis of the corresponding gel using an anti-CotA antibody. The position of the main forms of CdeM (in A) or CotA (in B) is shown by red arrowheads. C: spores produced by the yabGC207A mutant and by derivatives expressing P cotE -cdeM or P cotE -cotA were imaged by TEM. The polar region of yabGC207A/P cotE -cotA and yabGC207A/P cotE -cdeM spores is magnified in the panels to the right. Cr, core; Cx, cortex; Ap, spore appendage region. Green arrowheads, coat and exosporium material peeling off the spore; blue arrowheads, regions with an exposed cortex. Scale bars: 100 nm for the magnified images, 500 nm for all other panels. https://doi.org/10.1371/journal.ppat.1011741.g004 To determine if and to what extent expression of cdeM or cotA corrected the phenotype of yabGC207A spores we used TEM. In spores of the yabGC207A mutant expressing of P cotE -cdeM the lamellar appearance of the polar appendage region was lost (Fig 4C, middle panel); instead, the appendage region appeared compact and electrondense (Fig 4C, Ap), consistent with the accumulation of CdeM and its role in formation of the polar appendage [25]. Other features of yabG spores, however, such as the peeling off of significant sections of the coat/exosporium were maintained (Fig 4C, blue arrowheads). Expression of P cotE -cotA in the yabGC207A background resulted in spores with visible juxtaposed sheets in the polar appendage region (Fig 4C, panels in the right), similar to yabG spores and most likely due to the absence of CdeM (see also above). These results show that YabG acts at the level of the cotA and cdeM promoters to influence transcription of these genes. In addition, since the expression of cdeM and cotA from P cotE result in the detection of CdeM and CotA in coat/exosporium extracts prepared from yabG spores, we infer that YabG is not a strict requirement for the localization of CotA or CdeM.

Auto-regulatory assembly of YabG We then wanted to monitor the sub-cellular localization of YabG. To this end, a translational fusion of the WT protein to the SNAPCd tag, YabGWT-SNAPCd, was constructed and introduced into the WT strain. As detailed below, the YabGWT-SNAPCd fusion is largely functional. Cells were grown in 70:30 agar plates and imaged by phase contrast and fluorescence microscopy 14 and 20 hours after inoculation. At hour 20 in the WT background, i.e., in the presence of the yabGWT allele, YabGWT-SNAPCd was detected around 10% of the phase-dark forespores, in 42% of the phase-grey forespores and in 81% of the phase-bright forespores (Fig 5A, yellow arrowheads). A similar localization pattern was observed for YabGWT-SNAPCd at hour 14 (S13A Fig, yellow arrowheads on the left set of panels) and at the two time points in a ΔyabG mutant (Figs 5A and S13A). Thus, consistent with an association of the protein with the coat and/or exosporium layers YabGWT-SNAPCd localizes to the forespore after engulfment completion and remains associated with the developing spore at late stages in morphogenesis. We note that for both YabGWT-SNAPCd and YabGC207A-SNAPCd, in either the WT or ΔyabG background, a haze of fluorescence is detected in the mother cell cytoplasm (Figs 5A and S13A, white arrowheads). This signal may result from release of the SNAPCd moiety through proteolysis or otherwise indicate that some of the fusion proteins remain in the mother cell (see below). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. Localization of YabG-SNAPCd in sporulating cells. A: Localization of YabGWT-SNAPCd and YabGC207A-SNAPCd in C. difficile 630Δerm (WT) and ΔyabG strains. Cells were collected after 20h of growth in 70:30 agar plates [53], stained with the SNAP substrate TMR-Star and examined by phase contrast and fluorescence microscopy (red channel for TMR signal and green channel for autofluorescence signal). The numbers refer to the percentage of cells at the represented stage showing SNAP fluorescence. Yellow and white arrowheads point to the position of the forespore and the mother cell respectively. At least 150 cells were analysed for each strain, in three independent experiments. Scale bar, 1 μm. B: Accumulation of YabG-SNAPCd and YabGC207A-SNAPCd in sporulating cells of strains 630Δerm (WT) and ΔyabG at 20h of growth in 70:30. Proteins in whole cell extracts were resolved by SDS-PAGE and the gel subject to immunoblot analysis with anti-SNAP antibodies. Samples collected from the WT and the ΔyabG mutant bearing no SNAPCd fusion were used to control for antibody specificity. The Coomassie-stained gel is included as a loading control. Red arrowheads point to the position of YabGC207A-SNAP (52 kDa). Asterisks denote possible degradation products that include the SNAP moiety (~19.4 kDa). The black arrowhead shows the position of a cross-reactive species. https://doi.org/10.1371/journal.ppat.1011741.g005 To determine whether the activity of YabG was involved in the association of the protein with the developing spore, we monitored the localization of the catalytically-inactive YabGC207A-SNAPCd fusion. At 20 hours of growth, YabGC207A-SNAPCd localized as 2 caps at the mother cell proximal and distal poles in 47% of phase-dark spores (Fig 5A). YabGC207A-SNAPCd formed a ring of fluorescence around 70% of phase-grey forespores and around 94% of phase bright-forespores (Fig 5A, yellow arrowheads). A similar pattern of localization was observed at hour 14, except that YabGC207A-SNAPCd was detected even earlier, as a single cap of fluorescence in 6% of the sporangia during engulfment (S13A Fig., yellow arrowheads). The localization of YabGC207A-SNAPCd in cells during engulfment and the higher percentages of localization of the fusion protein in sporangia of phase-dark, phase-grey and phase-bright spores suggests increased stability of the catalytically inactive protein even in the presence of the WT yabG allele. This observation suggests that the auto-proteolytic activity of YabG, detected for both the B. subtilis [52] and the C. difficile proteins (Fig 1E) controls the accumulation and localization of the protein. If so, the localization of the fusion proteins could increase in cells of a ΔyabG mutant. In comparison to the WT background, however, the localization of YabGWT-SNAPCd in ΔyabG sporangia only increased slightly around phase-dark forespores (25% as opposed to 10% in the WT at hour 20; Fig 5A) and for phase-grey forespores (65% as opposed to 55% at hour 14; S13A Fig). For the localization of YabGC207A-SNAPCd in the ΔyabG background, the main difference relative to the WT background was the increase in the single cap pattern in sporangia during engulfment (from 0 to 11% at hour 20 and from 6 to 12% at hour 14; Figs 5A and S13A). The immunoblot analysis of whole cell extracts is in good agreement with the microscopy results. YabGC207A-SNAPCd is detected with an anti-SNAP monoclonal antibody at higher levels than YabGWT-SNAPCd (both proteins run as 52 kDa species) in both the WT and the ΔyabG background with the highest accumulation corresponding to the catalytically inactive fusion in the ΔyabG mutant (red arrowheads in Figs 5B and S13B.). Bands just above and below the 20 kDa marker are likely to result from cleavage of the fusion protein close to the C-terminus of YabG (asterisks in Figs 5B and S13). The increased accumulation of YabGWT-SNAPCd or YabGC207A-SNAPCd in yabG mutants does not seem to result from augmented transcription of yabG: control experiments show that the transcription of a P yabG -SNAPCd fusion increases only slightly in ΔyabG sporangia (S14 Fig). Thus, YabGWT-SNAPCd appears to degrade itself; since no major difference was detected for YabGC207A-SNAPCd in ΔyabG sporangia in comparison to the WT, it may be that the degradation of YabG requires at least one in cis cleavage event. In any event, since YabGC207A-SNAPCd localizes earlier than the WT, the catalytic activity of YabG controls, at least in part, the localization of the protein during sporulation. In that sense, assembly of YabG is auto-regulatory.

YabG localizes asymmetrically in mature spores To gain insight onto the localization of YabG in mature spores, we used Super Resolution Structured Illumination microscopy (SR-SIM), in which the lateral resolution is increased to about 110 nm, as compared with the diffraction limit of 250 nm of conventional light microscopy [83]. Spores of strains producing either YabGWT- or YabGC207A-SNAPCd in the WT or ΔyabG backgrounds were labelled with MTG, which decorates the spore body and with TMR-Star prior to SR-SIM imaging. Both YabGWT- and YabGC207A-SNAPCd localized around the entire contour of the spore (Fig 6A, white arrowheads; see also S15 Fig), in confirmation of the conventional fluorescence microscopy data where the signal for YabG-SNAP was detected around the forespore in sporangia of phase-bright forespores (above). Strikingly, in the WT, YabGWT-SNAPCd localized to and showed a strong signal overlapping the spore polar appendage in 63% of the spores showing this structure, and YabGC207A-SNAPCd localized in 85% of those spores (Fig 6A, blue arrowheads; see also S15 Fig). In the ΔyabG mutant, YabGWT-SNAPCd localized in 63% of the appendage-bearing spores, and YabGC207A-SNAPCd localized in 46% of those spores (Fig 6A, blue arrowheads; S15 Fig). The impaired localization of the YabGC207A fusion in the ΔyabG background suggests that the WT protein facilitates the localization of the mutant form to free spores and/or its maintenance. The localization of the fusion proteins to the spore polar appendage is consistent with the role of YabG in the morphogenesis of this structure (see above). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. Localization of YabGWT- or YabGC207A-SNAPCd in mature spores. A: Localization of YabGWT- or YabGC207A-SNAPCd in mature spores using SR-SIM, in either the WT or ΔyabG backgrounds. The spores were stained with the membrane dye MTG (green) and with TMR-Star (red) prior to imaging. The blue arrows point to the SNAP signal at the spore poles and the white arrows to the signal along the side of the spore (see also S15 Fig). The distribution of the fluorescence signal (in arbitrary units, AU) in three dimensional intensity graphs is shown below the microscopy images. Scale bar, 500 nm. B: Coomassie stained SDS-PAGE gel of the proteins extracted from the cortex/coat/exosporium and core/cortex fractions of purified spores of the WT strain, the ΔyabG mutant, and of strains producing YabGWT-SNAPCd or YabGC207A-SNAPCd in either the WT or ΔyabG backgrounds. The gel was subjected to immunoblot analysis with anti-SNAP, anti-YabG, anti-SleC, anti-CdeC, anti-CotA and anti-CdeM antibodies. The arrowheads points to the position of the relevant proteins. YabG-SNAP denotes the position of either full-length WT or the C207A variant fused to the SNAP tag. Asterisks denote possible degradation products or cross-reactive species. https://doi.org/10.1371/journal.ppat.1011741.g006 Proteins in a cortex/coat/exosporium and a core/cortex fraction were resolved by SDS-PAGE and analysed by immunoblotting. Full-length YabGWT-SNAPCd, with an expected size of about 52 kDa, was barely detected in the coat/exosporium fraction of WT or ΔyabG spores with anti-SNAP or anti-YabG antibodies and was not detected in the core/cortex fraction of either strain (Fig 6B, red arrowhead). In contrast, YabGC207A-SNAPCd was detected in the cortex/coat/exosporium fraction of both WT and ΔyabG spores, and in trace amounts in the core/cortex fraction of ΔyabG spores (Fig 6B). Since full-length YabGWT-SNAPCd does not accumulate even in the ΔyabG mutant, it seems that the protein undergoes auto-proteolysis and thus, that the fusion protein is largely functional with respect to this activity (see also below). Auto-proteolysis may occur, at least in part, in cis, since full-length YabGC207A-SNAPCd was detected in the WT but at significantly higher levels in the ΔyabG mutant (Fig 6B, middle panel, red arrowhead). Because the fluorescence signal from YabGWT- or YabGC207A-SNAPCd in the SR-SIM images is comparable (Fig 6A), we infer that YabGWT-SNAPCd, although accessible to the TMR-Star substrate, is less extractable than YabGC207A-SNAPCd. Bands below the 20 kDa marker, detected in the WT for YabGWT-SNAPCd and for YabGC207A-SNAPCd, may result from cleavage of the fusion to release the SNAP moiety or fragments of it, since the C-terminally located reporter has a predicted molecular mass of 19.4 kDa (Fig 6B). As above (Fig 2A), SleC was only detected in the cortex/coat/exosporium fraction [74], whereas GPR was only detected in the core/cortex fraction (Fig 6B). Importantly, the analysis of the cortex/coat/exosporium extracts by Coomassie staining and immunoblotting showed that the YabGWT-SNAPCd fusion, but not YabGC207A-SNAPCd, restored both CotA and CdeM assembly and significant processing of Pre-pro-SleC to spores of a ΔyabG mutant and no other major differences to WT spores were noticed (Fig 6B). We infer that with respect to the assembly of the spore surface, YabGWT-SNAPCd is largely functional.

[END]
---
[1] Url: https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1011741

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.