The WalRK Two-Component System Is Essential for Proper Cell Envelope Biogenesis in Clostridioides difficile

Abstract

The WalR-WalK two-component regulatory system (TCS) is found in all Firmicutes, in which it regulates the expression of multiple genes required for remodeling the cell envelope during growth and division. Unlike most TCSs, WalRK is essential for viability, so it has attracted interest as a potential antibiotic target. In this study, we used overexpression of WalR and CRISPR interference to investigate the Wal system of Clostridioides difficile, a major cause of hospital-associated diarrhea in high-income countries. We confirmed that the wal operon is essential and identified morphological defects and cell lysis as the major terminal phenotypes of altered wal expression. We also used transcriptome sequencing (RNA-seq) to identify over 150 genes whose expression changes in response to WalR levels. This gene set is enriched in cell envelope genes and includes genes encoding several predicted PG hydrolases and proteins that could regulate PG hydrolase activity. A distinct feature of the C. difficile cell envelope is the presence of an S-layer, and we found that WalR affects expression of several genes which encode S-layer proteins. An unexpected finding was that some Wal-associated phenotypic defects were inverted in comparison to what has been reported for other Firmicutes. For example, downregulation of Wal signaling caused C. difficile cells to become longer rather than shorter, as in Bacillus subtilis. Likewise, downregulation of Wal rendered C. difficile more sensitive to vancomycin, whereas reduced Wal activity is linked to increased vancomycin resistance in Staphylococcus aureus. IMPORTANCE The WalRK two-component system (TCS) is essential for coordinating synthesis and turnover of peptidoglycan in Firmicutes. We investigated the WalRK TCS in Clostridioides difficile, an important bacterial pathogen with an atypical cell envelope. We confirmed that WalRK is essential and regulates cell envelope biogenesis, although several of the phenotypic changes we observed were opposite to what has been reported for other Firmicutes. We also identified over 150 genes whose expression is controlled either directly or indirectly by WalR. Overall, our findings provide a foundation for future investigations of an important regulatory system and potential antibiotic target in C. difficile.

Keywords: cell envelope; regulation of gene expression; signal transduction.

FIG 1

The wal operon of C. difficile. Based on studies in other bacteria, WalK (red) is a bifunctional signal-transducing enzyme that can phosphorylate or dephosphorylate WalR (green). WalK is predicted to have an extracellular Cache domain, as well as intracellular HAMP, HisKA (phospho-acceptor), and HATPase domains. Unlike the B. subtilis ortholog (inset), it does not have an intracellular PAS domain. WalR~P binds DNA to activate or repress expression of Wal-regulated genes. The lipoprotein (WalA [blue]), which is unique to C. difficile, may modulate WalK activity and is predicted to have a beta-propeller lyase domain. The truA2 gene codes for a pseudouridylate synthase and is not thought to play a role in Wal signaling. The operon locus is cdr20291_1676-1679 in R20291 and cd630_17810-17840 in 630Δerm.

FIG 2

The wal operon is required for viability and normal rod morphology of C. difficile 630Δerm. (A) Viability assay. Serial dilutions of overnight cultures were spotted onto TY plates with or without 1% xylose to induce the expression of dCas9. Plates were photographed after incubation overnight. Strains carry chromosomal copies of the CRISPRi components and the following guides: sgRNA-neg (UM554), sgRNA-walA1 (UM555), and sgRNA-walA2 (UM556). (B) Growth curves in liquid TY containing 0, 0.01, 0.1, or 1% xylose as indicated. For clarity, the sgRNA-neg control strain is graphed only at 0% xylose. Samples were taken at 4 h (arrow) for phase-contrast microscopy (C), cell length measurements (D), and determination of curvature (E). Numbers at the bottom of micrographs in panel C refer to percent xylose. Note cell debris indicative of lysis at 1% xylose. Bar = 10 μm. Cell length and percent curved cells were based on about 500 cells per condition. The sgRNA-walA1 strain is longer than the sgRNA-neg control in all pairwise comparisons as determined by a t test (P < 0.0001).

FIG 3

Overexpression of walR from a multicopy plasmid impairs growth, alters morphology, and slows autolysis in 630Δerm. (A) Viability assay. Overnight cultures of 630Δerm harboring pCE691 (P_xyl::walR) or the empty vector control plasmid pBZ101 (P_xyl EV) were serially diluted and spotted onto TY-Thi plates with or without 1% xylose. Plates were photographed after incubation overnight. (B) Growth curve of 630Δerm/pCE691 (P_xyl::walR). Duplicate cultures were grown to an OD₆₀₀ of 0.2, at which time one was induced with 3% xylose (green arrow). Both cultures were harvested after 3 h (red arrow) for analysis by phase-contrast microscopy (C) and a lysis assay (D). Bar in panel C = 10 μm. For the lysis assay, cells were suspended in buffer containing 0.01% Triton X-100 and OD₆₀₀ was monitored over time. Error bars depict SD of 3 technical replicates. Data shown are representative of results from at least 3 independent experiments.

FIG 4

Prolonged overexpression of walR from a single-copy chromosomal P_xyl::walR allele affects cell shape and slows lysis. (A) Growth curves. An overnight culture of strain UM626 (P_xyl::walR integrated at pyrE in 630Δerm) was subcultured 1:50 into TY with varied xylose concentrations as indicated, grown for 6 h, and then analyzed in a lysis assay (B) or by phase-contrast microscopy (C). For the lysis assay, cells were suspended in buffer containing 0.01% Triton X-100 and OD₆₀₀ was monitored over time. Data are graphed as the mean and SD of 3 technical replicates, but most error bars are smaller than the symbols. Cells from cultures grown in the presence of 0 to 1% xylose appeared normal and are not shown, but in cultures induced with 3% xylose, about 10% of the cells had irregular or bent morphologies, examples of which are shown in panel C. Bar = 10 μm. Data shown are representative of those from at least 3 independent experiments.

FIG 5

Comparison of transcript changes between Wal-ON and Wal-OFF conditions. (A) Heat map showing the 79 genes whose transcript abundance changed ≥4-fold when walR was overexpressed (Wal-ON) in comparison to when the wal operon was silenced with CRISPRi (Wal-OFF). (B) Heat map showing the 49 genes whose transcript abundance changed ≥ 4-fold when the wal operon was silenced (Wal-OFF) in comparison to when walR was overexpressed (Wal-ON). Green, induced; red, repressed.

FIG 6

Confirmation of RNA-seq results with RFP reporter fusions to selected genes. Cells harboring plasmids with transcriptional fusions of the red fluorescent protein mCherryOpt to the indicated promoters were grown as per RNA-seq conditions, either Wal-ON (green) or Wal-OFF (red). Red fluorescence was measured by flow cytometry and is graphed as the mean log₂ fold change from two independent experiments with two technical replicates. Error bars represent the SD of all four measurements. For comparison, the mean log₂ fold change of the same genes as determined by RNA-seq is also shown. The dltD reporter was graphed only under Wal-OFF conditions because this fusion was induced by xylose even in the absence of P_xyl::walR. Genes with asterisks represent fusions that were constructed to the corresponding promoter regions from strain R20291.

FIG 7

Perturbation of individual Wal-ON regulon genes does not replicate the Wal-ON phenotype. Overnight cultures of 630Δerm harboring overexpression plasmids were subcultured into TY-Thi to an OD₆₀₀ of 0.03, grown to an OD₆₀₀ of 0.2, and induced with 3% xylose for 3 h. Overnight cultures of 630Δerm harboring CRISPRi plasmids were subcultured into TY-Thi containing 1% xylose to an OD₆₀₀ of 0.03 and grown to an OD₆₀₀ of ~0.8. Each culture was then examined by microscopy (bar = 5 μm) (A) and tested in the lysis assay (B). Induction of walR is the only condition that achieved phase-bright cells and slowed lysis. The plasmids used were pBZ101 (EV), pCE691 (P_xyl::walR), pIA112 (P_xyl::cd630_07380), pIA113 (P_xyl::cd630_07390), pIA114 (P_xyl::cd630_07380-07390), pIA115 (P_xyl::cd630_07390-07380), pIA34 (sgRNA-neg), pCE744 (sgRNA-cd630_27680-1), and pCE745 (sgRNA-cd630_27680-2).

FIG 8

Perturbation of individual Wal-OFF regulon genes does not replicate the Wal-OFF phenotype. (A) Phase-contrast microscopy of strains harboring overexpression plasmids or CRISPRi plasmids after induction with xylose as described in the legend to Fig. 7. Bar = 5 μm. (B) Viability assay. Overnight cultures were serially diluted and spotted onto TY plates with or without 1% xylose. Plates were photographed after incubation overnight. CRISPRi knockdown of the wal operon was the only condition that resulted in curved cells, lysis, or a viability defect. The plasmids used were pBZ101 (EV), pCE791 (P_xyl::cwpV), pIA34 (sgRNA-neg), pIA50 (sgRNA-walA1), and pCE738 (sgRNA-dltD).