Introduction

In mammals, the major route for the metabolism of sulfur-containing amino acids occurs through the reverse transsulfuration pathway (RTP) via the action of cystathionine b-synthase (CBS) and cystathionine g-lyase (CTH) (Aitken and Kirsch 2005; Chiku et al. 2009; Singh et al. 2009). These cytosolic enzymes utilize pyridoxal 5′-phosphate as a cofactor and homocysteine as a first substrate to generate cystathionine. More specifically, CBS catalyzes the condensation of homocysteine with serine to form cystathionine, whereas CTH condenses cysteine with homocysteine to generate cystathionine and hydrogen sulfide (H2S). Cystathionine is then converted into cysteine, a-ketobutyrate, and ammonia by CTH. Of importance, CTH can also generate H2S by a, b-elimination of cysteine to generate serine, a, g-elimination of homocysteine to synthesize homoserine, b-replacement of cysteine to form lanthionine, and g-replacement of homocysteine leading to homolanthionine synthesis.

The biological activity of CTH in homeostatic conditions relies principally on i) the biosynthesis of the gasotransmitter H2S, which supports endothelium-dependent vasorelaxation (Yang et al. 2008), reduces oxygen- and nitrogen-mediated oxidations (Liu et al. 2022), and regulates inflammation (Stummer et al. 2023); and ii) generation of the semi-essential amino-acid cysteine, a precursor of the formation of the antioxidant tripeptide glutathione and of the broad-spectrum cytoprotective agent taurine (Huxtable 1992). However, CTH can be induced during pathological conditions and the dysregulation of the RTP may have pathophysiological consequences (Sbodio et al. 2019). Hence, we reported that pathogenic bacteria stimulate the expression of the gene Cth in macrophages through a PI3K/MTOR/SP1 signaling pathway (Gobert et al. 2019). Consequently, cystathionine synthesis is increased, thus facilitating survival of pathogens within myeloid cells, and macrophage activation is affected by putrescine accumulation and increased histone methylation (Gobert et al. 2019). Further, Cth-deficient mice exhibit reduced mucosal immune response and gastritis during infection with the gastric pathogen Helicobacter pylori (Latour et al. 2022). These data support the concept that CTH induction favors the pathogenicity of H. pylori and sustains inflammation in the stomach. In this context, the aim of the present paper was to determine the role of CTH in inflammation in the colon and, by extension, in colitis-associated carcinogenesis (CAC).

Herein, we found that Cth deletion protects mice from colitis induced by dextran sulfate sodium (DSS). However, tumorigenesis in animals treated with azoxymethane (AOM)-DSS, as a model of CAC, was similar between WT and Cth–/– mice. Mechanistically, we determined that mice lacking the gene Slc7a11, which encodes for the transporter of the anionic form of cysteine, cystine, namely solute carrier family 7 member 11 (SCL7A11, also termed xCT), are not protected from DSS colitis, suggesting that the RTP-dependent synthesis of cysteine is not involved in the pathophysiology of colitis. We also observed that CTH, but not SLC7A11, dysregulates the intestinal microbiota. Therefore, the RTP represents a potential target for clinical intervention in patients with inflammatory bowel disease (IBD).

Materials and methods

Mice and experimental models

Age-matched (8–12 wk) wild-type (WT), Cth–/– (Yang et al. 2008), and Slc7a11–/– mice (Sato et al. 2005) on the C57BL/6 background were house-bred in our animal facility and were fed 5L0D chow (LabDiet).

For the colitis model, age-matched (8–12 wk old) male mice were treated with 4% DSS (mol. wt. 36,000–50,000; TdB Labs) for 5 days, followed by 5 more days of regular water (Gobert et al. 2018, 2023; Singh et al. 2019). For the CAC model, mice received one intraperitoneal injection of AOM (12.5 mg/kg; Millipore) followed by 3 cycles of 4% DSS for 4 days beginning at day 5, 26, and 47, and were euthanized on day 56 or sooner if moribund (Gobert et al. 2021, 2022, 2023). Mice were weighed daily or weekly for the colitis and CAC model, respectively. At sacrifice, colons were removed, measured, opened longitudinally, washed, weighed, and Swiss-rolled for histology. In the AOM-DSS experiments, tumors were also counted and measured in two dimensions in the distal colon with electronic calipers under a dissecting microscope (Gobert et al. 2021, 2022, 2023). Tumor burden was determined by the sum of the areas of all tumors.

Animals were used under protocol M20000047, which was approved by the Institutional Animal Care and Use Committee at Vanderbilt University, the Vanderbilt University Institutional Biosafety Committee, and the Research and Development Committee of the Veterans Affairs Tennessee Valley Healthcare System. Procedures were performed in accordance with institutional policies, AAALAC guidelines, the AVMA Guidelines on Euthanasia, NIH regulations (Guide for the Care and Use of Laboratory Animals), and the United States Animal Welfare Act (1966).

Histological analysis

Colons were fixed in formalin and embedded in paraffin. Sections (5 μm) were stained with hematoxylin and eosin (H&E) and examined in a blinded manner by a gastrointestinal pathologist (M.K.W.). The comprehensive histological injury score for DSS colitis involves both inflammatory and epithelial damage parameters: inflammation severity (0–3) and inflammation extent (0–3) were each multiplied by the percent involvement (1 = 0–25%, 2 = 25–50%, 3 = 50–75%, and 4 = 75–100%) and added together to yield the inflammation score (0–24); the parameter of crypt damage (0–4) was also multiplied by the percent involvement to yield an epithelial injury score (0–16). These scores were then added together to yield the histologic injury score (0–40) (Gobert et al. 2018, 2022, 2023). The severity of dysplasia was determined as described for the AOM-DSS model (Gobert et al. 2022, 2023; Hardbower et al. 2017).

Immunostaining

Immunohistochemistry (IHC) was performed on the Swiss-rolled colons. Sections were deparaffinized and incubated at room temperature with 3% hydrogen peroxide in phosphate-buffered saline to block endogenous peroxidase and blocked for 1 h in Protein Block Serum-Free (Dako). Slides were sequentially incubated with i) a rabbit polyclonal anti-CTH antibody (MyBiosource; 1:100) and MACH 2 Rabbit HRP-Polymer (Biocare Medical) or ii) a rabbit monoclonal anti-IL-1b antibody (Abcam; 1:500) and with EnVision+/HRP, Rabbit, HRP (Dako). Detection was performed using 3,3′-diaminobenzidine and tissues were counterstained by hematoxylin.

ELISA

Colon tissues were lysed by the CelLytic MT Cell Lysis Reagent (Sigma) and the protein concentration was determined by Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). The concentration of IL-1b in these lysates was quantified using the Mouse IL-1 beta ELISA Kit (Proteintech).

Measurement of cysteine concentration

Frozen colon biopsies were homogenized in 0.1 M trichloroacetic acid containing 10− 2 M sodium acetate, 10− 4 M EDTA, and 10.5% methanol (pH 3.8). After a 10,000 g centrifugation for 20 min, the supernatants were used for protein assay using BCA and for LC/MS, as we reported (Gobert et al. 2019).

Analysis of mRNA expression

Total RNA was isolated using the RNeasy Mini kit (Qiagen). Reverse transcription was then performed using Superscript II Reverse Transcriptase (Invitrogen) and Oligo dT, and mRNAs were amplified using the PowerUp SYBR Green Master Mix (Thermo Fisher Scientific) and the following primers: mouse Cth: F: TGCTAAGGCCTTCCTCAAAA, R: GTCCTTCTCAGGCACAGAGG; mouse Cbs: F: TCATCCTGCCTGACTCTGTG, R: CAGCTCTTGAACACGCAGAC; mouse Il1b: F: ACCTGCTGGTGTGTGACGTTCC, R: GGGTCCGACAGCACGAGGCT; mouse Il17: F: ATCCCTCAAAGCTCAGCGTGTC, R: GGGTCTTCATTGCGGTGGAGAG; mouse Actb: F: CCAGAGCAAGAGAGGTATCC, R: CTGTGGTGGTGAAGCTGTAG. The relative expression of target genes was calculated by the 2–DDCt method, using Actb as the housekeeping gene.

Fecal microbiota analysis

Mice were euthanized and feces in the colon were harvested, weighed, and genomic DNA was extracted using the QIAamp Fast DNA Stool Mini Kit (Qiagen). DNA was quantified using Qubit Fluorometric Quantification (Thermo Fisher Scientific) and the V4 region of the 16 S rRNA gene was sequenced with the Illumina MiSeq. Sequences were processed with Mothur, version 1.44 (Schloss et al. 2009), aligned to the SILVA database release 132, and taxonomically classified with the Ribosomal Database Project classifier 16 (Cole et al. 2009; Pruesse et al. 2007). Non-bacterial sequences and chimeric sequences detected by UCHIME were removed. Operational Taxonomic Unit clustering was performed with VSEARCH, using abundance-based greedy clustering (Rognes et al. 2016). Rarefaction followed by alpha-diversity, and beta-diversity calculations were repeated 1000 times, and the results were averaged.

The 16 S rRNA sequencing of the fecal microbiota has been deposited on the Sequence Read Archive website with the BioProject ID PRJNA1096754 (Temporary Submission ID: SUB14361309).

Statistics

Figures were designed and statistics were executed using GraphPad Prism 10.3.1. All the data are expressed as the mean ± SEM. Outliers were identified using the ROUT test (Q = 5%) and removed from the analysis. Data that were not normally distributed according to the D’Agostino & Pearson normality test were log or square root transformed, and distribution was re-assessed. Student’s t test or ANOVA with the Tukey test were used to determine significant differences between two or multiple groups, respectively. When data were not normally distributed, a Kruskal-Wallis test followed by a Mann-Whitney U test was performed. Contingency analysis was performed by the Fisher’s exact test.

The relative abundance of the phyla and genus of the intestinal microbiota was analyzed by the Kruskal-Wallis test after multiple comparison adjustment by the Benjamini & Hochberg method. Alpha-diversity was estimated with the inverse Simpson index. A beta-diversity dissimilarity matrix (Bray-Curtis) was computed over the multiple rarefactions and PERMANOVA was used to test for associations between the three groups.

Results

Induction of CTH in experimental colitis

We previously reported that CTH is induced in gastric macrophages from patients with gastric inflammation and cancer (Gobert et al. 2019), as well as mice with H. pylori-mediated gastritis (Latour et al. 2022). We now found that expression of the gene Cth was significantly induced in the colon of WT mice treated with DSS compared to sham control animals (Fig. 1A). As expected, Cth mRNA was not detected in the colon of Cth–/– mice that were given DSS or not (Fig. 1A). The expression of the gene Cbs that encodes for CBS, the second enzyme of the RTP, was not affected by DSS treatment nor Cth deletion (Fig. 1A).

Fig. 1
figure 1

Induction of CTH expression in murine colitis. (A) RNA was extracted from colon biopsies from WT or Cth–/– mice treated or not with DSS. Cth and Cbs mRNA levels were determined by RT-real-time PCR and semi-quantitative analysis. (B) The colon of these mice was also immunostained for CTH. These are representative images from 5 mice per group; scale bars, 100 μm. (C) Cysteine concentration measured by LC/MS. in A and C, each dot represents a mouse; *P < 0.05, **P < 0.01 by one-way ANOVA and Tukey test

Full size image

By IHC, we observed that CTH is expressed in colonic epithelial cells (CECs), at the upper part of the crypts in C57BL/6 mice (Fig. 1B). After DSS treatment, the level of CTH immunostaining was markedly increased in CECs and was expanded to the lower part of the crypts (Fig. 1B). Deletion of Cth led to complete loss of CTH protein in the colon of mice treated or not with DSS (Fig. 1B).

Although it has been shown that daily food intake is not altered by Cth deletion (Ishii et al. 2010), we observed that cysteine, which is biosynthesized through the RTP, was significantly reduced in the colon of Cth–/– mice compared to WT mice after DSS treatment (Fig. 1C). Cth–/– naïve mice also exhibited less cysteine concentration in the colon, although the difference did not reach statistical significance (Fig. 1C).

Deletion of Cth improves DSS colitis

WT and Cth–/– mice began losing weight on day 5 after starting DSS (Fig. 2A). However, at day 9 and 10, there was significantly less body weight loss in Cth–/– mice compared to WT animals (Fig. 2A). In addition, DSS-induced colon shortening in WT mice was significantly improved in Cth-deficient mice (Fig. 2B). In WT mice treated with DSS, we observed epithelial damage, crypt loss, and strong colonic inflammation evidenced by immune cell infiltration in the colonic mucosa (Fig. 2C), consistent with our prior studies (Gobert et al. 2018, 2022; Singh et al. 2019). These histologic injuries were markedly reduced in DSS-treated Cth–/– mice (Fig. 2C). Using a comprehensive scoring system quantifying inflammation and epithelial damage, we showed a significant reduction in histologic injury in Cth–/– mice compared with WT animals in response to DSS (Fig. 2D). There was no detectable inflammation or epithelial injury in sham-treated WT or Cth–/– mice (Fig. 2C and D). Assessment of the mRNA expression of the prototype Th17 gene Il17 and the innate immune response gene Il1b demonstrated a marked increase with DSS colitis in the WT mice that was significantly attenuated in Cth–/– mice (Fig. 2E and F). Further, we found by ELISA that IL-1b protein level was increased in the colon of mice that were given DSS; however, the concentration of IL-1b protein was significantly reduced in DSS-treated Cth–/– mice compared to WT animals (Fig. 2G). This was confirmed by IHC for IL-1b, which was mainly present in the inflammatory lesions of DSS-treated mice and markedly less abundant in Cth-deficient mice (Fig. 2H).

Fig. 2
figure 2

Impact of Cth deletion on DSS colitis. WT and Cth–/– C57BL/6 mice were treated or not with 4% DSS for 5 days and then kept for 5 more days. (A) Body weights were measured daily and are shown as percentage of initial body weight. *P < 0.05 compared to Cth–/– + DSS by two-way ANOVA and Tukey test; n = 13 untreated mice for both genotypes; n = 23 WT + DSS; n = 24 Cth–/– + DSS. (B) The length of the colon was measured. C-D. Colons were Swiss-rolled and stained with H&E (C) and scored for histologic injury (D); scale bars, 100 μm. E-F. The expression of Il17 (E) and Il1b (F) mRNA was analyzed in the colon by RT-real-time PCR. G. The concentration of IL-1b was determined by ELISA. H. Immunodetection of IL-1b by IHC in the colon. These are representative images of 3 controls and 5 DSS-treated mice per genotype. *P < 0.05, **P < 0.01, ****P < 0.0001 by one-way ANOVA and Tukey test (B to F) or one-way ANOVA with the Kruskal-Wallis test, followed by the Mann-Whitney U test (G); each dot represents a mouse

Full size image

CTH does not affect inflammation-mediated tumorigenesis

Because Cth deletion protected from DSS colitis, we reasoned that it could have also a protective effect on development of CAC. At the second cycle of DSS treatment, we found that Cth–/– mice lost significantly less weight than WT mice (Fig. 3A), as in the acute DSS model. However, we found that the number (Fig. 3B) and the size (Fig. 3C) of tumors, as well as the total tumor burden per colon (Fig. 3D) was similar in WT and Cth–/– mice. Histologic assessment of H&E staining (Fig. 3E) demonstrated that the development of dysplasia was not different between the WT and Cth–/– genotypes (Fig. 3F).

Fig. 3
figure 3

Effect of CTH on colon carcinogenesis. WT and Cth–/– mice were treated or not with AOM-DSS. A. Body weights were measured weekly and are depicted as percentage of initial body weight. *P < 0.05 versus Cth–/– + AOM-DSS. After 56 days, colons were removed and tumor number (B), tumor size (C), and tumor burden (D) were determined. E. Representative images of H&E staining; scale bars, 50 μm. F. Frequency of LGD and HGD; ND, no dysplasia. P was determined by Chi-square. n = 5 and n = 14 AOM-DSS-treated mice for both genotypes

Full size image

Loss of the cystine transporter does not influence DSS colitis

One main physiological role of CTH is to generate cysteine from cystathionine, which is supported by our discovery that DSS-treated Cth–/– mice have less cysteine in their colon compared to WT mice (Fig. 1C). Therefore, one mechanism by which Cth deletion could ameliorate DSS colitis might be by reducing intracellular cysteine. To mimic a context of intracellular cysteine deprivation, we used mice lacking the transporter SLC7A11 (Sato et al. 2005). This protein imports the anionic form of cysteine in exchange for glutamate and cells from Slc7a11–/– mice have reduced intracellular cysteine (Sato et al. 2005).

We first observed that Slc7a11–/– mice lost significantly more weight than WT mice upon DSS treatment between day 6 and 8 (Fig. 4A). However, colon length was not affected by Slc7a11 deletion (Fig. 4B). Further, we did not find a difference in histologic injury between WT and Slc7a11–/– mice that were treated with DSS (Fig. 4C and D).

Fig. 4
figure 4

Colitis in Slc7a11 mice. WT and Slc7a11–/– mice were treated or not with 4% DSS for 5 days and then kept for 5 more days. (A) Body weights were assessed daily; §P < 0.05, §§P < 0.01 compared to WT + DSS. (B) The length of the colon was measured. C-D. H&E staining of the colons (C) was used to score histologic injury (D); scale bars, 100 μm. In B and D, **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way ANOVA and Tukey test. n = 10 and n = 22 DSS-treated mice for both genotypes

Full size image

CTH regulates the composition of the intestinal microbiota

We then analyzed the composition of the intestinal microbiota, which can also regulate the development of colitis, in naïve WT, Cth–/– and Slc7a11–/– mice. Bacterial community diversity, assessed by Shannon and Inverse Simpson indexes (Fig. 5A), was significantly increased in Cth-deficient mice compared to both WT and Slc7a11–/– mice. The Chao1 index, which is an indicator of total richness, was modestly increased in Cth–/– mice compared to WT animals, although this did not reach statistical significance. We found that richness was reduced in Slc7a11-deficient mice compared to Cth–/– mice (Fig. 5A). Thus, although the total number of bacterial species is similar in WT and Cth–/– mice, the Cth-deficient animals exhibit a more diverse gut microbiota. According to the principal coordinate analysis (PcoA) based on the weighted UniFrac distance, the gut microbiota was clustered into three groups based on the genotype of the mice (Fig. 5B). The fecal microbiota of WT C57BL/6 mice and Slc7a11–/– mice was mainly dominated by Bacteroidetes (Fig. 5C). We observed a significant reduction of this phylum in Cth-deficient mice that was accompanied by a significant increase of bacteria from the Firmicutes phylum (Fig. 5C). Although they were a minority, the phyla Deferribacteres and Verrumicomicrobia were less represented in Cth–/– and Slc7a11–/– mice compared to WT animals (Fig. 5C). Prevotella, Porphyromonadaceae, and Bacteroides were the prevalent genera in WT, Cth–/–, and Slc7a11–/– mice, respectively (Fig. 5D). The Lachnospiraceae genus, which belongs to the Firmicutes phylum, was significantly increased in mice lacking Cth compared to WT and Slc7a11–/– mice (Fig. 5D). There was also a marked reduction of the abundance of Akkermansia in Cth–/– mice (Fig. 5D).

Fig. 5
figure 5

Composition of the gut microbiota. DNA was extracted from colonic feces of WT, Cth–/–, and Slc7a11–/– mice. A. Alpha diversity, evaluated by the Shannon index, Inverse Simpson index, and Chao1 index; **P < 0.01, ****P < 0.0001 by one-way ANOVA and Tukey test. B. PCoA plot based on the unweighted UniFrac metric; P was determined by PERMANOVA. C-D. Bacterial community composition at phylum (C) and genus (D) levels, expressed as a ratio to the total community. *P < 0.05 by PERMANOVA

Full size image

Discussion

CTH is a major enzyme of the RTP that contributes to cysteine and H2S synthesis and, consequently, to sulfur metabolism and redox regulation. We observed that CTH is induced in the colon of mice with DSS colitis, as previously observed in rats (Flannigan et al. 2013). We also found herein that Cth-deficient mice exhibit less DSS colitis than WT mice, demonstrating that CTH supports colon inflammation. This does not appear to occur through the regulation of the mucosal concentration of cysteine, since mice lacking the cystine transporter SLC7A11 were not protected from colitis. However, the intestinal microbiota of Cth–/– mice has increased diversity compared to WT and Slc7a11–/– mice, suggesting that the gut microbial community is shaped by the host RTP. Interestingly, we also showed that Cth deletion does not affect tumorigenesis in a murine model of CAC. Therefore, we contend that CTH is a critical contributor to intestinal inflammation, but does not play a role in inflammation-induced neoplastic transformation.

In this report, we observed that Cth deletion ameliorates DSS colitis, assessed by body weight loss, colon length, and, more importantly, histological injury score; this overall score includes inflammatory parameters, epithelial injury, and their extent (Gobert et al. 2018, 2022, 2023). Accordingly, we found that the expression of the genes encoding for IL-1b and IL-17, two major cytokines that play a pathogenic role in colitis (Coccia et al. 2012; Ito et al. 2008), was also reduced in the colonic mucosa of Cth–/– mice. We also confirmed that IL-1b protein was upregulated in DSS colitis and attenuated in Cth–/– mice, which we demonstrated by both ELISA and IHC. Altogether, these data evidence that CTH supports the development of colitis in our DSS model. Supporting the contention that this enzyme can mediate inflammation in the gastrointestinal tract, we recently reported that Cth deletion dampens macrophage and T cell activation in the stomach of H. pylori-infected mice, leading to a reduction of gastritis (Latour et al. 2022). In addition, human and murine T cell activation and proliferation are reduced by silencing CTH and/or CBS (Miller et al. 2012). However, Akahoshi et al. have shown that C57BL/6 mice lacking Cth or Mpst, which encodes for the H2S generating enzyme mercaptopyruvate sulfurtransferase (THTM), do not show significant alteration of trinitrobenzene sulfonic acid (TNBS)- and oxazolone-induced colitis (Akahoshi et al. 2023). In contrast, the deleterious effect of Cth deletion on the pro-inflammatory response, including NLRP3 inflammasome activation and innate cytokine production, has been also reported (Qin et al. 2019); unfortunately, histological colitis was not investigated in this study (Qin et al. 2019). Further, Thanki et al. have reported that clinical and histological parameters are worsened in Cth–/– mice, but these authors used a colitis score developed for ulcerative colitis (UC) patients (Geboes et al. 2000), which does not integrate epithelial injury nor extent of inflammation and damage (Thanki et al. 2022); moreover, these experiments were performed in mice with a C57BL/J6;129SvEv mixed background treated with 3% DSS (Thanki et al. 2022). Discrepancies between these studies and ours could be attributable to the differences of experimental procedures. It is also highly probable that the intestinal microbiota of the mice that have been used in these different studies differs due to dissimilar genetic backgrounds and animal facilities.

Because we found that Cth–/– mice were protected from DSS colitis, we reasoned that this gene deletion could also influence the development of CAC induced by AOM and DSS. Surprisingly, we found that tumorigenesis and severity of dysplasia were identical in WT mice and those lacking CTH. Similarly, WT and Cth–/– C57BL/J6;129SvEv mice treated with AOM-DSS displayed similar tumor burden after 80 days (Thanki et al. 2022). These data suggest that CTH is not a major contributor to carcinogenesis and tumor expansion in the inflamed colon. Nonetheless, it would be interesting to test the role of this enzyme in sporadic colorectal cancer (CRC), notably because CTH i) is stimulated by the WNT pathway in human CRC cells (Fan et al. 2014) and ii) is more highly expressed in human colonoids that harbor mutations in the genes implicated in CRC development, i.e., APC and TP53, TP53 and SMAD4, or TP53, SMAD4, and KRAS (Ascencao et al. 2022). The CTH/H2S pathway results in proliferation and migration of CRC cells (Ascencao et al. 2022; Fan et al. 2014; Szabo et al. 2013), and growth of tumor xenografts in nude mice (Fan et al. 2014).

The activity of CTH has a major impact on amino acid metabolism and contributes to generation of numerous metabolites. In this context, it is probable that the deleterious effect of CTH on DSS colitis that we have observed has a multifactorial etiology. First, CTH generates cysteine and we observed that Cth–/– mice have less cysteine in their colon compared to WT mice. Therefore, we reasoned that one mechanism by which CTH could support DSS colitis might be by reducing intracellular cysteine. Because cysteine is an antioxidant, the general concept is that this amino acid and its derivatives protect from inflammation. However, controversial results have emerged from a recent study showing that treatment with N-acetylcysteine increases the level of malondialdehyde, a reactive electrophile that we have implicated in colitis and CAC (Gobert et al. 2023), in the colon of mice with DSS colitis (da Paz Martins et al. 2023). To mimic a context of intracellular cysteine deprivation, we used mice lacking the cystine importer SLC7A11 (Sato et al. 2005). Note that de novo synthesis of cysteine and/or its import via other transporters cannot compensate for the need for intracellular cysteine in cells lacking SLC7A11 (Guo et al. 2023; Jyotsana et al. 2022). We found that DSS colitis was similar between WT and Slc7a11–/– mice, suggesting that the deleterious effect of CTH in our model occurs independently of intracellular cysteine level.

Second, H2S is a gasotransmitter with anti-inflammatory properties synthesized by CTH and CBS during the RTP. Rectal injection of the H2S donor, NaHS, protects Wistar rats from dinitrobenzene sulfonic acid colitis (Goyal et al. 2015) and daily intraperitoneal injections dampens DSS colitis in C57BL/6 mice (Qin et al. 2019); in this last study, the treatment with NaHS was not investigated in Cth-deficient mice, which exhibited increased colitis (Qin et al. 2019). Thus, these data demonstrate that exogenous H2S might be protective in the colon, but do not directly establish that CTH exerts a protective effect through H2S. In contrast, excess H2S production has been suggested to be involved in the pathogenesis of IBD (Mottawea et al. 2016; Rowan et al. 2009). Fecal sulfide concentration and production were found to be elevated in patients with UC (Pitcher et al. 2000; Levine et al. 1998). Thus, H2S may cause mucus disruption by reducing its disulfide bonds (Ijssennagger et al. 2016), and the administration of the H2S-producing bacterium Desulfovibrio indonesiensis and Atopobium parvulum exacerbates the development of TNBS colitis (Figliuolo et al. 2017) and spontaneous colitis in Il10−/− mice (Mottawea et al. 2016), respectively. Interestingly, we observed that WT mice with DSS colitis have increased CTH expression in their colon, which can in turn lead to increased H2S generation. In this context, Cth deletion could lead to the amelioration of DSS colitis by reducing endogenous H2S generation. However, it should be recognized that the cysteine aminotransferase/THTM metabolic pathway is a major source of colonic H2S compared to the RTP in healthy and inflamed colon (Flannigan et al. 2013).

Third, we investigated whether CTH can modulate the composition of the intestinal microbiota, which can play a major role in the pathophysiology of intestinal inflammation (Lee and Chang 2021). First, we found an overall increased diversity and richness in the fecal microbiota of Cth–/– mice compared to both WT and Slc7a11–/– mice, although the b-diversity analysis indicated that bacterial communities of the three genotypes are dissimilar. Loss of diversity is mainly associated with development of more severe colitis in mice (Singh et al. 2019) and is a hallmark of IBD pathogenesis (Frank et al., 2007; Willing et al., 2010), suggesting that CTH may support colon inflammation by reducing diversity. Further, the taxonomic distribution of the bacteria in each genotype indicated that the fecal microbiota of C57BL/6 mice, as we previously reported (Gobert et al. 2022, 2023; Singh et al. 2019), and Slc7a11–/– mice was mainly dominated by Bacteroidetes; within this phylum, Prevotella and Bacteroides were the prevalent genera in WT and Slc7a11–/– mice, respectively. Cth deficiency was associated with a reduction of Bacteroidetes and an increase of Firmicutes. At the genus level, Porphyromonadaceae (Bacteroidetes) and Lachnospiraceae (Firmicutes), which are associated with the resistance of mice to infectious colitis (Ferreira et al. 2011), were dominant in Cth–/– mice. Interestingly, the prevalence of Bacteroidetes is often increased and Firmicutes reduced in patients with IBD (Santana et al. 2022), leading to the concept that a decrease of the Firmicutes/Bacteroidetes ratio is considered as a dysbiosis and potentially contributes to the etiology/progression of colitis (Stojanov et al. 2020). Our present data showed an increased Firmicutes/Bacteroidetes ratio in Cth–/– mice, which are protected from DSS colitis, suggesting that CTH activity in the colon may sustain a pro-inflammatory microbiota through the dysbiosis of this ratio. In this global context, the mechanism by which host CTH activity affects the ecology of the microbiological intestinal community remains to be elucidated.

The RTP is essential for homeostasis. However, we found that dysregulation of this metabolic pathway, notably during infection, results in increased pathology (Latour et al. 2022). The present paper shows that a similar concept occurs for colitis. Further investigations are warranted to define the molecular/cellular mechanism by which CTH supports colonic inflammation. However, we present herein the first line of evidence that the host RTP regulates the composition of the intestinal microbiota as a potential causal factor in colitis.