The global redox-responsive transcriptional regulator Rex represses fermentative metabolism and is required for Listeria monocytogenes pathogenesis

The Gram-positive bacterium Listeria monocytogenes is the causative agent of the foodborne disease listeriosis, one of the deadliest bacterial infections known. In order to cause disease, L. monocytogenes must properly coordinate its metabolic and virulence programs in response to rapidly changing environments within the host. However, the mechanisms by which L. monocytogenes senses and adapts to the many stressors encountered as it transits through the gastrointestinal (GI) tract and disseminates to peripheral organs are not well understood. In this study, we investigated the role of the redox-responsive transcriptional regulator Rex in L. monocytogenes growth and pathogenesis. Rex is a conserved canonical transcriptional repressor that monitors the intracellular redox state of the cell by sensing the ratio of reduced and oxidized nicotinamide adenine dinucleotides (NADH and NAD+, respectively). Here, we demonstrated that L. monocytogenes Rex represses fermentative metabolism and is therefore required for optimal growth in the presence of oxygen. We also show that Rex represses the production of virulence factors required for survival and invasion of the GI tract, as a strain lacking rex was more resistant to acidified bile and invaded host cells better than wt. Consistent with these results, Rex was dispensable for colonizing the GI tract and disseminating to peripheral organs in an oral listeriosis model of infection. However, Rex-dependent regulation was required for colonizing the spleen and liver, and L. monocytogenes lacking the Rex repressor were nearly sterilized from the gallbladder. Taken together, these results demonstrated that Rex functions as a repressor of fermentative metabolism and suggests a role for Rex-dependent regulation in L. monocytogenes pathogenesis. Importantly, the gallbladder is the bacterial reservoir during listeriosis, and our data suggest redox sensing and Rex-dependent regulation are necessary for bacterial survival and replication in this organ. AUTHOR SUMMARY Listeriosis is a foodborne illness caused by Listeria monocytogenes and is one of the deadliest bacterial infections known, with a mortality rate of up to 30%. Following ingestion of contaminated food, L. monocytogenes disseminates from the gastrointestinal (GI) tract to peripheral organs, including the spleen, liver, and gallbladder. In this work, we investigated the role of the global redox-responsive regulator Rex in L. monocytogenes growth and pathogenesis. We demonstrated that Rex derepression coordinates expression of genes necessary in the GI tract during infection, including fermentative metabolism, bile resistance, and invasion of host cells. Accordingly, Rex was dispensable for colonizing the GI tract of mice during an oral listeriosis infection. Interestingly, Rex-dependent regulation was required for bacterial replication in the spleen, liver, and gallbladder. Taken together, our results demonstrate that Rex-mediated redox sensing and transcriptional regulation are important for L. monocytogenes metabolic adaptation and virulence.


53
To successfully colonize different niches, bacteria must be able to rapidly sense  116 Highlighted genes are predicted to be in an operon [20].

118
In silico promoter analysis of genes exhibiting Rex-dependent regulation was 119 performed to determine potential Rex binding sites using the Bacillus subtilis Rex 120 consensus sequence [14]. We identified potential Rex binding sites in the promoter 121 regions of 48 genes and/or operons repressed by Rex (S5 Table). Specifically, we 122 identified putative Rex binding sites upstream of lap, pflBC, and pflA, indicating Rex likely 123 repress fermentative metabolism directly. Rex binding sites were also predicted upstream 124 of bsh and inlAB, further suggesting the involvement of Rex in virulence gene regulation.

125
In the absence of rex, 110 transcripts were less abundant during aerobic growth, 126 indicating the presence of Rex is required to fully activate these genes (S3 Table). As 127 Rex is a canonical transcriptional repressor, we hypothesize these changes are due to 128 indirect effects. Indeed, promoter analysis did not identify any putative Rex binding sites 129 in the promoters of genes activated by Rex, suggesting these changes are likely due to 130 indirect Rex-dependent regulation.

131
132 Fermentative metabolism is repressed by Rex

133
Transcriptional analysis indicated that Rex-mediated repression functions to down-134 regulate fermentative metabolism during aerobic growth ( Fig 1A). To verify the role of Rex 135 in regulating metabolism, we first assessed growth of the wt and ∆rex strains during both 136 aerobic and anaerobic growth. A small, but significant, growth defect was observed for 137 rex-deficient L. monocytogenes, beginning 4 hours post-inoculation into aerobic shaking 138 flasks (Fig 1B). This defect was not due to a change in glucose uptake, as wt and ∆rex 139 consumed glucose at similar rates ( Fig 1C). In contrast, the ∆rex strain exhibited no 140 growth defect when incubated anaerobically (S1A Fig), demonstrating Rex-dependent 141 repression is dispensable during anaerobic growth.

142
To clarify the effect of Rex regulation on L. monocytogenes aerobic growth, 143 extracellular metabolites were quantified 4 hours post-inoculation. The ∆rex mutant 144 secreted approximately 90% more lactate and ~55% more formate than wt and the 145 complemented strain (∆rex pPL2.rex, Figs 1D and 1E). This was accompanied by a 146 concomitant decrease in the primary aerobic by-product acetate ( Fig 1F). These  172 monocytogenes strains lacking bsh or both rex and bsh, and assessed their survival 173 following a 24-hour exposure to porcine bile in BHI. L. monocytogenes colonizing the 174 gallbladder would be exposed to bile at neutral pH [9], which we found to have no effect 175 on the survival of any bacterial strain tested (Fig 2A). These results are consistent with 219 increased adherence of ∆rex to Huh7 hepatocytes did not translate to a significant 220 increase in internalization by these cells (Fig 3D). Together, these results demonstrated 221 that L. monocytogenes ∆rex invades human intestinal epithelial cells and adheres to 222 human hepatocytes better than wt as a result of increased inlAB transcription.

223
After invading host cells via receptor-mediated endocytosis or phagocytosis, L.
224 monocytogenes replicates intracellularly and spreads cell-to-cell using actin-based 225 motility [2-5]. To investigate the role of Rex regulation in these facets of pathogenesis, 226 we first measured intracellular growth in several relevant cell types. We found that the 227 ∆rex strain replicated intracellularly at the same rate as wt in activated bone marrow-228 derived macrophages, Huh7 human hepatocytes, and TIB73 murine hepatocytes (Figs 229 4A-C). These results suggested that Rex-regulated promoters are de-repressed in wt L.
230 monocytogenes during intracellular growth and therefore, deleting the Rex repressor had 231 no effect on growth.  268 approximately 2-logs on day 2 and 1-log on days 3 and 4 of the infection (Fig 5F).
269 Similarly, there was a 1-log decrease in ∆rex CFU in the liver 4 days post-infection (Fig   270 5G). The most dramatic attenuation was observed in the gallbladder, with ∆rex decreased 272 significant defect was surprising, as the ∆rex mutant survived similarly to wt when 273 exposed to bile at neutral pH, as would be encountered in the gallbladder (Fig 3). Taken 274 together, these results confirmed our hypothesis that Rex-dependent repression is

306
The results herein demonstrate that L. monocytogenes Rex functions to repress 307 fermentation during aerobic growth in order to maximize energy generation. We found the 308 ∆rex mutant over-expressed genes necessary for fermentative metabolism (lap, ldhA, and 309 pflBC/pflA) and accordingly, produced more lactate and formate than wt when replicating 310 aerobically. While acetate is the major end-product generated by wt L. monocytogenes 311 during aerobic growth [29,30], we observed a concomitant decrease in acetate production 312 by ∆rex. Together, these results suggest that in the absence of Rex repression, there is 313 an increased metabolic flux from pyruvate towards lactate and away from acetate 314 production (Fig 2A). The increased LdhA activity to produce lactate funnels NADH away 315 from the ETC, resulting in less ATP generation by respiration. Indeed, the L.