Advertisement

  • Loading metrics

Open Access

Peer-reviewed

Research Article

The RgaS-RgaR two-component system promotes Clostridioides difficile sporulation through a small RNA and the Agr1 system

  • Adrianne N. Edwards ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Writing – original draft

    anehrli@emory.edu

    Affiliation Department of Microbiology and Immunology, Emory University School of Medicine, Emory Antibiotic Resistance Center, Atlanta, Georgia, United States of America

    ⨯
  • Shonna M. McBride

    Roles Conceptualization, Data curation, Funding acquisition, Investigation, Resources, Visualization, Writing – review & editing

    Affiliation Department of Microbiology and Immunology, Emory University School of Medicine, Emory Antibiotic Resistance Center, Atlanta, Georgia, United States of America

    ⨯

Abstract

The ability to form a dormant spore is essential for the survival of the anaerobic pathogen, Clostridioides difficile, outside of the mammalian gastrointestinal tract. The initiation of sporulation is governed by the master regulator of sporulation, Spo0A, which is activated by phosphorylation. Multiple sporulation factors control Spo0A phosphorylation; however, this regulatory pathway is not well defined in C. difficile. We discovered that RgaS and RgaR, a conserved orphan histidine kinase and orphan response regulator, function together as a cognate two-component regulatory system to directly activate transcription of several genes. One of these targets, agrB1D1, encodes gene products that synthesize and export a small quorum-sensing peptide, AgrD1, which positively influences expression of early sporulation genes. Another target, a small regulatory RNA now known as SpoZ, impacts later stages of sporulation through a small hypothetical protein and an additional, unknown regulatory mechanism(s). Unlike Agr systems in many organisms, AgrD1 does not activate the RgaS-RgaR two-component system, and thus, is not responsible for autoregulating its own production. Altogether, we demonstrate that C. difficile utilizes a conserved two-component system that is uncoupled from quorum-sensing to promote sporulation through two distinct regulatory pathways.

Author summary

The formation of an inactive spore by the anaerobic, gastrointestinal pathogen, Clostridioides difficile, is required for its survival outside of the mammalian host. The sporulation process is induced by the regulator, Spo0A; yet, how Spo0A is activated in C. difficile remains unknown. To address this question, we investigated potential activators of Spo0A. Here, we demonstrate that the sensor RgaS activates sporulation, but not by direct activation of Spo0A. Instead, RgaS activates the response regulator, RgaR, which in turn activates transcription of several genes. We found two direct RgaS-RgaR targets independently promote sporulation: agrB1D1, encoding a quorum-sensing peptide, AgrD1, and spoZ, encoding a small regulatory RNA. Unlike most other characterized Agr systems, the AgrD1 peptide does not affect RgaS-RgaR activity, indicating that AgrD1 does not activate its own production through RgaS-RgaR. Altogether, the RgaS-RgaR regulon functions at multiple points within the sporulation pathway to tightly control C. difficile spore formation.

Citation: Edwards AN, McBride SM (2023) The RgaS-RgaR two-component system promotes Clostridioides difficile sporulation through a small RNA and the Agr1 system. PLoS Genet 19(10): e1010841. https://doi.org/10.1371/journal.pgen.1010841

Editor: Aimee Shen, Tufts University School of Medicine, UNITED STATES

Received: June 26, 2023; Accepted: October 2, 2023; Published: October 16, 2023

Copyright: © 2023 Edwards, McBride. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files, and genome and transcriptomic sequence files were deposited to the NCBI Sequence Read Archive (SRA) BioProject PRJNA986905.

Funding: This research was supported by the U.S. National Institutes of Health through research grants AI116933 and AI156052 to S.M.M. The funding agency had no role in the design or execution of this study, the analysis of the data, or the decision to publish the results.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Clostridioides difficile is an anaerobic gastrointestinal pathogen that causes severe diarrheal disease [1]. The vegetative cells of this strict anaerobe undergo a dramatic morphogenesis within the mammalian gastrointestinal tract to form dormant, highly resistant endospores. This infectious spore enables long-term survival outside of the host and facilitates transmission to new hosts [2]. The initiation of spore formation requires the activation of the highly conserved transcriptional regulator, Spo0A, which is encoded in all endospore-forming bacteria [2–4]. Spo0A DNA-binding activity is controlled by the phosphorylation of a conserved aspartate residue [5,6], allowing phosphorylated Spo0A (Spo0A~P) to directly activate early sporulation-specific gene expression and triggering the entry into sporulation [5–8].

In other well-studied spore formers, most notably the soil dweller Bacillus subtilis, Spo0A phosphorylation is tightly controlled by an expansive regulatory network comprised of kinases, phosphatases, phosphotransferases, and additional regulators [9]. B. subtilis Spo0A phosphorylation is achieved via a phosphorelay that begins with sensor histidine kinases that autophosphorylate and transmit a phosphoryl group through a response regulator (Spo0F) and a phosphotransferase (Spo0B) to Spo0A. Many of the key regulatory proteins that govern Spo0A activation in B. subtilis are not well conserved or are absent in C. difficile genome [10–13], suggesting that C. difficile utilizes unique factors and regulatory pathways to initiate spore formation. Three C. difficile phosphotransfer proteins (PtpA, PtpB, and PtpC), originally annotated as orphan histidine kinases, appear to inactivate Spo0A under the conditions tested [14,15]. Thus, the primary driver of C. difficile Spo0A phosphorylation and activation is unknown.

Environmental cues and nutrient availability influence C. difficile sporulation initiation in the host gastrointestinal tract [16–20]. Additionally, bacterial global signaling systems play a significant role in the decision to initiate sporulation. The intracellular nucleotide second messenger, c-di-GMP, is an inhibitor of early sporulation in C. difficile [21]. In other spore formers, quorum-sensing systems are well established for regulating sporulation initiation [22,23]. The RRNPP family of multifunctional proteins in B. subtilis and Bacillus cereus sp. directly bind small quorum-sensing peptides to regulate their activity, which impacts Spo0A phosphorylation [24,25]. The C. difficile RRNPP ortholog, RstA, positively influences spore formation by directly binding an inhibitor of sporulation [26,27]. However, RstA differs from most members of the RRNPP family in that its cognate quorum-sensing peptide is not adjacently encoded on the genome and has yet to be discovered [26], providing additional evidence that C. difficile employs distinct regulatory pathways to initiate sporulation.

Another conserved quorum sensing system implicated in spore formation is the accessory gene regulator (Agr) system, which utilizes an extracellular, cyclic autoinducing peptide (also generically known as the AIP) as the quorum-sensing signal. The prototypical Agr system encompasses the autoinducing peptide (AgrD), a transmembrane protease (AgrB) that processes and exports AgrD, a sensor histidine kinase (AgrC), which directly senses extracellular AgrD, and the cognate response regulator (AgrA) [28]. Prior studies in Clostridium botulinum, Clostridium sporogenes, and Clostridium perfringens demonstrated that disruption or knockdown of the agrBD genes resulted in decreased spore formation [29,30]. The majority of C. difficile genomes encode an agrB1D1 operon with no apparent cognate agrA1 or agrC1 present [29,31–33]. Recent work revealed that the agrB1D1 locus positively impacts early sporulation gene expression and spore formation [34]; however, the identity of the AgrD1 receptor remains elusive.

C. difficile encodes an orphan histidine kinase (CD0576; herein, RgaS) that is annotated as an ArgC-like sensor histidine kinase. Because the sporulation factor(s) that activate Spo0A are unclear, we asked if RgaS plays a role in C. difficile sporulation initiation. Our results reveal that RgaS signals through an orphan response regulator, RgaR, to promote C. difficile spore formation. We demonstrate that the conserved residues required for phosphoryl group transfer in both RgaS and RgaR are necessary for their function as a cognate two-component system. Further, RgaSR directly activate transcription of several loci, including agrB1D1 and a small regulatory RNA, n00620 (nc063; herein referred to as spoZ). We establish that RgaSR influences spore formation through direct regulation of both agrB1D1 and spoZ transcription. Additionally, we show that the effects of AgrD1 and spoZ are independent and impact different stages of sporulation. Unlike prototypical Agr systems, RgaS does not respond to the AgrD1 peptide, indicating that AgrB1D1 has adapted distinct functions that are independent of the RgaSR two-component system. Finally, we present evidence that RgaSR promotion of spore formation through AgrD1 and SpoZ is functionally conserved in diverse C. difficile strains.

Results

The orphan histidine kinase RgaS and the RgaR response regulator promote C. difficile sporulation

The master response regulator of sporulation, Spo0A, is activated by phosphorylation. Multiple kinases activate sporulation in other spore-forming organisms; however, these factors are not conserved in C. difficile. Although three phosphotransfer proteins are proposed to act as phosphatases to inhibit Spo0A activity [15], a factor that directly phosphorylates Spo0A has not been discovered [13]. We sought to determine if other encoded kinases could activate Spo0A to drive sporulation. We identified an additional, putative orphan histidine kinase within the C. difficile genome, rgaS (CD0576), that has limited similarity to the other sporulation-associated phosphotransfer proteins, including in the dimerization and histidine autophosphorylation domain (H-box; S1 Fig). To examine whether RgaS contributes to C. difficile sporulation, we first utilized CRISPR interference (CRISPRi) to directly repress rgaS gene expression. As the addition of xylose to sporulation medium reduces sporulation frequency [15], we modified the C. difficile CRISPRi plasmid [35] to drive expression of the nuclease-deactivated version of the dCas9 gene with the nisin-inducible cprA promoter. A scrambled sgRNA (sgRNA-neg) was included as a control. We assessed the sporulation frequency of 630Δerm expressing either sgRNA-neg or sgRNA-rgaS by determining the ratio of spores to vegetative cells after 24 h of growth on 70:30 sporulation agar, with and without nisin. Expressing the anti-rgaS target resulted in a ~6-fold decrease in spore formation compared to the control strain (Fig 1A). To ensure that rgaS was directly targeted, rgaS transcripts were measured by quantitative reverse-transcription PCR (qRT-PCR), and as expected, rgaS transcripts were decreased ~25-fold in the presence of nisin (S2A Fig). This data suggests that RgaS positively impacts C. difficile sporulation.

thumbnail
Download:
Fig 1.

CRISPRi knockdown of rgaS (CD0576) and rgaR (CD3255) results in decreased sporulation frequency. (A, C) Ethanol-resistant spore formation at H24 in 630Δerm psgRNA-neg (MC2065), 630Δerm psgRNA-rgaS (MC2066), and 630Δerm psgRNA-rgaR (MC2227) grown on 70:30 agar supplemented with 2 μg/ml thiamphenicol and 1 μg/m nisin, as indicated. The means and standard deviation of three independent biological replicates (A) or standard error of the means of four independent biological replicates (C) are shown. *, P < 0.01 by a Student’s t-test. (B) Percent identity and similarity of Clostridium perfringens VirS, C. difficile AgrC2 (CDR20291_3188), and Staphylococcus aureus AgrC to the histidine kinase RgaS in C. difficile 630.

https://doi.org/10.1371/journal.pgen.1010841.g001

RgaS possesses a histidine kinase-like ATPase domain that is similar to the AgrC histidine kinase of Staphylococcus aureus (Fig 1B). Additional AgrC-like kinases have been identified in Clostridia, including the Clostridium perfringens VirS [36] and the C. difficile AgrC2, which is encoded only in the epidemic ribotype 027 strains [37]. Recent work established that C. perfringens VirS, which directly phosphorylates its cognate response regulator, VirR [38], responds to the Agr quorum sensing peptide produced by C. perfringens to activate VirR-dependent gene expression [39]. RgaS shares high similarity and identity with C. perfringens VirS and C. difficile R20291 AgrC2 (Fig 1B), suggesting that RgaS functions as a histidine kinase that directly senses an extracellular quorum sensing peptide.

We hypothesized that if RgaS is a VirS-like histidine kinase, then its cognate response regulator would be similar to C. perfringens VirR. We scanned the C. difficile 630 genome for orphan response regulators and located two VirR orthologs: CD1089 (RgbR) and CD3255 (RgaR). A prior investigation showed that overexpression of rgaR, but not rgbR, in a C. perfringens virR mutant partially complemented toxin production, suggesting that RgaR functions as a VirR ortholog [40]. Further, they discovered that RgaR directly activated transcription of three C. difficile operons, including the partial agrB1D1 locus, while an rgbR mutant did not exhibit any transcriptional changes [40]. To determine whether RgaR affects C. difficile sporulation, we knocked down rgaR gene expression by CRISPRi. C. difficile sporulation was significantly decreased (~12-fold) in the 630Δerm strain expressing sgRNA-rgaR compared to the parental control, indicating that RgaR positively influences spore formation (Fig 1C). rgaR transcript levels were decreased ~4-fold (S2B Fig), confirming that the rgaR locus was directly targeted by CRISPRi.

To further explore the impact of RgaS and RgaR on C. difficile sporulation, we generated rgaS and rgaR deletion mutants using allele-coupled exchange [41] (see Methods; S3 Fig). Sporulation phenotypes of the mutants were assessed after 24 h growth on sporulation agar. The rgaS and rgaR single mutants sporulated at lower frequencies (~7%) than the 630Δerm parent (~21%; Fig 2A), corroborating the CRISPRi knockdown results. These sporulation phenotypes were also confirmed by phase contrast microscopy (Fig 2B), as fewer phase bright spores are visible in the rgaS and rgaR mutants. We also created a double rgaS rgaR mutant and found that the double mutant exhibited similar sporulation frequencies as the single rgaS and rgaR mutants (Fig 2C), suggesting that RgaS and RgaR function in the same regulatory pathway to affect sporulation.

thumbnail
Download:
Fig 2.

C. difficile rgaS and rgaR mutants exhibit low sporulation frequencies and require their conserved histidine or aspartate residue for function. (A) Ethanol-resistant spore formation and (B) representative phase-contrast micrographs at H24 in 630Δerm, 630Δerm ΔrgaS (MC2228), and 630Δerm ΔrgaR (MC2229) grown on 70:30 agar. Scale bar denotes 10 μm. (C) Ethanol-resistant spore formation at H24 of 630Δerm, 630Δerm ΔrgaS (MC2228), 630Δerm ΔrgaR (MC2229), 630Δerm ΔrgaS ΔrgaR (MC2236), 630Δerm ΔrgaS Tn916::rgaS (MC2278), 630Δerm ΔrgaS Tn916::rgaS-H256A (MC2318), 630Δerm ΔrgaR Tn916::rgaR (MC2319), and 630Δerm ΔrgaR Tn916::rgaR-D57A (MC2320) grown on 70:30 agar. The means and standard error of the means of at least four independent biological replicates are shown. *, P < 0.01 by a one-way ANOVA followed by Dunnett’s multiple comparisons test (A) or Tukey’s multiple comparisons test (C).

https://doi.org/10.1371/journal.pgen.1010841.g002

To ensure that the decreased sporulation frequencies of the rgaS and rgaR mutants are solely the result of the deleted genes, we complemented these mutants by expressing the wild-type rgaS and rgaR alleles from their native promoters on the chromosome via the Tn916 conjugative transposon. Expression of the rgaS and rgaR alleles in their respective mutants restored sporulation frequencies to greater than wild-type levels (Fig 2C).

Our data suggest that RgaS and RgaR function in the same regulatory pathway as a two-component regulatory system (TCS), where the RgaS histidine kinase undergoes autophosphorylation at the conserved histidine residue and transfers the phosphoryl group to the conserved aspartate residue of the RgaR response regulator [42]. To determine whether the predicted functional residues of RgaS and RgaR are essential for their activity, we replaced the conserved histidine residue of RgaS and the conserved aspartate residue of RgaR with alanine, using site-directed mutagenesis, and expressed these alleles in their respective mutants via the Tn916 conjugative transposon. Neither the rgaS-H256A nor the rgaR-D57A allele complemented the low sporulation phenotypes of the respective mutants (Fig 2C), demonstrating that these residues are required for RgaS and RgaR to promote sporulation.

Late-stage sporulation is downregulated in the rgaR mutant

To better understand how the RgaS-RgaR TCS promotes sporulation, we performed RNA-seq on the rgaR mutant during growth on sporulation agar (H12) to identify differentially expressed genes during sporulation (S1 Table). The development of spores requires the activation of Spo0A followed by the sequential activation of the sporulation sigma factors, SigF, SigE, SigG, and SigK [10,43]. RNA-seq analysis revealed that only a few Spo0A- or SigF-dependent transcripts were significantly decreased in the rgaR mutant (~22% of the Spo0A regulon and ~33% of the SigF regulon; Table 1), suggesting that RgaR does not strongly influence the initiation of sporulation. To confirm that Spo0A activity is not impacted in the rgaS and rgaR mutants, we measured transcript levels of sigE, a Spo0A~P-dependent gene, by qRT-PCR (S4A Fig). There were no significant differences in sigE transcript levels between the rgaS, rgaR, and parent strains, confirming that early sporulation gene regulation is not considerably impacted by RgaSR. In contrast, ~84% of the SigE regulon, 94% of the SigG, and 97% of the SigK regulons were significantly downregulated in the rgaR mutant (Table 1), revealing that RgaR promotes the progression to later stages of spore formation.

thumbnail
Download:
Table 1. Late-stage sporulation genes are significantly downregulated in the rgaR mutant.

https://doi.org/10.1371/journal.pgen.1010841.t001

The RNA-seq results also revealed that expression of the toxin genes tcdA and tcdB were largely unaffected in the rgaR mutant, with a modest ~2-fold decrease in tcdB transcripts (S1 Table). We further assessed toxin production by measuring the total TcdA and TcdB toxin present in the supernatants of the rgaS and rgaR mutants. Toxin levels were unchanged between the rgaS mutant, the rgaR mutant, and the parent strain (S4B Fig). Altogether, these data confirm that the RgaSR TCS does not substantially impact toxin production in the conditions tested and suggests that the effects of RgaSR on sporulation regulation are independent of toxin regulation.

RgaS and RgaR have the same effects on gene expression

Prior work revealed that RgaR directly binds and activates transcription from three promoters that contain highly conserved VirR-like consensus binding sites [40]. These direct RgaR targets include the promoters for agrB1D1, the hypothetical factor CD2098, and CD0587-CD0588, a two-gene operon encoding two hypothetical proteins. Bioinformatical mining uncovered two additional VirR-like binding sites located in intergenic regions within the C. difficile genome [40], which have since been annotated. One VirR-like binding site is located upstream of a small ORF encoding a hypothetical protein, CD15111, and the second site lies immediately upstream of a predicted small regulatory RNA we have dubbed SpoZ (Fig 3A). Notably, the RNA-seq data revealed that only sporulation genes were downregulated in the rgaR mutant with the exception of these known direct RgaR targets (S1 Table), further supporting that this list encompasses all genes directly regulated by RgaR.

thumbnail
Download:
Fig 3.

RgaR-dependent gene expression is similarly decreased in both the rgaS and rgaR mutants. (A) Alignment of the putative RgaR binding sites with the promoters of confirmed and predicted direct RgaR targets. The direct repeats are in bold. (B) qRT-PCR analysis of RgaR-dependent transcripts at H12 in 630Δerm, 630Δerm ΔrgaS (MC2228), and 630Δerm ΔrgaR (MC2229) grown on 70:30 agar. The means and standard error of the means of at least three independent biological replicates are shown. *, P < 0.0001 by a one-way ANOVA followed by Dunnett’s multiple comparisons test.

https://doi.org/10.1371/journal.pgen.1010841.g003

We next asked if these direct RgaR targets are similarly regulated in the rgaS mutant. Using qRT-PCR, we assessed transcript levels of RgaR-regulated genes in sporulating cells and observed the same decrease in transcript levels for the rgaS and rgaR mutants (Fig 3B). Altogether, these data strongly support that RgaS is the cognate histidine kinase of the RgaR response regulator and that these proteins comprise a functional two-component regulatory system.

RgaR regulates sporulation through the small regulatory RNA, SpoZ

To determine which direct RgaR target is responsible for the low sporulation phenotype in the rgaS and rgaR mutants, we used CRISPRi to individually knockdown transcription of CD0587, CD2098, CD15111, and spoZ, and assessed the impacts on sporulation. Knockdown of CD0587, CD2098, or CD15111 did not alter sporulation frequency compared to the control (Fig 4A). However, the knockdown of spoZ significantly increased spore formation (Fig 4A). This was a surprising result given that the sporulation phenotype of the spoZ knockdown is opposite of what we observe for the rgaS and rgaR mutants. However, spoZ is the only known RgaR target that differentially impacts sporulation when targeted by CRISPRi, suggesting that the RgaS and RgaR influence sporulation frequency through the SpoZ small regulatory RNA. As before, we confirmed direct targeting of transcription by CRISPRi using qRT-PCR (S5 Fig).

thumbnail
Download:
Fig 4.

CRISPRi knockdown of RgaR-dependent genes reveal that RgaR impacts sporulation through the regulatory sRNA, SpoZ (encoded by CD630_n00620). (A) Ethanol-resistant spore formation of 630Δerm expressing psgRNA-neg (MC2065) or CRISPRi targets for CD0587 (MC2269), CD2098 (MC2270), CD15111 (MC2271), or spoZ (MC2272) grown on 70:30 agar supplemented with 2 μg/ml thiamphenicol and 1 μg/m nisin at H24. The means and standard errors of the means for four biological replicates are shown. (B) Alkaline phosphatase activity of either the wildtype PspoZ::phoZ or a site-directed mutagenized PspoZG-51A/G-72A::phoZ reporter fusion expressed from a plasmid in 630Δerm (MC2329 and MC2357, respectively) or 630Δerm ΔrgaR (MC2330 and MC2358, respectively). The promoterless phoZ reporter carried by 630Δerm (MC448) was included as a negative control. The means and standard deviations for at least three biological replicates are shown. *, P < 0.01 by one-way ANOVA followed by Dunnett’s multiple comparisons test (A) or Tukey’s multiple comparisons test (B).

https://doi.org/10.1371/journal.pgen.1010841.g004

Because the spoZ locus was not annotated when the original RgaR regulon was elucidated by bioinformatics, RgaR-dependent regulation of this locus had not been confirmed [40]. We created a reporter fusion of the spoZ promoter region to the alkaline phosphatase (AP) gene, phoZ [44]. We expressed PspoZ::phoZ in 630Δerm and the rgaR mutant and found that AP activity was virtually eliminated in the rgaR mutant compared to the parent strain (Fig 4B), indicating that RgaR is required for efficient spoZ expression. To confirm that the identified VirR/RgaR direct repeats are responsible for direct RgaR-dependent transcription of spoZ, we mutated the critical “G” residue within each direct repeat of the PspoZ::phoZ reporter. AP activity from the PspoZG-51A/G-72A::phoZ reporter was reduced ~5-fold in the rgaR mutant compared to the parent strain (Fig 4B), indicating that these two nucleotides within the direct repeats are necessary for direct RgaR binding to the spoZ promoter. Finally, AP activity from the PspoZG-51A/G-72A::phoZ reporter was reduced even further in the rgaR background, suggesting that RgaR still has some affinity to the mutated spoZ promoter region. Altogether, these data affirm that RgaR directly activates transcription from the spoZ promoter.

The regulatory sRNA SpoZ influences C. difficile sporulation through multiple regulatory pathways

To determine how SpoZ impacts C. difficile sporulation, we first characterized the spoZ locus. Immediately downstream from the spoZ locus is a small ORF (58 amino acids), CD16671, which is encoded on the antisense strand (Fig 5A). To confirm the expression and determine the length of the spoZ transcript, we performed near-infrared northern blotting and detected a transcript corresponding to approximately 560 nt in the 630Δerm background (Fig 5B). A similarly sized transcript, although less abundant, was observed in the R20291 background. No corresponding spoZ transcript was detected in the rgaR mutant, which was expected as RgaR is required for efficient activation of spoZ transcription. We further analyzed the cDNA synthesized from this locus in the parent strain (S6A, S6B Fig), and these data confirmed that the spoZ transcript terminates within the CD16671 promoter region (S6B Fig). Altogether, these data corroborate the previously mapped transcription termination site for spoZ, slightly beyond the CD16671 ORF (528 nt) [45], indicating that spoZ is encoded antisense to the entire CD16671 ORF. The predicted secondary structure of SpoZ was determined using UNAFold [46] and is shown in S6C Fig.

thumbnail
Download:
Fig 5.

The regulatory sRNA SpoZ influences C. difficile sporulation through multiple regulatory pathways. (A) Genomic context of the sRNA, spoZ, in relation to the small ORF, CD16671. (B) Representative irNorthern blot of spoZ in 630Δerm, 630Δerm ΔrgaR (MC2229), 630Δerm ΔspoZ-CD16671 (MC2351), and R20291 grown on 70:30 agar at H12. (C) qRT-PCR analysis of CD16671 transcript levels in 630Δerm, 630Δerm ΔrgaS (MC2228), and 630Δerm ΔrgaR (MC2229) grown on 70:30 agar at H12. (D) Ethanol-resistant spore formation of 630Δerm, 630Δerm ΔspoZ-CD16671 (MC2351), and 630Δerm ΔCD16671 (MC2363) grown on 70:30 agar at H24. (E) Ethanol-resistant spore formation at H24 of and (F) qRT-PCR analysis at H7 of sigE and sigG transcripts in 630Δerm pMC211 (MC282) and 630Δerm pCD16671 (MC2561) grown on 70:30 agar supplemented with 2 μg/ml thiamphenicol and 1 μg/ml nisin. (G) CD16671 in 630Δerm expressing psgRNA-neg (MC2065) or psgRNA-spoZ (MC2272) grown on 70:30 agar supplemented with 2 μg/ml thiamphenicol and 1 μg/m nisin. The means and standard deviation of three independent biological replicates (C, G) or standard error of the means of four independent biological replicates (D-F) are shown. *, P < 0.01 by a Student’s t-test (E-G) or by an one-way ANOVA followed by a Dunnett’s multiple comparisons test (C, D).

https://doi.org/10.1371/journal.pgen.1010841.g005

SpoZ was originally identified as a non-coding intergenic RNA and is predicted to act as an antisense target for the mRNA transcript of the neighboring ORF, CD16671 [45,47,48]. Subsequent studies discovered that SpoZ copurifies with Hfq, an RNA-binding protein that mediates post-transcriptional RNA interactions [48]. We hypothesized that SpoZ functions as an antisense RNA that targets the CD16671 transcript for Hfq-dependent turnover. We assessed CD16671 transcript levels in the rgaS and rgaR mutants and found that the CD16671 transcript is significantly increased in these mutants (Fig 5C). These results suggest that SpoZ may promote sporulation by inhibiting accumulation of the CD16671 transcript and subsequent protein product.

To test this hypothesis, we generated deletion mutants of either the entire spoZ-CD16671 locus or only CD16671 (see Methods; Figs 5B; S7A and S7B) and assessed their sporulation phenotypes. Like the spoZ knockdown strain, deletion of spoZ-CD16671 resulted in increased sporulation frequency (~2.7-fold) compared to the parent strain (Fig 5D). However, deletion of CD16671 did not impact sporulation frequency (Fig 5D). Because the decreased sporulation phenotype of the rgaS and rgaR mutants coincided with increased CD16671 transcript levels, it is possible that the absence of CD16671 does not affect C. difficile spore formation but rather the accumulation of CD16671 inhibits sporulation. To explore this possibility, we overexpressed CD16671 from an inducible promoter in the 630Δerm background and found that sporulation frequency was reduced (~1.8-fold; Fig 5E). CD16671 transcript levels were increased ~10-fold over wild-type levels (S7C Fig), suggesting that accumulation of CD16671 prevents the formation of spores. We also measured transcript levels of sigE and sigG, encoding sporulation-specific sigma factors, and found sigE and sigG transcript levels were also decreased ~1.6-fold when CD16671 was overexpressed (Fig 5F), Altogether, this data indicates that increased CD16671 abundance inhibits C. difficile sporulation-specific gene expression and spore formation.

Because the rgaS and rgaR mutants have fewer spoZ transcripts and lower sporulation frequencies, and the spoZ-CD16671 mutant and spoZ knockdown strains also have fewer spoZ transcripts but hypersporulate, we asked if the difference in the sporulation phenotypes between these strains is due to differences in CD16671 transcript levels. We revisited the spoZ CRISPRi knockdown strain, which hypersporulated (Fig 4A), and discovered that CD16671 transcript levels were only modestly increased (~3.4-fold; P = 0.06; Fig 5G), in contrast to the significant increase of CD16671 transcript levels in the rgaS and rgaR mutants (~20-fold; Fig 5C) and the CD16671 overexpression strain (~10-fold; S7C Fig). These data support the hypothesis that CD16671 accumulation inhibits sporulation, and provides additional evidence that SpoZ has an additional unidentified target(s) that mediates its effects on sporulation when CD16671 is not expressed at high levels.

Interestingly, 630Δerm encodes a CD16671 ortholog, CD19903, which shares 80% nucleotide identity and 81% amino acid identity (S8A Fig). We asked whether SpoZ also affects CD19903 transcript levels and found that there was no significant difference in CD19903 transcript levels in the rgaR mutant (S8B Fig). We observed a statistically significant, but highly variable, ~1.8-fold increase in CD19903 transcript levels in the spoZ-CD16671 mutant. These data indicate that SpoZ does not influence CD19903 in the conditions tested. It is likely that CD19903 functions similarly to CD16671 but is under different regulatory control.

Finally, the spoZ-CD16671 mutant was confirmed via whole genome sequencing (see Methods). However, despite numerous attempts with various constructs and approaches, we were unable to complement the spoZ-CD16671 mutant. In all, the regulatory portion of SpoZ is not well defined, and the regulation of the spoZ-CD16671 locus is complex. Additional studies are needed to delineate the multifaceted molecular roles SpoZ plays in C. difficile spore formation.

AgrD1 does not signal through the RgaS-RgaR two component system

The prototypical Agr signaling systems form a regulatory feedback loop, in which the quorum peptide signal (AgrD) activates the sensor kinase to initiate Agr-dependent regulation. As RgaS-RgaR directly activates agrB1D1 transcription and a previously published agrB1D1 mutant exhibits lower sporulation frequency [34], we hypothesized that the AgrD1 quorum-sensing peptide is recognized by RgaS to activate RgaR-dependent gene expression. To test this, we created an agrB1D1 deletion mutant by allelic exchange (see Methods; S9A Fig). The agrB1D1 mutant made fewer spores, similar to the rgaS and rgaR mutants, and expressing agrB1D1 from an inducible promoter restored the sporulation frequency to greater than wild-type levels (Fig 6A).

thumbnail
Download:
Fig 6.

AgrB1D1 does not activate RgaR-dependent gene expression and promotes sporulation upstream of SpoZ. (A) Ethanol-resistant spore formation of 630Δerm, 630Δerm ΔagrB1D1 (MC2433), and 630Δerm ΔagrB1D1 Tn916::PcprA-agrB1D1 (MC2524) grown on 70:30 agar at H24. (B) Transcript levels of RgaR-dependent genes, CD0587, CD2098, CD15111, and spoZ in 630Δerm and 630Δerm ΔagrB1D1 (MC2433) grown on 70:30 agar at H12. (C) Ethanol-resistant spore formation of 630Δerm, 630Δerm ΔrgaR (MC2229), 630Δerm ΔrgaR Tn916::PcprA-agrB1D1 (MC2424), and 630Δerm ΔrgaR Tn916::PcprA-spoZ-CD16671 (MC2425) grown on 70:30 agar at H24. The means and standard errors of the means for at least four biological replicates are shown. *, P < 0.01 by Student’s t-test (B) or an one-way ANOVA followed by Tukey’s multiple comparisons test (A, C).

https://doi.org/10.1371/journal.pgen.1010841.g006

We next asked whether the RgaS-RgaR-dependent gene expression was regulated by AgrD1. To answer this question, we assessed RgaR-dependent gene expression in the agrB1D1 mutant. Surprisingly, qRT-PCR analysis revealed that transcript levels of CD0587, CD2098, CD15111, and spoZ were not dramatically reduced in the agrB1D1 mutant (Fig 6B). Thus, RgaS is not the sensor for the AgrD1 peptide, as expression of RgaR-dependent genes is not affected by AgrD1. This unconnected decrease in sporulation for the agrB1D1 mutant indicates that AgrB1D1 promotes sporulation through a mechanism that is independent of RgaSR.

Prior analyses of transcription in agrB1, agrD1, and agrB1D1 mutants revealed that AgrD1 promotes transcription of early sporulation genes [34]. In contrast, disruption of rgaR most significantly impacted expression of late stage sporulation genes (Table 1), implying that SpoZ affects later stages of sporulation. As the rgaR mutant has significantly decreased transcript levels of both agrB1D1 and spoZ (Fig 3B), we asked whether AgrD1 or SpoZ promote different stages of sporulation by performing an epistasis study in the rgaR mutant. We overexpressed either the agrB1D1 or the spoZ-CD16671 locus in the rgaR mutant and assessed sporulation frequency. We anticipated that if SpoZ is epistatic to AgrD1 (that is, if SpoZ functions downstream from AgrD1 in the sporulation regulatory pathway) that only spoZ overexpression will restore sporulation in the rgaR mutant. As expected, overexpression of agrB1D1 did not increase sporulation of the rgaR mutant; however, overexpression of spoZ-CD16671 in the rgaR mutant doubled spore formation (Fig 6C), demonstrating that spoZ is able to partially complement the low sporulation phenotype of the rgaR mutant. Since AgrD1 supports early sporulation, overexpression of agrB1D1 cannot compensate for an rgaR mutant, as SpoZ must be required for later sporulation.

The RgaS-RgaR regulatory network is conserved in the epidemic R20291 background

As the agrB1D1 locus is highly conserved in almost all sequenced C. difficile genomes, we asked whether the RgaS-RgaR system is present and functions similarly in the epidemic ribotype 027 strain, R20291. The R20291 genome encodes highly conserved RgaS and RgaR orthologs, CDR20291_0503 and CDR20291_3113. Null rgaS and rgaR mutants were created in the R20291 background (see Methods; S9B Fig). We assessed sporulation frequency and observed that the rgaS and rgaR mutants produced fewer spores (~1.6-fold and ~2.1-fold, respectively) than the R20291 parent strain (Fig 7A). Transcriptional analysis of the direct RgaR targets, agrB1D1, the CD2098 ortholog (CDR20291_2005), the CD15111 ortholog (unannotated in R20291; herein referred to as CD15111^), and spoZ, revealed significantly decreased transcript levels in the rgaS and rgaR mutants (Fig 7B), indicating that the RgaS-RgaR orthologs in R20291 function similarly as in 630Δerm.

thumbnail
Download:
Fig 7.

The RgaS-RgaR-AgrB1D1-SpoZ regulatory network is conserved in R20291. (A) Ethanol-resistant spore formation at H24 and (B) qRT-PCR analyses of orthologous RgaR-dependent transcripts at H12 in R20291, R20291 ΔrgaS (MC3278), and R20291 ΔrgaR (MC3279) grown on 70:30 agar. The means and standard deviations of three independent biological replicates are shown. (C) qRT-PCR analyses of orthologous RgaR-dependent transcripts in R20291 psgRNA-neg (MC2237) and R20291 psgRNA-agrB1 (MC2534) grown on 70:30 agar at H12. (E) Representative phase-contrast micrographs of R20291 and R20291 Tn916::PcprA-spoZ-CD16671 (MC2533) and 630Δerm and 630Δerm Tn916::PcprA-spoZ-CD16671 (MC2532) grown on 70:30 agar at H24. Numbers below represent the means and standard error of the means of four independent biological replicates from ethanol-resistant sporulation assays at H24. Scale bar denotes 10 μm. ^, The CD15111 and CD16671 loci in 630Δerm are not annotated in the R20291 genome used for this study, so the 630Δerm locus tag is used as a placeholder. *, P < 0.01 by one-way ANOVA followed by Dunnett’s multiple comparisons test (A, B, D) or a Student’s t-test (C, E).

https://doi.org/10.1371/journal.pgen.1010841.g007

To further evaluate the RgaS-RgaR and AgrD1 regulatory circuitry in the R20291 background, we next asked whether the R20291 RgaS-RgaR system was responsive to the AgrD1 quorum-sensing peptide. To do this, we utilized CRISPRi to knockdown agrB1 transcription, which significantly decreased transcript levels of both agrB1 and agrD1 (Fig 7C). We then assessed transcript levels of the direct RgaR targets, CDR20291_2005, CD15111^, and spoZ, which remained unchanged in the agrB1D1 knockdown (Fig 7C). These data show that the RgaS-RgaR system also does not respond to AgrD1 in the R20291 background, indicating that this uncoupled signaling circuitry is conserved in diverse C. difficile strains.

To determine whether the regulatory sRNA SpoZ also impacts sporulation in R20291 through CD16671 (again, unannotated in R20291; herein referred to as CD16671^), we first examined CD16671^ transcript levels in the R20291 rgaS and rgaR mutants. The R20291 rgaS and rgaR mutants had modestly increased CD16671^ transcript levels (~3.8-fold and ~5.6-fold, respectively; Fig 7D), which was not as elevated as CD16671 transcript levels were in the 630Δerm background. As overexpression of CD16671 inhibits spore formation in 630Δerm, the modest increase in CD16671^ transcript levels in R20291 likely explain why the R20291 rgaS and rgaR mutants do not exhibit as dramatic sporulation defects as the 630Δerm rgaS and rgaR mutants (Fig 7A vs Fig 2A). Finally, we overexpressed the spoZ-CD16671 locus in both the R20291 and 630Δerm background via the Tn916 transposon. Sporulation frequency in the R20291 spoZ-CD16671 overexpression strain was increased 2.5-fold (Figs 7E; S10A), demonstrating that SpoZ significantly impacts spore formation in R20291 as well. Interestingly, overexpression of the spoZ-CD16671 locus in the 630Δerm background did not affect sporulation (Figs 7E; S10B). This difference in sporulation phenotypes observed between R20291 and 630Δerm when spoZ is overexpressed may be because of the relative abundance of native spoZ is less in R20291 than in 630Δerm (Fig 5B). Overall, these data indicate that the unusual RgaS-RgaR, AgrD1, and SpoZ regulatory circuitry uncovered in 630Δerm are conserved and functional in the R20291 background.

Discussion

Although the morphological changes throughout spore formation are conserved between clostridia and the well-studied bacilli, the regulatory pathways and molecular mechanisms that govern sporulation have significantly diverged. Further, the mechanism of Spo0A phosphorylation, which activates sporulation, is unclear in C. difficile. While searching for sporulation factors that facilitate C. difficile sporulation initiation, we uncovered a complex regulatory circuit that promotes sporulation through an orthologous accessory gene regulator (Agr) system. We discovered a non-contiguous, cognate two-component system, comprised of the transmembrane histidine kinase RgaS and the response regulator RgaR. RgaS-RgaR directly activate transcription of several loci, including the agrB1D1 operon and the regulatory small RNA, spoZ. We further demonstrated that the gene products of these loci, the quorum-sensing peptide, AgrD1, and the regulatory sRNA, SpoZ, impact C. difficile spore formation at different stages of sporulation. Finally, our investigation revealed that AgrD1 does not activate the RgaS-RgaR two-component system, breaking the conventional positive feedback featured in many characterized Agr systems.

To our knowledge, this is the first identification of an Agr-like two-component system that activates the transcription of the gene encoding the AIP precursor but does not serve as the receptor for the AIP. Classical Agr systems are activated through a positive feedback loop in which the histidine kinase (AgrC) autophosphorylates in response to the accumulation of AIP during high cell density conditions and subsequently transfers the phosphoryl group to the response regulator (AgrA), which in turn, activates transcription of target genes, including the genes that encode the Agr system components [49–52]. Our findings reveal an unusual Agr system in C. difficile that is not autoregulatory. Previous work demonstrated that an agrD1 mutant does not impact agrB1 transcript levels [34], corroborating that the AgrD1 peptide does not influence its own transcription. C. perfringens encodes a single agrB1D1 locus, and the VirS-VirR two component system is the AIP receptor [53,39]; however, VirR does not activate agrB1D1 transcription [54], also breaking the Agr autoregulatory loop. These examples suggest that clostridial Agr systems are not autoregulatory in general, although different mechanisms have evolved to unlink AIP accumulation to its synthesis. Considering that the C. difficile AgrD1 AIP is important for promoting spore formation independently of RgaS-RgaR, there must be another unidentified transmembrane sensor that serves as the AgrD1 receptor. Discovery of the C. difficile AgrD1 receptor will be important in elucidating the regulatory pathway through which AgrD1 promotes sporulation.

A striking feature of the RgaS-RgaR regulon is the direct transcriptional activation of a regulatory RNA. The regulon of Agr systems generally includes a gene encoding a small regulatory RNA. The most notable examples include the RNAIII regulatory RNA of Staphylococcus aureus, which is activated by AgrC-AgrA, and the VR-RNA, VirT, and VirU small regulatory RNAs in C. perfringens, which are activated by VirS-VirR [55–57]. These regulatory RNAs function as global regulators in both S. aureus and C. perfringens by influencing transcript stability and translational efficiency of target mRNAs [58–62]. Our discovery that RgaS-RgaR directly activates spoZ transcription suggests that C. difficile also coordinately regulates multiple targets through a single regulatory sRNA.

The function of SpoZ during C. difficile sporulation appears to be dependent on the accumulation of the hypothetical protein, CD16671. Low levels of spoZ transcript levels in the rgaS and rgaR mutants (~5% compared to the parent strain) led to low sporulation frequencies, while deletion or knockdown of spoZ resulted in hypersporulation. These sporulation phenotypes correlate with the abundance of the CD16671 transcript: strains with low abundance of spoZ and high levels of CD16671 exhibit low sporulation frequencies while those with low spoZ transcript levels and wild-type levels or lower of CD16671 transcripts hypersporulate (Table 2). Altogether, these data suggest that SpoZ function is hierarchal, and the inhibitory effects of CD16671 mask the regulatory functions of other SpoZ targets. The inability to complement the spoZ-CD16671 mutant may also be due to complex interplay between the spoZ and CD16671 transcripts. Additionally, a recent study employing Hfq RIL-seq identified three direct sRNA/mRNA interactions with SpoZ [63], indicating that SpoZ has multiple direct targets. These direct targets include the mRNAs of two genes, CDIF630_01110 and CDIF630_02206), encoding hypothetical proteins, and another small regulatory RNA, CDIF630nc_165, which, in turn, binds to dozens of additional targets [63]. Intriguingly, only one of these direct SpoZ targets from the Hfq RIL-seq is encoded in R20291, perhaps offering another explanation of why overexpression of the spoZ-CD16671 locus resulted in different sporulation phenotypes between the R20291 and 630Δerm background (Fig 7E). Future directions will focus on the molecular mechanisms utilized by SpoZ to influence C. difficile sporulation.

thumbnail
Download:
Table 2. CD16671 transcript levels correlate with the C. difficile sporulation phenotype in various spoZ mutants.

https://doi.org/10.1371/journal.pgen.1010841.t002

Surprisingly, we found that our 630Δerm agrB1D1 mutants exhibited a variable sporulation phenotype that was selectable on plates. However, this variation was not noted with the previously published agrB1D1 mutant, which was also constructed in the 630Δerm background. Since both low and high sporulation phenotypes can be differentiated on plates after passaging a colony exhibiting a single colony morphology, we propose that there is a phase variation event occurring with higher frequency in the agrB1D1 mutants that influences sporulation. Phase variation is a common strategy used by C. difficile to exhibit significant phenotypic heterogeneity [64–70], and there is a possibility that the Agr1 system may regulate a recombinase or other regulatory factor that influences the phase variation of sporulation.

The components of the RgaS-RgaR-SpoZ-AgrB1D1 regulatory circuitry are highly conserved in C. difficile genomes, suggesting these regulatory pathways are important in controlling earlier and later sporulation events across the species. The universal function of this system in C. difficile is further supported by data from the R20291 strain, which demonstrated conservation of the RgaS-RgaR regulon and sporulation phenotypes. In addition, bioinformatic analysis of the R20291 genome revealed the same RgaR targets, with no additional RgaR binding sites compared to the 630Δerm strain.

Our working model of the Rga regulatory system (Fig 8) begins with RgaS autophosphorylation and phosphoryl group transfer to RgaR, which then directly activates both agrB1D1 and spoZ transcription. The resulting pro-AgrD1 is processed and exported by AgrB1, accumulates extracellularly in high cell density conditions and promotes the initiation of sporulation through an unknown factor. The small regulatory RNA, SpoZ, targets one or more sRNA/mRNA(s) to influence late stage sporulation. One of these targets includes CD16671, whose expression and/or transcript stability is increased in the absence of SpoZ. CD16671 is proposed to inhibit late stage sporulation through an unknown mechanism. Finally, in the absence of CD16671 accumulation, SpoZ affects at least one additional target (denoted as X) that likely positively impacts spore formation. In all, the RgaS-RgaR regulon integrates at multiple points within the sporulation pathway to control C. difficile spore formation. Understanding when RgaSR is most active and which signals activate (or inhibit) this system will aid in understanding the environmental cues that promote C. difficile sporulation, permitting its survival in the environment and efficient transmission to new hosts.

thumbnail
Download:
Fig 8.

The model of RgaSR influence on C. difficile sporulation. The transmembrane sensor histidine kinase RgaS autophosphorylates in response to an unknown stimulus and transfers the phosphoryl group to the cytosolic response regulator RgaR. RgaR~P activates transcription of its regulon, including agrB1D1 and spoZ. The gene products of the agrB1D1 locus produce the extracellular quorum-sensing peptide, AgrD1, which promotes early stage sporulation through an unknown regulatory pathway. spoZ encodes a small regulatory RNA which influences CD16671 transcript levels, and the presumed accumulation of CD16671, which inhibits late stage spore formation. In the absence of CD16671 accumulation, SpoZ negatively impacts sporulation via unidentified sRNA/mRNA targets. Known direct interactions are depicted with solid lines while unknown regulatory pathways are denoted with dashed lines.

https://doi.org/10.1371/journal.pgen.1010841.g008

Methods

Bacterial strains and growth conditions

The bacterial strains and plasmids constructed and used in this study are listed in S2 Table. Plasmid construction details are in S3 Table. C. difficile was routinely cultured in a Coy anaerobic chamber at 37°C in an atmosphere of 10% H2, 5% CO2, and balanced with N2, as previously described and grown in brain-heart infusion (BHI) broth supplemented with yeast extract [71–73]. For plasmid maintenance or selection in C. difficile, 2–10 μg/ml thiamphenicol, 100–500 μg/ml spectinomycin, or 2–5 μg/ml erythromycin were supplemented in the media as needed. Overnight cultures were supplemented with 0.1% taurocholate to allow present spores to germinate and 0.2% fructose to prevent sporulation within the culture, as indicated [72,74]. Escherichia coli were grown at 37°C in LB medium supplemented with 100 μg/ml ampicillin, 100 μg/ml spectinomycin, 300 μg/ml erythromycin, and/or 20 μg/ml chloramphenicol as needed, and Bacillus subtilis were grown at 37°C in LB [75] supplemented with 100 μg/ml spectinomycin or 1 μg/ml erythromycin as needed. B. subtilis was also supplemented with 3.5 mM KNO3 to support anaerobic growth. Kanamycin (100 μg/ml) was used to counterselect against E. coli and B. subtilis after conjugation with C. difficile [76].

Strain construction

C. difficile 630 (GenBank accession no. NC_009089.1) and R20291 (GenBank accession no. NC_013316.1) were used as the templates for primer design, and C. difficile 630Δerm and R20291 genomic DNA were used as the template for PCR amplification and mutant construction. The C. difficile 630Δerm ΔrgaS, 630Δerm ΔrgaR, 630Δerm ΔrgaS ΔrgaR, 630Δerm ΔspoZ-CD16671, 630Δerm ΔCD16671, 630Δerm ΔagrB1D1, R20291 ΔrgaS, and R20291 ΔrgaR strains were constructed using allele-coupled exchange using either the pMSR or pMSR0 pseudo-suicide vectors for the 630 and R20291 backgrounds, respectively [41]. Briefly, pMSR/pMSR0 vectors containing the targeted gene’s 5′ and 3′ homology arms, with a selective antibiotic resistance cassette in between, were conjugated into the respective parent strain and were selected for growth on 10 μg/ml thiamphenicol and 100 μg/ml kanamycin. Colonies were then passaged to either 500 μg/ml spectinomycin or 5 μg/ml erythromycin and subsequently grown in 10 ml BHIS with the selective antibiotic and 100 μg/ml anhydrotetracycline (ATC) to induce the double recombination event. After ATC treatment, isolated colonies were screened for successful allelic exchange by PCR. Complementation strains for the rgaS and rgaR mutants were constructed by either expressing the wild-type or site-directed mutated rgaS or rgaR alleles via the conjugative transposon Tn916. rgaR mutants expressing either PcprA-agrB1D1 or PcprA-spoZ-CD16671 and the 630Δerm and R20291 strains expressing PcprA-spoZ-CD16671 were also constructed using the conjugative transposon Tn916. All strains were confirmed by PCR analysis, and most were further confirmed by whole genome sequencing (SeqCenter, Pittsburgh, PA; see below).

The Benchling CRISPR Guide RNA Design tool was used to create sgRNAs targeting rgaS, rgaR, CD0587, CD2098, CD15111, spoZ, and agrB1 [35], which were generated by PCR and subsequently cloned into pMC1123. The details of vector construction are given in the supplemental material (see S1 Text). Plasmids pMC1249 and pMC1250 were synthesized by Genscript (Piscataway, NJ).

Finally, amino acid alignments were performed with Clustal Omega [77].

Illumina whole genome sequencing

C. difficile strains were grown overnight in and harvested from 10 ml BHIS. After storing cell pellets overnight at -80°C, genomic DNA was isolated using a modified Bust ‘N’ Grab protocol [26,78]. Contaminating RNA was removed by incubation with RNase A (Ambion) for 1 h at 37°C. Library prep and Illumina sequencing were performed by SeqCenter (Pittsburgh, PA). Briefly, sample libraries were prepared using the Illumina DNA Prep Kit and IDT 10bp UDI indices, and sequenced on an Illumina NextSeq 2000, producing 2x151bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl-convert (v3.9.3; Illumina; https://support-docs.illumina.com/SW/BCL_Convert/Content/SW/FrontPages/BCL_Convert.htm). Using Geneious Prime v2022.3, the resulting reads were paired and trimmed using the BBDuk plug-in and subsequently mapped to the reference genome (630Δerm; NC_009089.1). The Bowtie2 plug-in was used to search for the presence of SNPs and InDels under default settings with a minimum variant frequency set at 0.85. Genome sequence files were deposited to the NCBI Sequence Read Archive (SRA) BioProject PRJNA986905.

Sporulation assays and phase-contrast microscopy

C. difficile cultures were grown overnight in BHIS medium supplemented with 0.1% taurocholate, to promote germination of existing spores, and 0.2% fructose, to prevent sporulation during growth [72]. C. difficile cultures were subsequently diluted with fresh BHIS, grown to mid-exponential phase (OD600 ≅ 0.5), and plated onto 70:30 agar supplemented with 2 μg/ml thiamphenicol and 1 μg/ml nisin as needed. After 24 h (H24), ethanol-resistant sporulation assays were performed as previously described [14,79]. Briefly, cells were scraped from 70:30 agar and suspended in BHIS medium to an OD600 ≅ 1.0. Total vegetative cells per milliliter were determined by immediately serially diluting suspended cells in BHIS, plating onto BHIS plates, and enumerating colony forming units (CFU) after at least 36 h of growth. Simultaneously, spores per milliliter were enumerated after a 15 min incubation in 28.5% ethanol by serially diluting in 1X PBS with 0.1% taurocholate, plating onto BHIS with 0.1% taurocholate, and counting after at least 36 h of growth. The sporulation frequency was calculated as the total number of spores divided by all viable cells (spores plus vegetative cells). A spo0A mutant was used as a negative sporulation control. Statistical significance was determined using a one-way ANOVA, followed by a Dunnett’s or Tukey’s multiple-comparison test, as indicated (GraphPad Prism 9.5.0). Phase-contrast microscopy was performed H24, using the suspended cells from 70:30 agar, with a Ph3 oil immersion objective on a Nikon Eclipse Ci-L microscope. At least two fields of view were captured with a Ds-Fi2 camera. Results represent a minimum of three independent biological experiments.

Enzyme-linked immunosorbent assay (ELISA)

C. difficile TcdA and TcdB from culture supernatants was quantified after 24 h growth of C. difficile strains in TY, pH 7.4, as previously described [80]. Supernatants, diluted in the provided dilution buffer, were assayed in technical duplicates using the tgcBIOMICS kit for simultaneous detection of both TcdA and TcdB, according to the manufacturer’s instructions. The results represent the means and standard error of the means of four independent biological replicates, and statistical significance was assessed via a one-way ANOVA, followed by a Dunnett’s multiple-comparison test (GraphPad Prism 9.5.0).

Quantitative reverse transcription PCR analysis (qRT-PCR)

C. difficile strains were cultured on 70:30 agar as described above, and cells were harvested at H12 (defined as 12 h after the cultures are applied to the plate) into 6 ml of 1:1:2 ethanol:acetone:dH2O solution and stored at -80°C. RNA was isolated and Dnase-I treated (Ambion), and cDNA was synthesized (Bioline) using random hexamers as previously described [16,18,81]. qRT-PCR analysis was performed in triplicate on 50 ng of cDNA using the SensiFAST SYBR & Fluorescein kit (Bioline) with a Roche Lightcycler 96. The results were analyzed by the comparative cycle threshold method [82], using the rpoC transcript as the normalizer, and are presented as the means and standard error of the means for a minimum of three independent biological replicates. An one-way ANOVA, followed by a Student’s t-test or a Dunnett’s multiple-comparison test as applicable (GraphPad Prism v9.5.0), was used to assess statistical significance.

RNA sequencing (RNA-seq) analysis

C. difficile strains were cultured on 70:30 sporulation agar as described above and harvested at H12 into 6 ml of 1:1:2 ethanol:acetone:dH2O solution and stored at -80°C. RNA was isolated and Dnase-I treated (Ambion). Samples were sent to Microbial Genomics Sequencing Center (MiGS; Pittsburgh, PA) where library preparation was performed using Illumina’s Stranded Total RNA prep Ligation with Ribo-Zero Plus kit and 10bp IDT for Illumina indices. Sequencing was done on a NextSeq200 giving 2x50bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl-convert (v3.9.3; Illumina; see reference above). Using Geneious Prime v2022.2.2, the reads were mapped to the reference genome (630Δerm; NC_009089.1). The expression levels were calculated and then subsequently compared using DESeq2 [83]. DESeq2 utilizes the Wald test to calculate P values which are then adjusted using the Benjamini-Hochberg test [83]. RNA-seq raw sequence reads were deposited to the NCBI Sequence Read Archive (SRA) BioProject PRJNA986905.

Alkaline phosphatase activity assays

C. difficile strains were cultured on 70:30 sporulation agar as described above and harvested at H8 (defined as eight hours after the cultures are applied to the plates). Alkaline phosphatase (AP) assays were performed as previously described [44] excepting the exclusion of chloroform. Technical duplicates were averaged and results represent the means and standard error of the means of three independent biological replicates. An one-way ANOVA, followed by a Tukey’s multiple-comparison test (GraphPad Prism v9.5.0), was used to assess statistical significance.

Northern blot analysis

Near-infrared fluorescent northern (irNorthern) blot analysis was performed as previously described [84]. Briefly, 2.5 nmol of a spoZ DNA probe (oMC3901), containing an internal azide modification (IDT, Coralville, IA), was fluorescently labeled with 50 nmol IRdye 800 CW (DBCO) in 1X phosphate buffered saline, pH 7.4 for 6 h at 25°C. IRdye-labeled oligonucleotides were purified with Ampure XP beads (Beckman Coulter) as described [84], eluted in 50 μl nuclease-free dH2O, and stored at -70°C. For the irNorthern blot, 20 μg of RNA was diluted 1:1 with Glyoxal Load Dye (ThermoFisher Scientific), heated to 50°C for 30 min, and electrophoresed on a 1.4% agarose gel. The RNA was transferred to a BrightStar-Plus positively charged nylon membrane (ThermoFisher Scientific) using downward capillary transfer, and the RNA was UV crosslinked to the membrane at 120 mJ/cm2 using a Stratagene UV Stratalinker 1800. The membrane was prehybridized overnight at 37°C in ULTRAhyb Hybridization Buffer (ThermoFisher Scientific), and hybridized with 1.3 nM IRdye-labeled DNA probe for 4 h at 37°C. The membrane was washed once with Low Stringency Wash Solution 1 (ThermoFisher Scientific) for 10 min at room temperature, twice for 3 min at 37°C. and subsequently visualized on a BioRad ChemiDoc MP. The size of the SpoZ transcript was estimated based on a standard curve calculated by the migration of the RNA Century Plus Marker (Ambion).

Supporting information

S1 Fig. Amino acid alignment of RgaS with the sporulation-associated phosphotransfer proteins, PtpA, PtpB, and PtpC.

The dimerization and histidine autophosphorylation domain (DHpt, or H-box) of the three sporulation-associated phosphotransfer proteins, PtpA (CD630_14920), PtpB (CD630_24920), and PtpC (CD630_15790) aligned with RgaS using Clustal Omega.

https://doi.org/10.1371/journal.pgen.1010841.s001

(TIFF)

S2 Fig. CRISPRi knockdown of rgaS and rgaR decreases rgaS and rgaR transcript levels.

Transcript levels of rgaS (A) and rgaR (B) at H12 in 630Δerm psgRNA-neg (MC2065), 630Δerm psgRNA-rgaS (MC2066), and 630Δerm psgRNA-rgaR (MC2227) grown on 70:30 agar supplemented with 2 μg/ml thiamphenicol and 1 μg/m nisin, as indicated. The means and standard deviation of three independent biological replicates are shown. *, P < 0.01 by a Student’s t-test.

https://doi.org/10.1371/journal.pgen.1010841.s002

(TIFF)

S3 Fig. PCR confirmation of rgaS and rgaR mutants.

PCR verification of allelic replacement of rgaS (CD0576) and rgaR (CD3255) with the spectinomycin (aad9) or erythromycin (ermB) cassette, respectively in 630Δerm, 630Δerm ΔrgaS (MC2228), and 630Δerm ΔrgaR (MC2229). Expected PCR products’ sizes are: 3319 bp for the wildtype rgaS allele and 3078 bp for the ΔrgaS::aad9 allele (primers oMC3123/3124); 2668 bp for the wildtype rgaR allele and 3436 bp for the ΔrgaR::ermB allele (primers oMC3261/3262).

https://doi.org/10.1371/journal.pgen.1010841.s003

(TIFF)

S4 Fig. Spo0A-dependent gene expression and toxin production are unaffected in C. difficile rgaS and rgaR mutants.

(A) qRT-PCR analysis of sigE in 630Δerm, 630Δerm ΔrgaS (MC2228), and 630Δerm ΔrgaR (MC2229) grown on 70:30 agar at H12. (B) ELISA of TcdA and TcdB present in the supernatant of 630Δerm, 630Δerm ΔrgaS (MC2228), and 630Δerm ΔrgaR (MC2229) grown in TY medium, pH 7.4, at H24. The means and standard error of the means of four independent biological replicates are shown. The statistical analysis used for these data sets was a one-way ANOVA followed by Dunnett’s multiple comparisons test.

https://doi.org/10.1371/journal.pgen.1010841.s004

(TIFF)

S5 Fig. CRISPRi knockdown of RgaR-dependent genes specifically and significantly decreases their transcript levels.

Transcript levels and of (A) CD0587, CD2098, CD15111, and spoZ at H12 in 630Δerm expressing psgRNA-neg (MC2065) or CRISPRi targets for CD0587 (MC2269), CD2098 (MC2270), CD15111 (MC2271), or spoZ (MC2272) grown on 70:30 agar supplemented with 2 μg/ml thiamphenicol and 1 μg/m nisin. The means and standard deviations of three independent biological replicates are shown. *, P < 0.0001 by a Student’s t-test.

https://doi.org/10.1371/journal.pgen.1010841.s005

(TIFF)

S6 Fig. PCR validation of spoZ transcript length and predicted secondary structure.

(A) Primers spanning the regions within the spoZ-CD16671 locus depicted in (B), were used to amplify cDNA and RNA (minus reverse transcriptase negative control; -RT) from 630Δerm grown on 70:30 agar at H12 and 630Δerm genomic DNA (gDNA). (C) Predicted secondary structure of the 528 nt SpoZ sRNA as calculated by UNAFold. Shown is the structure with the lowest free energy.

https://doi.org/10.1371/journal.pgen.1010841.s006

(TIFF)

S7 Fig. PCR confirmation of spoZ-CD16671 and CD16671 mutants and qRT-PCR analysis of transcript levels of CD16671 in a CD16671 overexpression strain.

(A, B) PCR verification of allelic replacement of the spoZ-CD16671 locus with the spectinomycin (aad9) cassette in 630Δerm, 630Δerm ΔspoZ-CD16671 (MC2351; A) and 630Δerm ΔCD16671 (MC2363; B). Expected PCR products’ sizes are: 2165 bp for the wildtype spoZ-CD16671 allele, 2745 bp for the ΔspoZ-CD16671::aad9 allele, and 3033 bp for the ΔCD16671::aad9 allele (primers oMC3457/3458). (C) qRT-PCR of CD16671 transcript levels in 630Δerm pMC211 (MC282) and 630Δerm pCD16671 (MC2561) grown on 70:30 agar at H7. The means and standard error of the mean is shown for five independent biological replicates. *, P < 0.001 by a Student’s t-test.

https://doi.org/10.1371/journal.pgen.1010841.s007

(TIFF)

S8 Fig. Transcript levels of CD19903, encoding a CD16671 ortholog, are not impacted by RgaR or SpoZ.

(A) Amino acid alignment of CD16671 and CD19903 using Clustal Omega (Sievers, et al. 2011 Mol Sys Biol. 7:539). (B) qRT-PCR analysis of CD19903 transcript levels in 630Δerm, 630Δerm ΔrgaR (MC2229), and 630Δerm ΔspoZ-CD16671 (MC2351) at H12 on 70:30 agar. *, P = 0.03 by one-way ANOVA followed by Dunnett’s multiple comparisons test.

https://doi.org/10.1371/journal.pgen.1010841.s008

(TIFF)

S9 Fig. PCR confirmation of 630Δerm agrB1D1 mutants and R20291 CDR20291_0503 (rgaS) and CDR20291_3113 mutants (rgaR).

(A) PCR verification of allelic replacement of the agrB1D1 locus with the spectinomycin cassette (aad9) in 630Δerm and 630Δerm ΔagrB1D1 Expected PCR sizes are 2585 bp for the wild-type allele and 2930 bp for the agrB1D1::aad9 allele (primers oMC3360/3361). (B) PCR verification of allelic replacement of rgaS and rgaR with the spectinomycin (aad9) cassette in R20291, R20291 ΔrgaS (MC2378), and R20291 ΔrgaR (MC2379). Expected PCR products’ sizes are: 3215 bp for the wildtype rgaS allele and 2974 bp for the ΔrgaS::aad9 allele (primers oMC3124/3554); 2667 bp for the wildtype rgaR allele and 3056 bp for the ΔrgaR::aad9 allele (primers oMC3261/3262).

https://doi.org/10.1371/journal.pgen.1010841.s009

(TIFF)

S10 Fig. Overexpression of spoZ differentially impacts sporulation in the R20291 background compared to 630Δerm.

Ethanol-resistant sporulation assays in (A) R20291 and R20291 Tn916::PcprA-spoZ-CD16671 (MC2533) and (B) 630Δerm and 630Δerm Tn916::PcprA-spoZ-CD16671 (MC2532) on 70:30 agar at H24. The means and standard errors of the means of four independent biological replicates are shown.

https://doi.org/10.1371/journal.pgen.1010841.s010

(TIFF)

S1 Text. DNA cloning and vector details.

https://doi.org/10.1371/journal.pgen.1010841.s011

(DOCX)

S1 Table. RNA-seq of rgaR mutant compared to 630Δerm.

https://doi.org/10.1371/journal.pgen.1010841.s012

(XLSX)

S2 Table. Bacterial strains and plasmids.

https://doi.org/10.1371/journal.pgen.1010841.s013

(DOCX)

S3 Table. Oligonucleotides.

https://doi.org/10.1371/journal.pgen.1010841.s014

(DOCX)

Acknowledgments

We are grateful to the members of the McBride lab for their helpful suggestions, comments, and discussions throughout the course of this study. We are also thankful to Craig Ellermeier for the gift of pIA33. The content of this manuscript is solely the responsibility of the authors and does not necessarily reflect the official views of the National Institutes of Health.

References

  1. 1. Smits WK, Lyras D, Lacy DB, Wilcox MH, Kuijper EJ. Clostridium difficile infection. Nat Rev Primer. 2016;2: 16020. pmid:27158839
  2. 2. Deakin LJ, Clare S, Fagan RP, Dawson LF, Pickard DJ, West MR, et al. The Clostridium difficile spo0A gene is a persistence and transmission factor. Infect Immun. 2012;80: 2704–11. pmid:22615253
  3. 3. Ferrari FA, Trach K, LeCoq D, Spence J, Ferrari E, Hoch JA. Characterization of the spo0A locus and its deduced product. Proc Natl Acad Sci U S A. 1985;82: 2647–51. pmid:3157992
  4. 4. Rosenbusch KE, Bakker D, Kuijper EJ, Smits WK. C. difficile 630Δerm Spo0A regulates sporulation, but does not contribute to toxin production, by direct high-affinity binding to target DNA. PLoS One. 2012;7: e48608. pmid:23119071
  5. 5. Ireton K, Rudner DZ, Siranosian KJ, Grossman AD. Integration of multiple developmental signals in Bacillus subtilis through the Spo0A transcription factor. Genes Dev. 1993;7: 283–294. pmid:8436298
  6. 6. Sonenshein AL. Control of sporulation initiation in Bacillus subtilis. Curr Opin Microbiol. 2000;3: 561–6.
  7. 7. Fimlaid KA, Bond JP, Schutz KC, Putnam EE, Leung JM, Lawley TD, et al. Global Analysis of the Sporulation Pathway of Clostridium difficile. PLoS Genet. 2013;9: e1003660. pmid:23950727
  8. 8. DiCandia MA, Edwards AN, Jones JB, Swaim GL, Mills BD, McBride SM. Identification of Functional Spo0A Residues Critical for Sporulation in Clostridioides difficile. J Mol Biol. 2022;434: 167641. pmid:35597553
  9. 9. Hoch JA. Regulation of the phosphorelay and the initiation of sporulation in Bacillus subtilis. Annu Rev Microbiol. 1993;47: 441–465. pmid:8257105
  10. 10. Paredes CJ, Alsaker KV, Papoutsakis ET. A comparative genomic view of clostridial sporulation and physiology. Nat Rev Microbiol. 2005;3: 969–978. pmid:16261177
  11. 11. Edwards AN, McBride SM. Initiation of sporulation in Clostridium difficile: a twist on the classic model. FEMS Microbiol Lett. 2014;358: 110–8. pmid:24910370
  12. 12. Shen A, Edwards AN, Sarker MR, Paredes-Sabja D. Sporulation and Germination in Clostridial Pathogens. Microbiol Spectr. 2019;7. pmid:31858953
  13. 13. Lee CD, Rizvi A, Edwards AN, DiCandia MA, Vargas Cuebas GG, Monteiro MP, et al. Genetic mechanisms governing sporulation initiation in Clostridioides difficile. Curr Opin Microbiol. 2022;66: 32–38. pmid:34933206
  14. 14. Childress KO, Edwards AN, Nawrocki KL, Woods EC, Anderson SE, McBride SM. The Phosphotransfer Protein CD1492 Represses Sporulation Initiation in Clostridium difficile. Infect Immun. 2016. pmid:27647869
  15. 15. Edwards AN, Wetzel D, DiCandia MA, McBride SM. Three Orphan Histidine Kinases Inhibit Clostridioides difficile Sporulation. J Bacteriol. 2022;204: e0010622. pmid:35416689
  16. 16. Dineen SS, McBride SM, Sonenshein AL. Integration of metabolism and virulence by Clostridium difficile CodY. J Bacteriol. 2010;192: 5350–62. pmid:20709897
  17. 17. Antunes A, Martin-Verstraete I, Dupuy B. CcpA-mediated repression of Clostridium difficile toxin gene expression: C. difficile toxin regulation by CcpA. Mol Microbiol. 2011;79: 882–899. pmid:21299645
  18. 18. Edwards AN, Nawrocki KL, McBride SM. Conserved oligopeptide permeases modulate sporulation initiation in Clostridium difficile. Infect Immun. 2014;82: 4276–91. pmid:25069979
  19. 19. Nawrocki KL, Edwards AN, Daou N, Bouillaut L, McBride SM. CodY-Dependent Regulation of Sporulation in Clostridium difficile. J Bacteriol. 2016;198: 2113–30. pmid:27246573
  20. 20. Wetzel D, McBride SM. The impact of pH on Clostridioides difficile sporulation and physiology. Appl Env Microbiol. 2019; AEM.02706-19. pmid:31811041
  21. 21. Edwards AN, Willams CL, Pareek N, McBride SM, Tamayo R. c-di-GMP inhibits early sporulation in Clostridioides difficile. mSphere. 2021 6(6): e0091921. pmid:34878288
  22. 22. Lazazzera BA, Solomon JM, Grossman AD. An exported peptide functions intracellularly to contribute to cell density signaling in B. subtilis. Cell. 1997;89: 917–25. pmid:9200610
  23. 23. Pottathil M, Lazazzera BA. The extracellular Phr peptide-Rap phosphatase signaling circuit of Bacillus subtilis. Front Biosci. 2003;8: d32–45. pmid:12456319
  24. 24. Rocha J, Flores V, Cabrera R, Soto-Guzmán A, Granados G, Juaristi E, et al. Evolution and some functions of the NprR–NprRB quorum-sensing system in the Bacillus cereus group. Appl Microbiol Biotechnol. 2012;94: 1069–1078. pmid:22159892
  25. 25. Neiditch MB, Capodagli GC, Prehna G, Federle MJ. Genetic and Structural Analyses of RRNPP Intercellular Peptide Signaling of Gram-Positive Bacteria. Annu Rev Genet. 2017;51: 311–333. pmid:28876981
  26. 26. Edwards AN, Tamayo R, McBride SM. A novel regulator controls Clostridium difficile sporulation, motility and toxin production. Mol Microbiol. 2016;100: 954–71. pmid:26915493
  27. 27. DiCandia MA, Edwards AN, Lee CD, Monteiro MP, Cuebas GNV, Bagchi P, et al. A Conserved Switch Controls Virulence, Sporulation, and Motility in C. difficile. BioRxiv [Preprint]. 2023 bioRxiv 534590 [posted 2023 Mar 28; cited 2023 Sep 14] Available from: https://www.biorxiv.org/content/10.1101/2023.03.28.534590v1. pmid:37034656
  28. 28. Ji G, Beavis RC, Novick RP. Cell density control of staphylococcal virulence mediated by an octapeptide pheromone. Proc Natl Acad Sci U S A. 1995;92: 12055–12059. pmid:8618843
  29. 29. Cooksley CM, Davis IJ, Winzer K, Chan WC, Peck MW, Minton NP. Regulation of neurotoxin production and sporulation by a Putative agrBD signaling system in proteolytic Clostridium botulinum. Appl Env Microbiol. 2010;76: 4448–60. pmid:20453132
  30. 30. Li J, Chen J, Vidal JE, McClane BA. The Agr-like quorum-sensing system regulates sporulation and production of enterotoxin and beta2 toxin by Clostridium perfringens type A non-food-borne human gastrointestinal disease strain F5603. Infect Immun. 2011;79: 2451–2459. pmid:21464088
  31. 31. Sebaihia M, Wren BW, Mullany P, Fairweather NF, Minton N, Stabler R, et al. The multidrug-resistant human pathogen Clostridium difficile has a highly mobile, mosaic genome. Nat Genet. 2006;38: 779–786. pmid:16804543
  32. 32. Stabler RA, He M, Dawson L, Martin M, Valiente E, Corton C, et al. Comparative genome and phenotypic analysis of Clostridium difficile 027 strains provides insight into the evolution of a hypervirulent bacterium. Genome Biol. 2009;10: R102. pmid:19781061
  33. 33. Okada Y, Okugawa S, Ikeda M, Kobayashi T, Saito R, Higurashi Y, et al. Genetic diversity and epidemiology of accessory gene regulator loci in Clostridioides difficile. Access Microbiol. 2020;2: acmi000134. pmid:32974597
  34. 34. Ahmed UKB, Shadid TM, Larabee JL, Ballard JD. Combined and Distinct Roles of Agr Proteins in Clostridioides difficile 630 Sporulation, Motility, and Toxin Production. mBio. 2020;11. pmid:33443122
  35. 35. Müh U, Pannullo AG, Weiss DS, Ellermeier CD. A Xylose-Inducible Expression System and a CRISPR Interference Plasmid for Targeted Knockdown of Gene Expression in Clostridioides difficile. J Bacteriol. 2019;201. pmid:30745377
  36. 36. Lyristis M, Bryant AE, Sloan J, Awad MM, Nisbet IT, Stevens DL, et al. Identification and molecular analysis of a locus that regulates extracellular toxin production in Clostridium perfringens. Mol Microbiol. 1994;12: 761–777. pmid:8052128
  37. 37. Martin MJ, Clare S, Goulding D, Faulds-Pain A, Barquist L, Browne HP, et al. The agr locus regulates virulence and colonization genes in Clostridium difficile 027. J Bacteriol. 2013;195: 3672–81. pmid:23772065
  38. 38. Cheung JK, Awad MM, McGowan S, Rood JI. Functional analysis of the VirSR phosphorelay from Clostridium perfringens. PloS One. 2009;4: e5849. pmid:19513115
  39. 39. Li J, McClane BA. Evidence That VirS Is a Receptor for the Signaling Peptide of the Clostridium perfringens Agr-like Quorum Sensing System. mBio. 2020;11: e02219–20. pmid:32934089
  40. 40. O’Connor JR, Lyras D, Farrow KA, Adams V, Powell DR, Hinds J, et al. Construction and analysis of chromosomal Clostridium difficile mutants. Mol Microbiol. 2006;61: 1335–51. pmid:16925561
  41. 41. Peltier J, Hamiot A, Garneau JR, Boudry P, Maikova A, Hajnsdorf E, et al. Type I toxin-antitoxin systems contribute to the maintenance of mobile genetic elements in Clostridioides difficile. Commun Biol. 2020;3: 718. pmid:33247281
  42. 42. Hoch JA. Two-component and phosphorelay signal transduction. Curr Opin Microbiol. 2000;3: 165–70. pmid:10745001
  43. 43. Haldenwang WG. The sigma factors of Bacillus subtilis. Microbiol Rev. 1995;59: 1–30. pmid:7708009
  44. 44. Edwards AN, Pascual RA, Childress KO, Nawrocki KL, Woods EC, McBride SM. An alkaline phosphatase reporter for use in Clostridium difficile. Anaerobe. 2015;32: 98–104. pmid:25576237
  45. 45. Fuchs M, Lamm-Schmidt V, Sulzer J, Ponath F, Jenniches L, Kirk JA, et al. An RNA-centric global view of Clostridioides difficile reveals broad activity of Hfq in a clinically important gram-positive bacterium. Proc Natl Acad Sci U S A. 2021;118. pmid:34131082
  46. 46. Zuker M. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 2003;31: 3406–3415. pmid:12824337
  47. 47. Soutourina OA, Monot M, Boudry P, Saujet L, Pichon C, Sismeiro O, et al. Genome-wide identification of regulatory RNAs in the human pathogen Clostridium difficile. PLoS Genet. 2013;9: e1003493. pmid:23675309
  48. 48. Boudry P, Piattelli E, Drouineau E, Peltier J, Boutserin A, Lejars M, et al. Identification of RNAs bound by Hfq reveals widespread RNA partners and a sporulation regulator in the human pathogen Clostridioides difficile. RNA Biol. 2021;18: 1931–1952. pmid:33629931
  49. 49. Novick RP, Projan SJ, Kornblum J, Ross HF, Ji G, Kreiswirth B, et al. The agr P2 operon: an autocatalytic sensory transduction system in Staphylococcus aureus. Mol Gen Genet MGG. 1995;248: 446–458. pmid:7565609
  50. 50. Lina G, Jarraud S, Ji G, Greenland T, Pedraza A, Etienne J, et al. Transmembrane topology and histidine protein kinase activity of AgrC, the agr signal receptor in Staphylococcus aureus. Mol Microbiol. 1998;28: 655–662. pmid:9632266
  51. 51. Koenig RL, Ray JL, Maleki SJ, Smeltzer MS, Hurlburt BK. Staphylococcus aureus AgrA binding to the RNAIII-agr regulatory region. J Bacteriol. 2004;186: 7549–7555.
  52. 52. Queck SY, Jameson-Lee M, Villaruz AE, Bach T-HL, Khan BA, Sturdevant DE, et al. RNAIII-independent target gene control by the agr quorum-sensing system: insight into the evolution of virulence regulation in Staphylococcus aureus. Mol Cell. 2008;32: 150–158. pmid:18851841
  53. 53. Ohtani K, Yuan Y, Hassan S, Wang R, Wang Y, Shimizu T. Virulence gene regulation by the agr system in Clostridium perfringens. J Bacteriol. 2009;191: 3919–3927. pmid:19363118
  54. 54. Shimizu T, Ohtani K, Hirakawa H, Ohshima K, Yamashita A, Shiba T, et al. Complete genome sequence of Clostridium perfringens, an anaerobic flesh-eater. Proc Natl Acad Sci U S A. 2002;99: 996–1001. pmid:11792842
  55. 55. Novick RP, Ross HF, Projan SJ, Kornblum J, Kreiswirth B, Moghazeh S. Synthesis of staphylococcal virulence factors is controlled by a regulatory RNA molecule. EMBO J. 1993;12: 3967–3975. pmid:7691599
  56. 56. Shimizu T, Yaguchi H, Ohtani K, Banu S, Hayashi H. Clostridial VirR/VirS regulon involves a regulatory RNA molecule for expression of toxins. Mol Microbiol. 2002;43: 257–265. pmid:11849553
  57. 57. Okumura K, Ohtani K, Hayashi H, Shimizu T. Characterization of genes regulated directly by the VirR/VirS system in Clostridium perfringens. J Bacteriol. 2008;190: 7719–7727. pmid:18790863
  58. 58. Le KY, Otto M. Quorum-sensing regulation in staphylococci—an overview. Front Microbiol. 2015;6. pmid:26579084
  59. 59. Bronesky D, Wu Z, Marzi S, Walter P, Geissmann T, Moreau K, et al. Staphylococcus aureus RNAIII and Its Regulon Link Quorum Sensing, Stress Responses, Metabolic Adaptation, and Regulation of Virulence Gene Expression. Annu Rev Microbiol. 2016;70: 299–316. pmid:27482744
  60. 60. Ohtani K, Hirakawa H, Tashiro K, Yoshizawa S, Kuhara S, Shimizu T. Identification of a two-component VirR/VirS regulon in Clostridium perfringens. Anaerobe. 2010;16: 258–264. pmid:19835966
  61. 61. Obana N, Shirahama Y, Abe K, Nakamura K. Stabilization of Clostridium perfringens collagenase mRNA by VR-RNA-dependent cleavage in 5’ leader sequence. Mol Microbiol. 2010;77: 1416–1428. pmid:20572941
  62. 62. André G, Haudecoeur E, Monot M, Ohtani K, Shimizu T, Dupuy B, et al. Global regulation of gene expression in response to cysteine availability in Clostridium perfringens. BMC Microbiol. 2010;10: 234. pmid:20822510
  63. 63. Fuchs M, Lamm-Schmidt V, Lenče T, Sulzer J, Bublitz A, Wackenreuter J, et al. A network of small RNAs regulates sporulation initiation in Clostridioides difficile. EMBO J. 2023;42: e112858. pmid:37140366
  64. 64. Emerson JE, Reynolds CB, Fagan RP, Shaw HA, Goulding D, Fairweather NF. A novel genetic switch controls phase variable expression of CwpV, a Clostridium difficile cell wall protein. Mol Microbiol. 2009;74: 541–56. pmid:19656296
  65. 65. Reynolds CB, Emerson JE, de la Riva L, Fagan RP, Fairweather NF. The Clostridium difficile cell wall protein CwpV is antigenically variable between strains, but exhibits conserved aggregation-promoting function. PLoS Pathog. 2011;7: e1002024. pmid:21533071
  66. 66. Sekulovic O, Ospina Bedoya M, Fivian-Hughes AS, Fairweather NF, Fortier L-C. The Clostridium difficile cell wall protein CwpV confers phase-variable phage resistance. Mol Microbiol. 2015;98: 329–342. pmid:26179020
  67. 67. Anjuwon-Foster BR, Tamayo R. A genetic switch controls the production of flagella and toxins in Clostridium difficile. PLoS Genet. 2017;13: e1006701. pmid:28346491
  68. 68. Sekulovic O, Mathias Garrett E, Bourgeois J, Tamayo R, Shen A, Camilli A. Genome-wide detection of conservative site-specific recombination in bacteria. PLoS Genet. 2018;14: e1007332. pmid:29621238
  69. 69. Anjuwon-Foster BR, Maldonado-Vazquez N, Tamayo R. Characterization of Flagellum and Toxin Phase Variation in Clostridioides difficile Ribotype 012 Isolates. J Bacteriol. 2018;200: e00056–18. pmid:29735765
  70. 70. Garrett EM, Sekulovic O, Wetzel D, Jones JB, Edwards AN, Vargas-Cuebas G, et al. Phase variation of a signal transduction system controls Clostridioides difficile colony morphology, motility, and virulence. PLoS Biol. 2019;17: e3000379. pmid:31658249
  71. 71. Smith CJ, Markowitz SM, Macrina FL. Transferable tetracycline resistance in Clostridium difficile. Antimicrob Agents Chemother. 1981;19: 997–1003. pmid:7271279
  72. 72. Sorg JA, Dineen SS. Laboratory maintenance of Clostridium difficile. Curr Protoc Microbiol. 2009;Chapter 9: Unit9A 1. pmid:19235151
  73. 73. Edwards AN, Suarez JM, McBride SM. Culturing and maintaining Clostridium difficile in an anaerobic environment. J Vis Exp. 2013; e50787. pmid:24084491
  74. 74. Putnam EE, Nock AM, Lawley TD, Shen A. SpoIVA and SipL are Clostridium difficile spore morphogenetic proteins. J Bacteriol. 2013;195: 1214–25. pmid:23292781
  75. 75. Luria SE, Burrous JW. Hybridization between Escherichia coli and Shigella. J Bacteriol. 1957;74: 461–476. pmid:13475269
  76. 76. Purcell EB, McKee RW, McBride SM, Waters CM, Tamayo R. Cyclic diguanylate inversely regulates motility and aggregation in Clostridium difficile. J Bacteriol. 2012;194: 3307–16. pmid:22522894
  77. 77. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7: 539. pmid:21988835
  78. 78. Harju S, Fedosyuk H, Peterson KR. Rapid isolation of yeast genomic DNA: Bust n’ Grab. BMC Biotechnol. 2004;4: 8. pmid:15102338
  79. 79. Edwards AN, McBride SM. Determination of the in vitro Sporulation Frequency of Clostridium difficile. Bio-Protoc. 2017;7. pmid:28516125
  80. 80. Edwards AN, Krall EG, McBride SM. Strain-Dependent RstA Regulation of Clostridioides difficile Toxin Production and Sporulation. Federle MJ, editor. J Bacteriol. 2020;202. pmid:31659010
  81. 81. McBride SM, Sonenshein AL. Identification of a genetic locus responsible for antimicrobial peptide resistance in Clostridium difficile. Infect Immun. 2011;79: 167–76. pmid:20974818
  82. 82. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc. 2008;3: 1101–8. pmid:18546601
  83. 83. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15: 550. pmid:25516281
  84. 84. Miller BR, Wei T, Fields CJ, Sheng P, Xie M. Near-infrared fluorescent northern blot. RNA N Y N. 2018;24: 1871–1877. pmid:30201850