CdsH Contributes to the Replication of Salmonella Typhimurium inside Epithelial Cells in a Cysteine-Supplemented Medium

Salmonella Typhimurium is a facultative, intracellular pathogen whose products range from self-limited gastroenteritis to systemic diseases. Food ingestion increases biomolecules’ concentration in the intestinal lumen, including amino acids such as cysteine, which is toxic in a concentration-dependent manner. When cysteine’s intracellular concentration reaches toxic levels, S. Typhimurium expresses a cysteine-inducible enzyme (CdsH), which converts cysteine into pyruvate, sulfide, and ammonia. Despite this evidence, the biological context of cdsH’s role is not completely clear, especially in the infective cycle. Since inside epithelial cells both cdsH and its positive regulator, ybaO, are overexpressed, we hypothesized a possible role of cdsH in the intestinal phase of the infection. To test this hypothesis, we used an in vitro model of HT-29 cell infection, adding extra cysteine to the culture medium during the infective process. We observed that, at 6 h post-invasion, the wild type S. Typhimurium proliferated 30% more than the ΔcdsH strain in the presence of extra cysteine. This result shows that cdsH contributes to the bacterial replication in the intracellular environment in increased concentrations of extracellular cysteine, strongly suggesting that cdsH participates by increasing the bacterial fitness in the intestinal phase of the S. Typhimurium infection.


Introduction
Salmonellosis corresponds to one of the most common food diseases, affecting an estimated ten million people annually, with more than 100.000 deaths worldwide. The disease symptoms are noticeable between 6 and 72 h after consuming Salmonella, with a duration of 2-7 days [1].
Salmonella enterica serovar Typhimurium (S. Typhimurium) is one of the most relevant etiologic agents of salmonellosis. The infection begins with the consumption of contaminated food or water, where bacteria reach the distal ileum and induce their internalization into epithelial cells and M cells, among others, in the so-called intestinal phase of the infection [2]. S. Typhimurium produces self-limited gastroenteritis in healthy humans at this phase, whereas, in immunosuppressed individuals, infection [24]. Due to the induction of cdsH expression inside epithelial cells and considering that this gene is overexpressed under high-cysteine concentrations, we wondered whether cdsH contributes to the S. Typhimurium replication inside epithelial cell lines in the presence of exogenous cysteine in vitro. We found that, at 6 h post-invasion, the wild type S. Typhimurium strain proliferates 30% more when extra cysteine is added to the medium, compared with the ∆cdsH strain in HT-29 cells. The contribution of cdsH to the replication of S. Typhimurium inside epithelial cells in response to cysteine is discussed.
Cysteine minimum inhibitory concentration (MIC). An overnight culture of each strain was used to generate a new culture with approximately 1 × 10 4 CFU/mL (the equivalent of a 0.5 McFarland standard). Bacterial cultures were disposed in 96-well microtiter plates and treated with decreasing concentrations of freshly prepared cysteine solutions. Microtiter plates were incubated in a Thermo shaker incubator (HILAB ® MSC-100, Curitiba, Brazil) at 37 • C with shaking for 16 h, and OD 600 was measured in a microtiter plate reader. The MIC was defined as the lowest concentration that completely inhibited visible growth compared with the cysteine-free control. In other words, a lower MIC is interpreted as an increased susceptibility.
Cysteine toxicity assay (Kirby-Bauer Assay). To qualitatively assess the toxic effect of cysteine under aerobic conditions, both LB and M9-glucose plates were used. Overnight cultures were used to generate a 3 mL bacterial suspension in M9-glucose with agar-agar (0.7%) with approximately 10 6 CFU. This bacterial suspension was added over an M9-glucose plate (agar-agar 1.5%) to generate a bacterial suspension layer. A filter-paper disc was then placed in the center of each bacterial layer before being loaded with 15 µL of a fresh solution of cysteine (0.5 M). The inhibition zone diameter was measured after 16 h incubation at 37 • C. Growth curves. Growth curves were performed in LB and M9-glucose media with or without cysteine supplementation (0.2 mM). Bacteria were grown aerobically until the stationary phase, and optical density at 600 nM (OD 600 ) was measured in a microtiter plate reader (Guangzhou Jet Bio-Filtration Co ® , Guangzhou, China) every hour.
RNA isolation and mRNA detection. Total RNA extraction and RT-qPCR analyses were performed as previously described by Álvarez et al. (2015) [14]. Cultures of S. Typhimurium LT2 wild type and ∆ybaO mutant strain were grown in M9-glucose broth and incubated with shaking at 37 • C until OD 600 =~0.4. When necessary, bacterial cultures were exposed to 0.2 mM cysteine for 20 min. RNA was obtained by the acid-phenol method. Briefly, RNA was precipitated with isopropanol for 10 min at room temperature, washed with ice-cold 70% v/v ethanol and resuspended in DEPC-treated water before treating with 2U DNase I (Promega) to remove any DNA trace. The purity of the extracted RNA was determined by spectrometry (Absorbance 260/280). cDNA synthesis was carried out at 37 • C for 1 h in 25 µL of a mixture that contained 2.5 pmol of the specific reverse primers, 1 µg of template RNA, 0.2 mM dNTPs, 1 mL of nuclease-free water, 4 µL of M-MLV Reverse Transcriptase 5× Reaction Buffer (250 mM Tris-HCl pH 8.3, 375 mM KCl, 15 mM MgCl2, 10 mM 1,4-dithiothreitol (DTT), 40 U of RNasin) and 100 U of M-MLV reverse transcriptase (Promega ® , Madison, WI, USA). The relative quantification of transcript levels was performed by RT-qPCR using the Brilliant II SYBR Green QPCR master reagent kit and the Mx3000P detection system (Stratagene ® , San Diego, CA, USA). The 16s rRNA levels were used for normalization. The RT-qPCR mixture (20 µL) contained 50 ng of cDNA template, 120 nM each primer (cdsHRTF-cdsHRTR for cdsH, ybaORTF-ybaORTR for ybaO, and 16SF-16SR for 16s gene, see Table 1) and 1 µL of ROX reference dye (1:200). RT-qPCR was performed under the following conditions: 10 min at 95 • C followed by 40 cycles of 15 s at 95 • C, 15 s at 60 • C, and 25 s at 72 • C, followed by a melting cycle from 65 to 95 • C to check for amplification specificity. A previous standard quantification curve with serial dilutions of RT-qPCR products was constructed for each gene to calculate the amplification efficiency. These values were used to obtain the ratio between the gene of interest and the 16s rRNA gene expression. The experiment was performed for three biological and technical replicates. These data were analyzed with the MxPRO software (Agilent Technologies ® , Santa Clara, CA, USA).
Western blot. Total protein extraction was performed as previously described by Nevermann et al. (2019) [27]. Briefly, cultures of S. Typhimurium LT2 wild type and ∆ybaO mutant strain were grown in M9-glucose broth and incubated with shaking at 37 • C until OD 600 =~0.4. When necessary, bacterial cultures were exposed to cysteine 200 µM for 1 h before centrifugation, and the pellet was resuspended in Tris-HCl 100 mM buffer. Total proteins were extracted by sonication on ice (30 s pulses, amplitude 70%) with sonicator Vibra-Cell. Protein quantification was performed with the classical Bradford method [28].
Western blot was performed as previously described by Jofre et al. (2014) [29]. Briefly, 100 µg of protein extracts of each sample were loaded by SDS-PAGE (12% acrylamide). Proteins were transferred to a nitrocellulose membrane (Bio-Rad) and blocked overnight at 4 • C with milk in TBS (1.4% NaCl, 0.6% glycine pH 7.4). The membrane was washed three times with TBS-Tween (NaCl 1.4%, glycine 0.6%, Tween 0.1%). The western blots were probed with a 1:5000 dilution of primary antibody anti-FLAG for 12 h at 4 • C. The membrane was rewashed with TBS-Tween and then probed with a mix of 1:10,000 dilution of secondary antibody anti-mouse-horseradish peroxidase (HRP) conjugate (Promega) and 1:25,000 dilution anti-mouse-HSP for 2 h with shaking at room temperature. Finally, the membrane was washed with TBS-Tween prior to being revealed by the BCIP/NTB (Invitrogen ® ) kit according to the manufacturer's instructions. Hsp60 was used as a control load, as described [30]. The Western blot was repeated at least three independent times and analyzed by densitometry to quantify the relative amounts of protein by ImageJ Software.
Deoxycholate survival assay. A 25 mL OD 600~0 .5 culture of each strain growing on LB broth was washed three times with phosphate saline buffer (PBS) and finally resuspended in 5 mL of PBS. Then, 20 µL of each bacterial suspension was resuspended in a 96-well microplate with 180 µL of deoxycholate (Sigma-Aldrich ® , Merck ® KGaA, Darmstadt, Germany) 0.5% or PBS (control) before incubating them for 2 h. Finally, each well was serially diluted in PBS and seeded in LB plates for CFU counting.
Bacterial viability in the HT-29 cell line with extra supplementation of cysteine. Cell viability was determined by neutral red uptake assay, as previously described by Repetto et al. (2008) [31]. Briefly, HT-29 cell cultures (DMEM) were washed with PBS and refilled with DMEM alone or supplemented with cysteine. This procedure was repeated at 1 h and 3 h of incubation. After 24 h of incubation, cells were washed with PBS, and a neutral red solution diluted in DMEM (Sigma-Aldrich ® , Merck ® KGaA, Darmstadt, Germany) (60 µg/mL) was added before incubation at 37 • C for 2 h. Next, each well was washed twice with PBS, and a destaining solution (50% ethanol (Merck ® KGaA, Darmstadt, Germany) 49% deionized water, 1% glacial acetic acid (Merck ® KGaA, Darmstadt, Germany)) was used to remove the neutral red solution from cells. Finally, OD 540 was measured for each well. A well was treated with deoxycholate 0.5% for 5 min before neutral red was added as a control.
Epithelial cell line replication assay. S. Typhimurium LT2 invasion and replication were determined using the gentamycin-protection assay described by Fuentes et al. (2008) with some modifications [32]. Briefly, the S. Typhimurium strains used to infect HT-29 monolayer were grown in microaerophilic conditions to an OD 600 = 0.2 at 37 • C. Then, bacteria were washed three times with PBS and finally resuspended in DMEM with or without cysteine (0.2 mM). This suspension was immediately used for HT-29 monolayer infection in 96-well plates using a multiplicity of infection of 100:1. One hour after infection, cells were washed three times with PBS and a DMEM solution supplemented with gentamycin (200 µg/mL), with or without cysteine, was added to each well. The previous procedure was repeated three hours after infection but without gentamycin, and infected cells were incubated for 3, 6 or 24 h post-infection (p.i.). To assess bacterial invasion and replication, the CFU were determined in each time and condition. Bacteria were recovered from HT-29 cells at 3 (invasion), 6 and 24 h (replication) p.i. using a deoxycholate solution (0.5% in PBS) and mechanical disruption. The bacterial suspension was seeded in LB plates, and CFU were counted after 16 h of incubation at 37 • C.
Statistical analysis. All analyses were performed using IBM SPSS and GraphPad Prism 8 software. ANOVA with Tuckey's test as post hoc and Mann-Whitney tests were performed in the cysteine Kirby-Bauer and mRNA expression experiments, respectively. Two-way ANOVA with post hoc Tukey's multiple comparisons test was performed for the replication experiment analysis.

The cdsH and ybaO Genes Participate in the Response of S. Typhimurium against High Concentrations of Cysteine in Minimal Medium
Since CdsH responds to an excess of cysteine and YbaO controls the expression of CdsH under anaerobic conditions [11], we wondered whether S. Typhimurium ∆cdsH exhibited increased susceptibility to exogenous cysteine under aerobic growth. For this aim, a Kirby-Bauer test was carried out in M9-glucose minimal medium, a culture medium that does include cysteine. A cysteine solution of 0.5 M was added to a filter-paper disc placed on the M9-glucose minimal medium plates, previously seeded with the tested bacteria, before incubation for 16 h at 37 • C to obtain growth inhibition haloes. Growth inhibition haloes of approximately 40 mM diameter were observed for the mutant strains, while the wild type exhibited no growth inhibition. The mutant phenotype was fully complemented by the corresponding plasmid ( Figure 1). This result argues for the role of cdsH in the resistance to high cysteine concentrations. In order to set the sublethal concentration of cysteine for the following experiments, the MIC of cysteine was determined for the S. Typhimurium wild type strain and the mutant derivatives (Table  3). Compared to the wild type strain, mutant strains (and mutants harboring the empty vector) showed a 16-fold lower MIC. On the other hand, the respective complemented mutants showed MICs even higher than those presented by the WT, plausibly due to the increased gene dosage provided by the respective cloning vector (Table 3). All these results support the participation of both cdsH and ybaO in cysteine susceptibility under the tested conditions. Based on these results, we set the work concentration of cysteine to 0.2 mM, half of the MIC regarding S. Typhimurium ΔcsdH and ΔybaO (Table 3). This concentration is similar to the higher concentration inferred for cysteine in the ileum (as extrapolated from the methionine concentration, see above) [9]. As a control, and to determine any additional detrimental effects of cysteine, we analyzed the growth curve of S. Typhimurium wild In order to set the sublethal concentration of cysteine for the following experiments, the MIC of cysteine was determined for the S. Typhimurium wild type strain and the mutant derivatives (Table 3). Compared to the wild type strain, mutant strains (and mutants harboring the empty vector) showed a 16-fold lower MIC. On the other hand, the respective complemented mutants showed MICs even higher than those presented by the WT, plausibly due to the increased gene dosage provided by the respective cloning vector (Table 3). All these results support the participation of both cdsH and ybaO in cysteine susceptibility under the tested conditions. Based on these results, we set the work concentration of cysteine to 0.2 mM, half of the MIC regarding S. Typhimurium ∆csdH and ∆ybaO (Table 3). This concentration is similar to the higher concentration inferred for cysteine in the ileum (as extrapolated from the methionine concentration, see above) [9]. As a control, and to determine any additional detrimental effects of cysteine, we analyzed the growth curve of S. Typhimurium wild type and mutant derivatives, either in M9-glucose supplemented with 0.2 mM cysteine or in LB (rich medium) supplemented with 0.2 mM cysteine (37 • C with shaking). As shown in Figure S1, we found no differences in the growth curves of the wild types and mutants in LB, despite the presence of cysteine. In the case of M9, we observed a slightly extended lag phase, although no differences in the growth rate (µ) were observed. No differences were observed when bacteria were cultured in the absence of supplemented cysteine ( Figure S2). These results show that the addition of 0.2 mM of exogenous cysteine exerts no detrimental effects in the S. Typhimurium ∆cdsH or ∆ybaO in a rich medium (LB). By contrast, in a poorer environment (i.e., minimal medium M9), the lack of either cdsH or ybaO seems to slightly delay the adaptation of strains, although no significant effects in the growth rate were observed under the tested conditions (Table S1).

The Expression of the cdsH Gene is Induced in the Presence of Cysteine
Since our results show that cdsH and ybaO participate in the adaptation to the presence of exogenous cysteine, we explored whether this amino acid induces the expression of these genes. For this purpose, RT-qPCR and Western blot assays were carried out from cultures in M9-glucose medium alone (control) or M9-glucose supplemented with 0.2 mM cysteine. As shown in Figure 2, the addition of cysteine induced the expression of cdsH only when the ybaO gene was present (Figure 2), corroborating the previous work of Oguri et al. (2012) [11]. By contrast, in a ∆ybaO context, this induction was not observed ( Figure 2C). Regarding the expression of ybaO, we found that, under the tested condition, the addition of cysteine produced no effect compared with the M9-glucose alone ( Figure 2B), which can be explained by the negative self-regulation of the LRP regulator family [33].
The S. Typhimurium ybaO-3×FLAG strain was also used to assess the ybaO expression at the protein level. Nevertheless, the Western blot assay could not detect this protein, even in the presence of cysteine (data not shown), strongly suggesting that the amount of YbaO is too low to be detected by the performed procedure. All these results agree with that previously described by Oguri et al. (2012) [11], who assessed the cdsH expression with a nutrient-deficient medium but under anaerobic conditions. Altogether, these results show that cdsH increased its expression in the presence of cysteine in a YbaO-dependent manner, even under aerobic conditions. Since ybaO showed no increase in its transcription or translation in the presence of cysteine, unlike what was observed for cdsH, we decided to continue assessing only the participation of cdsH. medium alone (control) or M9-glucose supplemented with 0.2 mM cysteine. As shown in Figure 2, the addition of cysteine induced the expression of cdsH only when the ybaO gene was present ( Figure  2), corroborating the previous work of Oguri et al. (2012) [11]. By contrast, in a ΔybaO context, this induction was not observed ( Figure 2C). Regarding the expression of ybaO, we found that, under the tested condition, the addition of cysteine produced no effect compared with the M9-glucose alone ( Figure 2B), which can be explained by the negative self-regulation of the LRP regulator family [33].  For the Western blot assay (D), the S. Typhimurium cdsH-3×FLAG was cultured in M9-glucose to OD 600 0.4-0.5 prior to supplementing with cysteine (0.2 mM, final concentration) or water (control). Hsp60 was used as a control load, as described [30]. After 1 min incubation, bacteria were harvested, and the Westen blot assay was performed as described in Materials and Methods (a representative result of n = 3).

Cysteine Promotes Replication of S. Typhimurium inside Epithelial Cells in a cdsH-Dependent Manner
According to the transcriptomic analysis carried out by Hautefort et al. (2008), cdsH is overexpressed when Salmonella is inside epithelial cells. Since this gene is induced in response to the presence of harmful cysteine concentrations, we wondered whether the increased cdsH expression is related to a bacterial defense response inside epithelial cells. To this end, both the invasion and the replication of S. Typhimurium strains were assessed in the HT-29 cell line, comparing the standard DMEM medium with DMEM supplemented with extra cysteine (0.2 mM).
Since HT-29 cells are typically cultured in standard DMEM medium, we first assessed whether the supplementation with additional cysteine (0.2 mM) affects cell viability. To that aim, we performed a neutral red uptake [31]. As shown in Figure S3A, extra cysteine in the DMEM medium produced no effects on the HT-29 s viability. Similarly, we found that the DMEM supplemented with additional cysteine exerted no detrimental effects on the growth of S. Typhimurium and derivatives, even after an incubation of 3, 6, 12, or 24 h ( Figure S3B), in agreement with what was found with LB supplemented with cysteine ( Figure S1). Having ruled out possible artifactual results, we tested the invasion and replication of S. Typhimurium and mutant derivatives in either DMEM or DMEM supplemented with cysteine by the gentamicin protection assays with HT-29. As shown in Figure 3A, we found indistinguishable invasions between the S. Typhimurium wild type and ∆cdsH, with or without additional cysteine. By contrast, the addition of extra cysteine to the DMEM medium slightly decreased the replication of ∆cdsH bacteria, while it increased the wild type strain replication up to 30% at 6 h, but not at 24 h, where the number of intracellular bacteria similarly increased for all the strains, even with the addition of extra cysteine cultures ( Figure 3B). As shown in Table 4, the generational time of intracellular bacteria was longer in the ∆cdsH strain only when extra cysteine was added, supporting our results.  All these results together argue for the participation of cdsH in S. Typhimurium replication within epithelial cells in high cysteine concentrations. Thus, cdsH contributes to improving the bacterial fitness inside epithelial cells.

Discussion
Salmonella is an enteric pathogen that produces a wide variety of diseases, from self-limited gastroenteritis to systemic infections and death. During the infective process, before the internalization into epithelial cells, Salmonella overcomes different stressful conditions such as acid pH, bile salts, high osmolarity, and low oxygen tension [2]. Inside epithelial cells, Salmonella responds to the intracellular environment to achieve a successful replication, which is an essential process for the bacteria's virulence and develops a successful infection [34]. Since intestinal epithelial cells are continually absorbing nutrients from the lumen, the intracellular amino acid concentration increases [9]. Under these conditions, intracellular bacteria could use these high amino acid concentrations for their own purpose. In this context, S. Typhimurium has a cysteine desulfhydrase enzyme, CdsH, that responds to high cysteine concentrations to avoid the toxicity of this amino acid. As cdsH increases its expression inside epithelial cells [24], S. Typhimurium may use this enzyme to prevent the toxic effect, or use cysteine to increase its fitness.

Cysteine Susceptibility Occurs only in Minimal Medium
It has been shown that cysteine affects the growth of multiple bacteria in minimal media, and S. Typhimurium is not an exception. S. Typhimurium is also susceptible to the toxic effect of excess cysteine in minimal media, although this is not the case in a rich medium ( Figures S1 and S3). The cysteine susceptibility increases when the CdsH enzyme cannot be synthesized in the ∆cdsH mutant, or as a result of the absence of its positive regulator YbaO in the ∆ybaO context ( Figure 2 and Table 3). When S. Typhimurium grows in minimal media, the expression of the genes involved in the transport of molecules is increased [35]. Accordingly, under sulfur-limited conditions, bacteria respond by synthesizing CysB-regulated sulfide transporters [36]. Thus, cysteine can enter bacteria in multiple ways, including derivatives such as cystine. For instance, in E. coli, extracellular cystine increases the intracellular cysteine concentration, exerting a toxic effect [12,17]. Therefore, the addition of both exogenous cysteine and cystine to the culture medium increases the concentration of intracellular cysteine when Salmonella is in a nutrient-deficient environment, and would cause the toxic effect produced by this amino acid. This could be explained by the differences in the global expression patterns when bacteria grow in nutrient-rich media versus nutrient-deficient media. For example, in E. coli, the number of transport genes overexpressed in minimal media is higher than the number of transport genes overexpressed in a rich medium when the expression is compared between both media [35]. Considering that cysteine enters the cell by specific transporters, the overexpression of these transporters could explain why, in minimal media, cysteine is toxic in lower concentration than in rich media.
Under aerobic conditions, we observed similar results to those previously reported in anaerobic conditions by Oguri et al. (2012) [11], i.e., S. Typhimurium is susceptible to excess cysteine only in minimal medium ( Figure S2), and cdsH is overexpressed in response to the addition of exogenous cysteine to the medium in a YbaO-dependent manner, this being an Lrp family regulator that does not change its expression in response to cysteine.

CdsH
Contributes to the Replication of S. Typhimurium in the HT-29 Line, against the Addition of Exogenous Cysteine As mentioned before, during the invasion of the intestine, Salmonella is exposed to different conditions [37]. These conditions produce changes in bacterial genetic profile, promoting the invasion and subsequent replication inside epithelial cells [38,39]. In both processes, the effector and regulatory genes from Salmonella pathogenicity island I and II are important, since mutation on these genes produces non-invasive phenotypes [40]. In the replication assay, at 3 h post-infection, no differences between wild type and ∆cdsH strains were observed in both DMEM and DMEM + cysteine media ( Figure 3A). Since the replication of Salmonella inside epithelial cells begins between 3 and 4 h after the invasion [41,42], this process was not affected by cdsH deletion in both conditions; therefore, cdsH does not contribute to the invasion of epithelial cells. This result is in accordance with the function of CdsH, which participates in the elimination of cysteine, and its expression is regulated in that context, unlike the expression of genes such as invA and sipA, among others, whose expression is directly related to the Salmonella-invasion of epithelial cells. Therefore, according to our results, CdsH does not participate in the invasion process.
Even though replication within epithelial cells mainly depends on the genes present in SPI-1 and SPI-2 [43,44], genes involved in other processes can also contribute to bacterial replication within this cell type. An example of this was reported in S. Typhi, where the mutation in genes related to anaerobic respiration causes lower bacterial replication inside epithelial cells, possibly because these genes would allow bacteria to perform more energy-efficient metabolic processes [45]. Other genes could also be indirectly related to the replication process, such as trxA and trxB, components of the thioredoxin pathway (a protein that acts as an antioxidant), where a mutation in these genes slightly decreases replication inside of epithelial cells [46].
In other cell types, non-classical factors that contribute to Salmonella replication have also been seen. The Tsx porin participates in nucleoside uptake, where a mutation in this gene has been shown to decrease Salmonella's replicative capacity inside a line of macrophages [47]. Likewise, the S. Typhimurium ∆cdsH strain's replication does not respond to the addition of cysteine to the medium as the wild type strain at 6 h ( Figure 3B). Since this gene is overexpressed in the presence of exogenous cysteine [11,48,49] and cystine [50], overexpression within epithelial cells [24] can be attributed to the entry of cysteine or cystine into bacteria. Although culture media do not have cysteine, they do have cystine. In this sense, the increased cdsH expression may be due to cystine uptake by bacteria. Related to this, YdjN, one of the primary transporters for cystine, was also found to have an increased expression within epithelial cells [24]. When comparing the replication of each strain in the supplemented versus the non-supplemented medium at different times, we found that the wild type strain exhibits an advantage in replication at 6 h p.i. when cysteine is added to the medium, increasing the replication. An explanation could be the pyruvate synthesis from cysteine by CdsH [11], which contributes to bacteria's energy production. In addition, it has been shown that Salmonella can use CdsH to synthesize thiamine, in a different way from the conventional, which is transformed into thiamine pyrophosphate, a coenzyme required for enzymes of the central metabolism [51]. In fact, in Listeria monocytogenes, it has been shown that replication inside epithelial cells requires the uptake and biosynthesis of thiamine precursors [52]. Thus, this could represent an advantage of the wild type over the ∆cdsH strain. On the other hand, at 24 h p.i., the mutant strain was able to use cysteine, reaching a replication ratio comparable to that of the wild strain, thus suggesting the existence of another redundant system that can fulfill this function. In this sense, it is known that other enzymes such as CysK or CysM also have cysteine desulfhydrase activity, although these are not cysteine-inducible [11].
Cysteine can exert its toxic activity mainly by producing oxidative stress through the reduction of Fe(III) to Fe(II) and the subsequent stimulation of the Fenton reaction, as well as hydroxyl radical production [12]. In this context, CdsH has been described as the major cysteine-degrading and sulfide-producing enzyme [11]. The degradation of cysteine could be the more immediate response in order to prevent the cysteine-dependent ROS. Nevertheless, in the absence of CdsH and ROS's eventual appearance, other mechanisms to circumvent the insult are plausibly triggered [53]. Besides, cysteine concentration might decrease over time due to its consumption in anabolic processes. In this context, the role of CdsH could be circumscribed to more early stages, providing a possible explanation of why the lack of CdsH does not affect bacterial replication at 24 h post-infection (Figure 3).
In addition to classic virulence factors, other factors related to various bacterial processes, such as molecule transport, protein synthesis, and metabolism under certain conditions, contribute to bacterial replication inside the host cells and thus favor the virulence. In this context, CdsH promotes bacterial replication, at least in an in vitro assay, when it is internalized in epithelial cells in a medium with extra cysteine, which could partially simulate the intestinal environment after consumption of a high-protein meal.