Systems Level Analyses Reveal Multiple Regulatory Activities of CodY Controlling Metabolism, Motility and Virulence in Listeria monocytogenes

Abstract

Bacteria sense and respond to many environmental cues, rewiring their regulatory network to facilitate adaptation to new conditions/niches. Global transcription factors that co-regulate multiple pathways simultaneously are essential to this regulatory rewiring. CodY is one such global regulator, controlling expression of both metabolic and virulence genes in Gram-positive bacteria. Branch chained amino acids (BCAAs) serve as a ligand for CodY and modulate its activity. Classically, CodY was considered to function primarily as a repressor under rich growth conditions. However, our previous studies of the bacterial pathogen Listeria monocytogenes revealed that CodY is active also when the bacteria are starved for BCAAs. Under these conditions, CodY loses the ability to repress genes (e.g., metabolic genes) and functions as a direct activator of the master virulence regulator gene, prfA. This observation raised the possibility that CodY possesses multiple functions that allow it to coordinate gene expression across a wide spectrum of metabolic growth conditions, and thus better adapt bacteria to the mammalian niche. To gain a deeper understanding of CodY's regulatory repertoire and identify direct target genes, we performed a genome wide analysis of the CodY regulon and DNA binding under both rich and minimal growth conditions, using RNA-Seq and ChIP-Seq techniques. We demonstrate here that CodY is indeed active (i.e., binds DNA) under both conditions, serving as a repressor and activator of different genes. Further, we identified new genes and pathways that are directly regulated by CodY (e.g., sigB, arg, his, actA, glpF, gadG, gdhA, poxB, glnR and fla genes), integrating metabolism, stress responses, motility and virulence in L. monocytogenes. This study establishes CodY as a multifaceted factor regulating L. monocytogenes physiology in a highly versatile manner.

Fig 1. RNA-Seq analysis of WT and ΔcodY L. monocytogenes strains grown in BHI and LBMM media.

A. Pie charts representing differentially transcribed genes in WT and ΔcodY L. monocytogenes bacteria grown in BHI or LBMM. B. Venn diagrams representing the different CodY regulated gene clusters in both BHI and LBMM media. C. Hierarchical clustering of differentially transcribed genes in WT and ΔcodY L. monocytogenes bacteria in both BHI and LBMM media.

Fig 2. Functional enrichment analysis of CodY regulated genes.

Categories functionally enriched in each cluster of CodY regulated genes, analyzed using the MIPS server [44]. All categories are significantly enriched (P-value <0.005).

Fig 3. ChIP-seq analysis of CodY binding sites under growth in BHI and LBMM media.

A. Venn diagram representing CodY binding sites associated with differentially expressed genes in BHI and LBMM. B. Genomic view of CodY binding sites in L. monocytogenes 10403S– the two outer circles represent the + and—strands of L. monocytogenes genome, each line is a gene. Orange lines represent CodY regulated genes, while grey lines represent genes unaffected by CodY (based on the RNA-Seq analysis). The third circle represents CodY binding sites associated with differentially expressed genes in BHI medium, while the forth circle represents CodY binding sites associated with differentially expressed genes in LBMM. The fifth circle represents CodY motifs associated with differentially expressed genes, identified by the MAST algorithm [46] using a CodY binding consensus sequence as an input based on previous experimental CodY-box analyses [36, 37]. The consensus sequence was generated using the MEME algorithm [47]. Only sites with E value < 10 and P value < 0.05 are shown.

Fig 4. Validation of CodY-dependent differential transcription and CodY binding to representative genes from each cluster using RT-PCR and ChIP RT-PCR.

A. RT-PCR transcription analysis of representative genes of each cluster presented by fold change values of mRNA relative quantity (RQ) in WT vs. ΔcodY bacteria grown in BHI, HBMM and LBMM media. Results are average of at least 3 independent biological repeats. Transcription levels were normalized to the transcription of rpoD. Error bars represent standard error (SE). B. ChIP RT-PCR analysis of CodY binding to representative genes from each cluster (corresponding panel A) in bacteria grown in BHI, HBMM and LBMM media. Fold enrichment of CodY association with each one of the tested sequences was first normalized to two control sequences: bglA and rpoD (normalization to multiple control genes was done by StepOne software of Applied biosystems) and then to its no-ChIP control. Results are average of at least 3 independent biological repeats. Error bars represent standard error (SE). Asterisk (*) represents p-value < 0.05.

Fig 5. EMSA analysis of CodY binding to select genes from the different clusters.

Binding of CodY to 11 regulatory regions of CodY regulated transcriptional units (hisZ, rsbV, glpF, argG, gadC, feoA, actA, gdhA, poxB, glnR and fliN) and a control probe (bglA). Primers used for amplification of DNA probes are described in S5 Table. Results are representative of at least 2 independent biological repeats. EMSA was performed with 10 mM BCAA for hisZ, rsbV, glpF, argG, gadC gdhA, poxB, glnR and fliN. For feoA and actA BCAAs were not added. Asterisks (*) point to unbound DNA probes.

Fig 6. ΔcodY mutant is impaired in motility and cell adhesion.

Swarming of L. monocytogenes WT and ΔcodY bacteria, as well a codY complemented strain on BHI A. and LBMM B. soft agar plates. ΔflaA strain was used as a negative control. Results are representative of at least 3 biological repeats. Bar charts represent the diameters of bacterial growth zones, based on 3 independent biological repeats. C. An attachment assay of WT, ΔcodY and codY complemented L. monocytogenes strains to Caco2 epithelial cells. ΔflaA is used as a negative control. The results represent at least 3 biological repeats Error bars represent standard error of the mean (SE). Asterisk (*) represents p-value < 0.05. Student’s t-test was used for statistical analysis.

Fig 7. CodY serves as both a repressor and as an activator of genes under rich and minimal growth conditions.

Unlike previous studies describing a function for CodY only under rich conditions, here we show that CodY exhibit a global regulatory role as a repressor and as an activator of genes under both rich and minimal conditions, the latter limiting for BCAAs. The coordinated regulation of metabolic, stress and virulence genes by CodY, most likely allows L. monocytogenes bacteria to swiftly switch from being a saprophyte to virulent bacteria. The color-coded oval represents CodY protein under the different conditions bound to regulatory regions of genes. R represents repressor and A activator.

Fig 8. A Model for CodY regulation of L. monocytogenes central metabolism.

Illustration of central metabolic pathways that are regulated by CodY in L. monocytogenes in BHI A. and LBMM B. In light blue—pyruvate metabolism, light green—nitrogen metabolism, light orange—arginine metabolism, in the middle the TCA cycle. Green arrows represent enzymes that are induced by CodY, while red arrows represent enzymes repressed by CodY, based on the RNA-Seq analysis.