Immunotherapy can elicit robust anticancer responses in the clinic. However, a large proportion of patients with colorectal cancer do not benefit from treatment. Although previous studies have shown that hydrogen sulfide (H2S) is involved in colorectal cancer development and immune escape, further insights into the mechanisms and related molecules are needed to identify approaches to reverse the tumor-supportive functions of H2S. Here, we observed significantly increased H2S levels in colorectal cancer tissues. Decreasing H2S levels by using CBS+/− mice or feeding mice a sulfur amino acid-restricted diet (SARD) led to a marked decrease in differentiated CD4+CD25+Foxp3+ Tregs and an increase in the CD8+ T-cell/Treg ratio. Endogenous or exogenous H2S depletion enhanced the efficacy of anti–PD-L1 and anti–CTLA4 treatment. H2S promoted Treg activation through the persulfidation of ENO1 at cysteine 119. Furthermore, H2S inhibited the migration of CD8+ T cells by increasing the expression of AAK-1 via ELK4 persulfidation at cysteine 25. Overall, reducing H2S levels engenders a favorable immune microenvironment in colorectal cancer by decreasing the persulfidation of ENO1 in Tregs and ELK4 in CD8+ T cells. SARD represents a potential dietary approach to promote responses to immunotherapies in colorectal cancer.

Significance:

H2S depletion increases the CD8+ T-cell/Treg ratio and enhances the efficacy of anti–PD-L1 and anti–CTLA4 treatment in colon cancer, identifying H2S as an anticancer immunotherapy target.

Colon cancer is the second leading cause of cancer mortality worldwide, with a minority of advanced-stage patients showing an objective response to immunotherapies based on immune-checkpoint inhibitors (ICI; refs. 1, 2). Microsatellite instability (MSI) is correlated with increased tumor mutation burden and is considered a marker of the response to ICIs in colon cancer (3). Anti–PD-1 therapies, including nivolumab, can significantly improve overall survival in some colon cancer patients with MSI tumors (4, 5). However, fewer than 15% of patients with advanced colon cancer have MSI tumors, and the remaining 85% with microsatellite stability (MSS) tumors hardly benefit from ICIs (6). The reduced expression of cancer antigens, upregulation of immune checkpoints and their ligands, and enrichment of immunosuppressive cells have been shown to contribute to the immunosuppressive tumor microenvironment (TME) in colon cancer (7–9). MSS colon cancer is characterized by a low mutation burden and decreased expression of neoantigens and tumor lymphocyte infiltration. Activating the immune microenvironment in MSS colon cancer and transforming it into an immune “hot tumor” has become the focus of current research (10). Unlike solid tumors at other sites, colon cancer is constantly exposed to large quantities of metabolites from the intestinal microbiota. Investigating the unique metabolites enriched in the colon cancer TME might provide novel approaches for improving the therapeutic effect of ICIs in colon cancer.

Regulatory T cells (Treg), a subset of immunosuppressive CD4+ T cells potentiated by activation of the forkhead box P3 (FOXP3) signaling pathway, play an essential role in fostering tumor immune evasion, and the enrichment of Tregs is associated with unfavorable prognosis in colon cancer (11, 12). The recruitment of Tregs is dependent on gradients of multiple chemokine and cytokine reagents, including chemotactic cytokines and their ligands (13, 14). Embedded in a relatively nutrient-deficient environment, Tregs in the TME utilize both metabolic reprogramming and defensive adaptations to gain a survival advantage over effector T cells. Tregs take up extra free fatty acids and lactate by upregulating CD36 and the lactate transporter MCT1 to thrive and initiate their immunosuppressive function (15, 16). Depleting Tregs through antibody-dependent cellular cytotoxicity has been suggested as a major contributor to the therapeutic effect of anti–CTLA4 antibodies (17).

As the third confirmed gaseous transmitter after NO and CO, hydrogen sulfide (H2S) promotes Treg differentiation in both the vascular endothelium and kidneys of patients with hypertension through LKB1 persulfidation (18). A previous study also revealed the role of H2S in the DNA demethylation of Treg-specific FOXP3 via nuclear transcription factor Y subunit beta (NFYB) persulfidation in systemic autoimmune diseases (19). Both intestinal epithelial cells and gut microbes produce H2S, utilizing dietary sulfur amino acids (methionine and cystine), and recently, higher sulfur microbial diet scores were shown to be associated with an increased risk for both early-onset colon adenomas and colon cancer (20–22). Methionine is a key nutritional factor that shapes Th-cell proliferation and functions in part through the regulation of histone methylation (23). Cysteine restriction via the administration of cyst(e)inase or interference with the expression of SLC7A11, the cytomembrane cystine/glutamate transporter, induced a drastic increase in ferroptosis in pancreatic tumor cells due to depletion of intracellular glutathione (GSH) and ensuing elevated reactive oxygen species levels (24, 25). Increased endogenous H2S production was also observed in diet restriction-mediated longevity (20, 26). Tumor-derived H2S, mainly produced by cystathionine-β-synthase (CBS), stimulates bioenergetics, cell proliferation, and peritumoral angiogenesis in colon cancer (27). Our previous studies have mainly focused on the involvement of decreased cellular H2S production in the pathogenesis of ulcerative colitis and increased H2S synthesis in acquired resistance to 5-fluorouracil in colon cancer cells (28, 29). Recently, we found that H2S inhibits ferroptosis in colon cancer cells by stabilizing SLC7A11 via OTUB1 persulfidation at cysteine 91 (30). However, the involvement of H2S in orchestrating the formation of an immunosuppressive TME in colon cancer and the therapeutic potential of depleting H2S through manipulating dietary sulfur amino acids have not been illustrated.

In the current study, we delineated the mechanisms underlying the immunosuppressive effect of H2S in colon cancer and investigated its diverse effects on both Tregs and CD8+ T cells and the underlying mechanisms. In addition, we show for the first time that depleting tumor H2S via a sulfur amino acid-restricted diet (SARD) dramatically improved the therapeutic effect of ICIs in both MSI and MSS orthotopic colon cancer models. Our results provide a foundation for exploiting H2S and sulfur amino acid-targeting strategies to promote responses to immunotherapies in colon cancer.

Cell lines and reagents

The C57BL/6 mouse colon cancer epithelial cell line MC38, BALB/c mouse colon cancer epithelial cell line CT26, human acute T lymphocytic leukemia cell line Jurkat, normal colon epithelial cell line NCM356, and human colon cancer cell lines DLD-1, SW620, HT-29, HCT-116, HCT-8, and Caco-2 were purchased from the National Biomedical Laboratory Cell Resource Bank. MC38, CT26, Jurkat, and NCM356 cells were cultured in the RPMI-1640 medium (Thermo Fisher Scientific). DLD-1, SW620, HT-29, HCT-8, and Caco-2 cells were cultured in modified DMEM (Thermo Fisher Scientific). HCT-116 cells were cultured in McCoy's 5A medium (Thermo Fisher Scientific). Cell lines were cultured with the above-mentioned media supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37°C and 5% CO2. Dichloromethane-free GYY4137 sodium (1975149-21-3, Sigma), a sustained H2S-release agent, was dissolved in sterile phosphate-buffered saline (PBS; Thermo Fisher Scientific). DMEM/F12 medium (11320033, Gibco) supplemented with B27 (17504044, Gibco), human EGF (20 ng/mL, AF-100-15, PeproTech), FGF10 (20 ng/mL, 100-26, PeproTech), human HGF (10 ng/mL, 100-39, PeproTech), and 1% penicillin/streptomycin was used in spheroid assays. All cell lines were genotyped by short tandem repeat analysis and were Mycoplasma-free at the time of testing according to the Mycoplasma Detection Kit (CA1080, Solarbio).

Mice

C57BL/6N or BALB/c wild-type (WT) mice were bred inside the Animal Center of Peking University First Hospital. The mouse CBS gene (GenBank accession number: NM_144855.3; Ensembl: ENSMUSG00000024039) is located on mouse chromosome 17 and exon 3 to exon 14 was selected as target sites. Cas9 mRNA and gRNA generated by in vitro transcription were injected into fertilized eggs for knockout mouse productions. The construction detail was shown in Supplementary Fig. S1A. Groups of 6 to 16 mice were used for the subcutaneous growth model and in situ growth model. All mouse experiments were approved by the Animal Ethics Committee of Peking University First Hospital. Genotype identification was performed by PCR using tail genomic DNA, and the primer sequences were as follows: mouse CBS-F: 5′-AGG GAG AAG TTA CAT CAT GCC TTG G-3′, mouse CBS-R: 5′-AGT ATC CAG GGC TTG ACA TCC TTA colon cancer-3′, and mouse CBS-Wt/He-F: 5′-CTG ATG CGG TTT CCT AGC AAC AG-3′. The annealing temperature was set to 65°C. The sizes of the PCR products from the CBS/+ mice were 1,055 bp and 810 bp. The size of the PCR product from the WT mice was 1055 bp, and the size of the PCR product from the CBS−/− mice was 810 bp.

Antibodies

The antibodies used in our study are listed in Supplementary Table S1.

Mouse genotype identification

For the genotype identification of the transgenic mice (WT or CBS/+), a mouse tail DNA extraction box (MiniBEST Universal Genomic DNA Extraction Kit ver. 5.0; 9765, Takara) and mouse tail cDNA-PCR amplification reagent (TaqTM Hot Start version; R007A, TaKaRa) were used in our study.

Tissue microarray and IHC

Twenty paired colon cancer samples and matched adjacent tissues collected from patients who received surgeries at Peking University First Hospital were included in this study. The data for the patients included are summarized in Supplementary Table S2. Briefly, these specimens were immersed in 10% formalin (ZLI-9381, ZSGB-BIO) immediately after separation. After routine deparaffinization and hydration, the antigens in the tissue microarray (TMA) specimens were repaired with pH 6.0 citrate buffer (ZLI-9065, ZSGB-BIO). The catalase activity of the colon cancer tissues was blocked using 3% hydrogen peroxide. Normal goat serum (ZLI-9056, ZSGB-BIO) was used to reduce nonspecific adsorption of the tissue to the detection antibody. Finally, the TMA was incubated with the primary and secondary antibodies (PV-6000, ZSGB-BIO) in sequence and stained with DAB solution (ZLI-9017, ZSGB-BIO).

Ethics approval

Healthy donors (HD) gave informed consent to participate in our study before taking part. This study was approved by the Institutional Review Board at Peking University First Hospital, Beijing, China.

Immunofluorescence staining

Colon cancer tissues were soaked in 10% formalin and fixed overnight. Next, the samples were dehydrated in 70% ethanol, embedded in paraffin, and cut into 4-μm sections. After routine deparaffinization and hydration, followed by antigen retrieval with pH 6.0 citrate buffer, the following primary antibodies were used: rabbit anti-human CBS and mouse anti-human FOXP3, followed by goat anti-rabbit IgG (H+L) cross-adsorbed secondary antibody, Alexa Fluor 488 (A-11008, Invitrogen) and goat anti-mouse IgG (H+L) cross-adsorbed secondary antibody, Alexa Fluor 647 (A-21235, Invitrogen). Fluorescence was visualized under an LSM710 confocal microscope (Carl Zeiss).

Measurement of H2S levels in colon cancer tissues using a sulfide-selective electrode

A sulfur ion–selective electrode (DX232, METTLER TOLEDO) was adopted to detect tissue H2S levels as described previously with slight modifications (31). The H2S concentrations of the homogenate were calculated utilizing a standard curve generated with serial dilutions of an NaHS solution following the manufacturer's instructions.

T-cell isolation and sorting

Written informed consent was collected from participants after approval by the institutional review board. Peripheral blood mononuclear cells (PBMC) were isolated from heparinized whole blood of HDs by density gradient centrifugation over Ficoll/Hystopaque (Sigma-Aldrich; ref. 32). According to the instructions of the EasySep Human CD4 Positive Selection Kit II (cat #17852) and EasySep Human CD8 Positive Selection Kit II (cat #17853), highly purified CD4+ and CD8+ T cells, respectively, were isolated from fresh human PBMC samples by immunomagnetic positive selection.

Treg induction

Using ImmunoCult-XF T-Cell Expansion Medium (STEMCELL Technologies, cat. #10981), ImmunoCult Human Treg Differentiation Supplement (STEMCELL Technologies, cat. #10977) supplemented with ImmunoCult Human CD3/CD28 T-Cell Activator (STEMCELL Technologies, cat. #10971), and human recombinant IL2 (CHO-expressed; STEMCELL Technologies, 40 ng/mL, cat #78036), the CD4+ cells purified from fresh human PBMCs samples and the Jurkat cell line were induced into Tregs following the manufacturers’ instructions.

CD8+ T-cell induction

Using ImmunoCult-XF T-Cell Expansion Medium (STEMCELL Technologies, cat #10981) supplemented with ImmunoCult Human CD3/CD28 T-Cell Activator and human recombinant IL2 (CHO-expressed), CD8+ T cells were purified from fresh PBMC samples from HDs.

Treg suppression assay

CD8+ T cells were cultured in 12-well plates with WT Treg or C119A Treg at a 4:1 ratio. Cell cultures were performed in ImmunoCult-XF T-Cell Expansion Medium supplemented with ImmunoCult Human CD3/CD28 T-Cell Activator, Human Recombinant IL2 (CHO-expressed) and GYY4137 (200 μmol/L). After 2 days of coculture, the proliferative activity of CD8+ cells was determined by flow cytometry. Cells were stained with CD8-FITC (BioLegend) for 20 minutes at room temperature. After surface staining, cells were fixed and permeabilized using the FOXP3/Transcription Factor Staining Buffer Set (BioLegend) and then were stained with ki67-PE(BioLegend) for 30 minutes. The samples were analyzed within 30 minutes by flow cytometry while kept on ice.

Western blotting

Total protein was extracted by RIPA supplemented with proteinase inhibitors, phosphatase inhibitors, and PMSF (Beyotime). The protein concentration was measured with a bicinchoninic acid (BCA) protein assay kit (KeyGEN) and adjusted such that the samples contained equal amounts of protein. Protein lysates were separated by SDS-PAGE (4%–12%) and transferred to polyvinylidene fluoride (PVDF) membranes. After blocking with 5% bovine serum albumin, the target proteins were detected with primary antibodies and secondary antibodies corresponding to the species.

Quantitative real-time PCR

Total RNA from the tissues or cells was isolated using TRIzol reagent according to the manufacturer's instructions. The cDNA was transcribed from RNA using PrimeScript RT Reagent Kit (Vazyme), which was used to determine the relative expression levels of GADPH, FOXP3, IFNγ, IL10, SQR, TST, and ETHE1 with qPCR SYBR Green Master Mix (Vazyme). Data were analyzed using the 2−ΔΔCt method, with GADPH serving as an internal control. Primers sequence were as follows: Human GAPDH-F: 5′-GGA GCG AGA TCC CTC CAA AAT-3′; Human GAPDH-R: 5′-GGC TGT TGT CAT ACT TCT CAT GG-3′; Human FOXP3-F: 5′-GTG GCC CGG ATG TGA GAA G-3′; Human FOXP3-R:5′-GGA GCC CTT GTC GGA TGA TG-3′; Human IFNγ-F: 5′-TCG GTA ACT GAC TTG AAT GTC CA-3′; Human IFNγ-R: 5′-TCG CTT CCC TGT TTT AGC TGC-3′; Human IL10-F: 5′- GAC TTT AAG GGT TAC CTG GGT TG-3′; Human IL10-R:5′-TCA CAT GCG CCT TGA TGT CTG-3′; Human SQR-F: 5′-AGC TAG AGT GAC TGA GTT GAA colon cancer-3′; Human SQR-R: 5′-AGC TGG ATT CCG AGA GCA ATA A-3′; Human TST-F: 5′- GTG GAT GTT CCG TGT GTT TGG-3′; Human TST-R: 5′-CAG CAC CTG CTC GTA GGT C-3′; Human TST-F: 5′- GTG GAT GTT CCG TGT GTT TGG-3′; Human TST-R:5′-CAG CAC CTG CTC GTA GGT C-3′; Human ETHE1-F: 5′-GAG GCC GTT CTG ATC GAC C-3′; Human ETHE1-R:5′- GCA GTG GGT ATT CAC AGC ATA G-3′.

Cell counting kit-8 assay

Cells were seeded into 96-well plates (5,000 cells/well). Cell viability was measured using Cell Counting Kit (CCK)-8 assay kits (Bimake, cat #B34304) in accordance with the manufacturer's instructions. The optical density (OD at 450 nm) was measured using a microplate reader.

Subcutaneous tumor model

For MC38 subcutaneous tumor formation, cells were inoculated subcutaneously (s.c.) into the armpits of male mice. On days 9 to 10 after tumor inoculation, WT mice and CBS/+ mice were randomized based on tumor size and treatment initiated when tumors reached 50 to 100 mm3, and then anti-IgG (10 mg/kg body weight), anti–PD-L1 (10 mg/kg body weight), and anti–CTLA4 (10 mg/kg body weight) antibodies were applied every 4 days for 4 doses. Digital calipers were used to measure the length and width of the tumors. Subcutaneous tumor growth was measured every 3 days, and tumor volume was calculated as follows: length × (width2)/2 (33). When the subcutaneous tumors reached 2 cm in every dimension or skin ulceration occurred, the mice were killed. The mice were euthanized by carbon dioxide asphyxiation, and the survival rate was also calculated.

Orthotopic murine colon cancer model (MC38 and CT26 cells)

All animal studies were approved by the Institutional Review Board at Peking University First Hospital. Six- to 8-week-old male C57BL/6 mice were purchased from Vital River. A two-step process was undertaken to induce the formation of orthotopic syngeneic colon tumors. In the first step, 1 × 106 MC38 tumor cells were inoculated s.c. into the right flank of the mice. At a tumor size of approximately 500 mm3, the donor mice were sacrificed, and the tumors were harvested. Necrotic areas were discarded, and the remaining tumor tissues were cut into 2-mm3 fragments and stored in ice-cold PBS. For surgical orthotopic implantation, mice were anesthetized with isoflurane (1.5%). The abdomen was shaved and disinfected. Subsequently, a 5-mm laparotomy was conducted. The cecum was exposed, and a tumor fragment was implanted below the serosa. The cecum was returned to the abdomen, and the peritoneum and skin were closed by suture. The same procedures were performed in BALB/c mice with CT-26 cell lines.

Tumor processing and flow cytometry

To determine the effect of CBS and anti–PD-L1 immunotherapy on various lymphocytes, tumor-bearing C57BL/6 and BALB/c mice (WT and CBS/+) with orthotopic MC38 and CT26 tumors, respectively, were used. All of the mice were treated with a single dose of IgG1 or anti–PD-L1 antibody on day 10 before being sacrificed on day 13 (34).

Tumors were cut into small pieces using scissors and digested with a mixture of 0.33 mg/mL DNase (Sigma-Aldrich) and 0.27 mg/mL Liberase (Sigma-Aldrich) in Hanks’ balanced salt solution for 30 minutes at 37°C before being dispersed through a 40-μm filter (17, 35, 36). The digested tumor suspension was washed twice with PBS and resuspended in PBS for subsequent staining.

To identify activated CD8+ T-cell populations, we dyed the surface markers CD8 and CD69 in the samples. For intracellular transcription factor staining, according to the manufacturer's protocol, cells were fixed and permeabilized after surface staining using the FOXP3/transcription factor staining buffer set (Invitrogen, lot 2052253). To identify Treg populations, the samples were surface stained with CD4 and CD25, and FOXP3 was used for nuclear staining (36, 37).

Multiplexed immunofluorescence staining

To assess the density of immune cells that make up the TME, multiplexed immunofluorescence (IF) staining of formalin-fixed paraffin-embedded (FFPE) slides containing MC38 and CT26 tumors was performed (38). Four-micron slices were cut from subcutaneous MC38 or CT26 tumors onto charged slides, and the slides were baked at 65°C for 2 hours. After routine deparaffinization and hydration, the tumor sections were washed in distilled water for 3 minutes and then fixed with 10% formalin to prevent fading. Multiplexed IF staining of FFPE tissues was performed using a PANO 5-plex IHC kit (cat. #0002100050, Panovue) according to the manufacturer's instructions (39). Antibodies against FOXP3, CD3, PD-1, CD8, and DAPI (Sigma‒Aldrich, D9542) were sequentially applied, followed by incubation with HRP-conjugated secondary antibody and tyramide signal amplification. The dyes corresponding to the four antibodies above were 520, 650, 540, and 620, respectively. To remove the already bound primary and secondary antibodies, AR9 antigen retrieval buffer (pH 9.0) was used for antigen retrieval. The fluorophore and tissue antigens were covalently bound and were not released due to microwave repair.

Multispectral imaging and analysis

The Mantra system (PerkinElmer) was used to obtain multispectral images. With the same exposure time, this device can capture fluorescence spectra from 420 to 720 nm at a wavelength interval of 20 nm. For each stained slide, we randomly selected 10 areas for image capture and subsequent data analysis. The unstained images were used to extract the autofluorescence spectrum of the tissue, whereas images of each single-stained section were used to extract the autofluorescence spectrum of each fluorescein label utilizing inForm image analysis software (PerkinElmer; ref. 40).

Magnetic resonance imaging

For magnetic resonance (MR) measurements, tumor-bearing mice were anesthetized with isoflurane (1.5%). The MR temperature was held constant at 37°C. With the help of professional technicians, tumor volume was quantified (41).

Spheroid assays

DLD-1 cells were digested completely, added to each well (1,000 cells/well) of a Costar ultralow-attachment 24-well plate (Corning), and cultured in a homemade DMEM/F12 medium. After 2 weeks, DLD-1 spheroids had formed. Next, PBMCs derived from HDs were added for coculture (3×106 cells/well). Before starting the coculture, CD8+ T cells sorted from PBMCs were stained with 0.1 μmol/L carboxyfluorescein succinimidyl ester (CFSE; Thermo Fisher) for 30 minutes. After 24 hours, the infiltrated spheroids were collected, washed with PBS, fixed with 4% polyformaldehyde at room temperature for 30 minutes, and incubated in a 30% sucrose solution overnight. Then, the spheroids were embedded in the Tissue-Tek OCT compound (Sakura). Next, the spheroid blocks were sliced into 10-μm-thick sections using a CM1520 cryostat (Leica), and the slides were mounted with DAPI-containing Fluoromount-G (Thermo Fisher; ref. 42). Finally, fluorescence was visualized under an LSM710 confocal microscope (Carl Zeiss).

Chromatin immunoprecipitation assays

With reference to previous research methods, we performed chromatin immunoprecipitation (ChIP) assays (43). In short, 1% formaldehyde was used to cross-link proteins and DNA, followed by ultrasonic disruption. Cell lysates were immunoprecipitated with normal rabbit serum (BD Biosciences) or anti-enolase 1 (ENO1) antibody (Santa Cruz Biotechnology)/anti-ELK4 antibody (Novus). With real-time PCR using an ABI 7700 instrument, the eluted immunoprecipitated DNA was amplified. Values were normalized to the corresponding input control and are expressed as the fold change compared with expression in normal rabbit serum for each experiment. The specific primer sequences corresponding to the FOXP3 gene-binding site were as follows: 5′-GGT TGG CCC TGT GAT TTA TTT TAG-3′ and 5′-GTG TGG AAG CCG CAG ACC TC-3′ (44). The specific primer sequences corresponding to the AAK1 gene-binding site were as follows: 5′-TTC TCG CGG TCG ACT CCT-3′ and 5′-CTC CAT CCA CCC AAC CAG GAA-3′.

Modified biotin switch assay

According to the instructions of a biotin switch assay kit (Abcam, ab236207), the modified biotin switch assay was performed in our study (45,46). In short, Jurkat cells were homogenized in HEN buffer supplemented with 1% Nonidet-P40 (NP-40), 150 μmol/L deferoxamine, and protease and phosphatase inhibitors. After the homogenate was sonicated, the total protein was extracted by centrifugation (13,000×rpm for 20 minutes). Next, HEN buffer supplemented with 20 mmol/L methyl methanethiosulfonate (MMTS; cat #23011, Thermo Fisher) and 2.5% SDS was used to block sulfhydryl groups. Cold acetone was used to remove the MMTS and precipitated proteins. The protein pellet was suspended in HEN buffer and labeled with 4 mmol/L biotin-HPDP (25°C for 3 hours). After precipitation with cold acetone and redissolution in HEN buffer, the protein enriched in the streptavidin–agarose beads (cat #L00353, GenScript) was subjected to Western blotting with anti-ENO1 and anti-ELK4 antibodies. Cells treated with dithiothreitol (DTT; 1 mmol/L, Beyotime Biotechnology) were used as a negative control group (30).

Statistical analysis

With GraphPad 8.0 software, the Mann–Whitney test and one-way ANOVA were performed to analyze data from two groups and more than two groups, respectively. Data are expressed as the mean ± standard error of the mean. Differences for which P < 0.05 were considered statistically significant.

Data availability

Publicly available data analyzed in this study were obtained from Gene-Expression Omnibus (GEO) at GSE59241. The RNA-seq data generated from this study have been uploaded to GEO with the accession number GSE214622.

Increased H2S levels correlate with a decreased CD8+ T-cell/Treg ratio in colon cancer

Significantly increased H2S levels were identified in colon cancer tumor tissues (Fig. 1A), and decreased infiltration of CD8+ T cells and increased infiltration of Tregs (stained with FOXP3 and CD4) were associated with elevated H2S levels (Fig. 1,B and C). The sulfide oxidation pathway (SOP), composed of SQR, TST, and ETHE1, might be involved in preventing the toxic accumulation of H2S in colon epithelial cells. We subsequently investigated the expression of these enzymes in a TMA composed of 20 paired colon cancer tissue samples and adjacent normal tissue samples, and the results indicated significantly increased expression of all three enzymes in the tumor tissues (Fig. 1D). The expression of these enzymes was further investigated in colon cancer cell lines (DLD-1, SW620, HT-29, HCT-226, and HCT-8 cells), a normal colon epithelial cell line (NCM356), Jurkat cells and CD8+ T cells. Intriguingly, increased expression of all three enzymes was revealed in the colon cancer cell lines in comparison with the normal colon epithelial cells, and the expression of all these enzymes was decreased in both Jurkat and CD8+ T cells (Fig. 1E). In addition, the mRNA levels followed the pattern set by the protein levels (Fig. 1F). We further investigated the effect of increasing doses of GYY4137, a slow-releasing H2S donor, on the proliferation of the above-mentioned cell lines, and the results indicated that low doses of GYY4137 significantly increased the proliferation of colon cancer cell lines (HCT-116 and SW620 cells). However, GYY4137 inhibited the proliferation of both the normal colon epithelial cell line and Jurkat cells in a dose-dependent manner (Fig. 1G). We further analyzed the levels of H2S levels in supernatants collected from different cell lines after adding increasing doses of GYY4137 for 72 hours. The results indicated that GYY4137 increased the H2S levels in a dose-dependent manner and the H2S levels in the supernatants of both HCT-116 and SW620 cell lines were significantly increased compared with NCM356 and T cells without GYY4137, depicting an increased endogenous production of H2S in colon cancer tumor cell lines (Fig. 1H). These results indicated that increased H2S levels are associated with both the decreased infiltration of CD8+ T cells and increased infiltration of Tregs in colon cancer.

Figure 1.

Increased H2S correlates with a decreased CD8+ T-cell/Treg ratio in colon cancer. A, Forty colon cancer tissues and paired normal colon epithelial tissues were collected, and a sulfide-selective electrode was utilized to detect the H2S levels in tissue homogenates. B, IHC analysis of CD8+ T-cell infiltration was performed in the 40 patients included in A, and a comparison of H2S levels between high CD8+T infiltration and low CD8+T infiltration tumors was performed. C, Immunofluorescence analysis of FOXP3+ CD4+ T-cell infiltration was performed in the 40 patients included in A, and a comparison of H2S levels between high Treg infiltration and low Treg infiltration tumors was performed. D, IHC analysis of SQR, TST, and ETHE1 was performed in a TMA composed of 20 colon cancer tissues and paired adjacent normal colon epithelial tissues. E, Western blot analysis of the expression of SQR, TST, and ETHE1 in NCM356, DLD-1, SW620, HT-29, HCT-116, HCT-8, Jurkat, and CD8+ T cells. GAPDH served as an internal control. F, Real-time PCR analysis of the expression of SQR, TST, and ETHE1, with GAPDH serving as an internal control. G, CCK-8 analysis of the effect of increasing doses of GYY4137 on the viability of HCT-116, SW620, NCM356, and CD8+ T cells. A total of 3000 cells were seeded per well and incubated with increasing doses of GYY4137 for 72 hours before collection for CCK-8 analysis. H, H2S levels in the supernatant of different cell lines after incubation with increasing doses of GYY4137 for 72 hours. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 1.

Increased H2S correlates with a decreased CD8+ T-cell/Treg ratio in colon cancer. A, Forty colon cancer tissues and paired normal colon epithelial tissues were collected, and a sulfide-selective electrode was utilized to detect the H2S levels in tissue homogenates. B, IHC analysis of CD8+ T-cell infiltration was performed in the 40 patients included in A, and a comparison of H2S levels between high CD8+T infiltration and low CD8+T infiltration tumors was performed. C, Immunofluorescence analysis of FOXP3+ CD4+ T-cell infiltration was performed in the 40 patients included in A, and a comparison of H2S levels between high Treg infiltration and low Treg infiltration tumors was performed. D, IHC analysis of SQR, TST, and ETHE1 was performed in a TMA composed of 20 colon cancer tissues and paired adjacent normal colon epithelial tissues. E, Western blot analysis of the expression of SQR, TST, and ETHE1 in NCM356, DLD-1, SW620, HT-29, HCT-116, HCT-8, Jurkat, and CD8+ T cells. GAPDH served as an internal control. F, Real-time PCR analysis of the expression of SQR, TST, and ETHE1, with GAPDH serving as an internal control. G, CCK-8 analysis of the effect of increasing doses of GYY4137 on the viability of HCT-116, SW620, NCM356, and CD8+ T cells. A total of 3000 cells were seeded per well and incubated with increasing doses of GYY4137 for 72 hours before collection for CCK-8 analysis. H, H2S levels in the supernatant of different cell lines after incubation with increasing doses of GYY4137 for 72 hours. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

CBS−/+ mice show an increased response to anti–PD-L1 therapy

Compared with those of WT mice, the levels of CBS and H2S in the colon and spleen of CBS−/+ mice (Supplementary Fig. S1A and S1B) were significantly downregulated (Supplementary Fig. S1C–S1E). Aminooxyacetic acid (AOAA) is a blocker of the cellular production of H2S. MC-38 cells were used to construct subcutaneous tumor models in CBS−/+ C57BL/6 mice, and no significant difference in tumor volume was observed between the anti–PD-L1 group and AOAA + anti–PD-L1 group (Fig. 2A). Besides, no significant difference in tumor volume was observed between the tumors inoculated in WT mice and CBS−/+ mice (Fig. 2B). The immune phenotype of these tumors was investigated utilizing polychromatic IHC analysis (Fig. 2C). No significant difference in CD8+ T-cell density was observed between tumors inoculated in WT and CBS−/+ mice (Fig. 2D). However, the enrichment of Tregs in tumors inoculated in CBS−/+ mice was significantly decreased compared with that in tumors inoculated in WT mice, and the CD8+ T-cell/Treg ratio was significantly increased in CBS−/+ mice (Fig. 2E and F). In addition, no difference in spleen weight was observed between the WT and CBS−/+ mice (Supplementary Fig. S1F). No significant difference in tumor growth was observed between the WT + IgG and CBS−/+ + IgG groups (Fig. 2G and H). Strikingly, significantly increased response to anti–PD-L1 therapy was observed in CBS−/+ mice. The tumor-free rates in the WT + anti–PD-L1 group and CBS−/+ + anti–PD-L1 group were 1/16 and 7/16, respectively (Fig. 2I and J). In addition, a significantly decreased tumor volume on day 25 after inoculation was observed in the CBS−/+ + anti–PD-L1 group compared with the WT+ anti–PD-L1 group (Fig. 2K). Prolonged survival was also evident in CBS−/+ + anti–PD-L1 mice compared with WT + anti–PD-L1 mice (Fig. 2L). Polychromatic IHC analysis (Fig. 3A) revealed significantly increased enrichment of CD8+ T cells in tumors from CBS−/+ + anti–PD-L1 mice (Fig. 3B). Decreased Treg infiltration was also evident in tumors collected from CBS−/+ + IgG mice compared with WT+ IgG mice (Fig. 3C). In addition, anti–PD-L1 significantly increased the CD8+ T-cell/Treg ratio in CBS−/+ mice compared with WT mice (Fig. 3D). Flow cytometry analysis indicated that anti–PD-L1 induced a significant increase in the infiltration of CD8+CD69+ T cells in CBS−/+ mice compared with WT mice (Fig. 3E). The results of Treg flow cytometry analysis was in accordance with the results of IHC analysis, depicting the decreased enrichment of Tregs in CBS−/+ + IgG mice compared with WT+ IgG mice (Fig. 3F). Online analysis of colon cancer TCGA data with GEIPIA 2.0 revealed a negative correlation between the expression of CBS and the effector Treg signature (FOXP3, CTLA4, CCR8, and TNFRSF9; Supplementary Fig. S2A). A positive correlation was also identified between the expression of CBS and the signature (HAVCR2, TIGIT, LAG3, PDCD1, CXCL13, and LAYN) of exhausted T cells (Supplementary Fig. S2A). Together, these results indicated that partial inhibition of CBS in a heterozygous mouse model induced increased responses to anti–PD-L1 therapy and that CBS might be involved in the activation of Tregs.

Figure 2.

CBS−/+ mice showed an increased response to anti–PD-L1 treatment. Subcutaneous tumors were constructed from MC-38 cells in WT and CBS−/+ mice generated with CRISPR-Cas9 in C57/BL 6 mice. A, Tumor growth curves from anti–PD-L1 and anti–PD-L1 + AOAA mice. B, Tumor growth curves from WT and CBS−/+ mice. C, Representative images from polychromatic IHC of tumors. D, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. E, Analysis of Treg infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. F, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. G, Tumor growth curves in WT mice that received IgG. H, Tumor growth curves in CBS−/+ mice that received IgG. I, Tumor growth curves in WT mice that received anti–PD-L1 antibodies. J, Tumor growth curves in CBS−/+ mice that received anti–PD-L1 antibodies. K, Volumes of tumors collected from the different groups of mice at day 25 after inoculation. L, Kaplan–Meier survival curves for WT and CBS−/+ mice that received IgG or anti–PD-L1 antibodies. Mice with tumors that reached 2,000 mm3 were euthanized, and tumor volume was recorded. **, P < 0.01 vs. WT + IgG; ##, P < 0.01 vs. WT + anti–PD-L1; ns, nonsignificant.

Figure 2.

CBS−/+ mice showed an increased response to anti–PD-L1 treatment. Subcutaneous tumors were constructed from MC-38 cells in WT and CBS−/+ mice generated with CRISPR-Cas9 in C57/BL 6 mice. A, Tumor growth curves from anti–PD-L1 and anti–PD-L1 + AOAA mice. B, Tumor growth curves from WT and CBS−/+ mice. C, Representative images from polychromatic IHC of tumors. D, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. E, Analysis of Treg infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. F, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. G, Tumor growth curves in WT mice that received IgG. H, Tumor growth curves in CBS−/+ mice that received IgG. I, Tumor growth curves in WT mice that received anti–PD-L1 antibodies. J, Tumor growth curves in CBS−/+ mice that received anti–PD-L1 antibodies. K, Volumes of tumors collected from the different groups of mice at day 25 after inoculation. L, Kaplan–Meier survival curves for WT and CBS−/+ mice that received IgG or anti–PD-L1 antibodies. Mice with tumors that reached 2,000 mm3 were euthanized, and tumor volume was recorded. **, P < 0.01 vs. WT + IgG; ##, P < 0.01 vs. WT + anti–PD-L1; ns, nonsignificant.

Close modal
Figure 3.

CBS deficiency elicited an activated immune microenvironment. A, Representative images from polychromatic IHC. B, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors. C, Analysis of Treg infiltration in polychromatic IHC images of tumors. D, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. E, Flow cytometry analysis of CD8+CD69+ T cells in tumors collected from different groups. F, Flow cytometry analysis of CD25+FOXP3+ Tregs in tumors collected from different groups. *, P < 0.05; **, P < 0.01 vs. WT + IgG; #, P < 0.05 vs. WT + anti–PD-L1.

Figure 3.

CBS deficiency elicited an activated immune microenvironment. A, Representative images from polychromatic IHC. B, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors. C, Analysis of Treg infiltration in polychromatic IHC images of tumors. D, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. E, Flow cytometry analysis of CD8+CD69+ T cells in tumors collected from different groups. F, Flow cytometry analysis of CD25+FOXP3+ Tregs in tumors collected from different groups. *, P < 0.05; **, P < 0.01 vs. WT + IgG; #, P < 0.05 vs. WT + anti–PD-L1.

Close modal

An SARD induced an increased response to ICIs in both MSI and MSS colon cancer models

Because both endogenous synthesis in tumor cells and bacterial sulfur reduction are sources of H2S in the colon cancer TME, we set out to investigate the feasibility of reducing H2S within colon cancer tumors by adopting a SARD (0.15% methionine and 0% cystine). Adopting a SARD for 2 weeks resulted in significantly decreased H2S levels in colon, spleen, and cecum orthotopic tumors (Supplementary Fig. S3A and S3B). First, we tested the effect of SARD on the therapeutic effect of anti–PD-L1 on immune-activated colon cancer tumor models constructed with MC-38 cells, an MSI colon cancer cell line derived from C57 BL/6 mice. An overview of the treatment protocol is illustrated in Fig. 4A. The results indicated that the SARD significantly increased responses to anti–PD-L1 in MC-38 cells (Fig. 4B). Representative magnetic resonance imaging (MRI) results are shown in Fig. 4C. Tumors collected on day 25 after inoculation are illustrated in Supplementary Fig. S3C. Polychromatic IHC analysis indicated that a SARD induced the increased infiltration of CD8+ T cells and decreased the infiltration of Tregs in orthotopic MC-38 tumors (Fig. 4DG). Second, we tested the effect of SARD on the therapeutic effect of combined anti–PD-L1 and anti–CTLA4 treatment in a highly aggressive colon cancer tumor model constructed with CT-26 cells, an MSS colon cancer cell line derived from BALB/c mice (Fig. 4H). Strikingly again, SARD significantly improved the MSS tumor response to combined treatment (Fig. 4I). The representative MRI results are shown in Fig. 4J. Polychromatic IHC analysis indicated that SARD induced increased infiltration of CD8+ T cells and decreased infiltration of Tregs in orthotopic CT-26 tumors (Fig. 4KN). In addition, immunotherapy did not affect H2S synthesis in the colon or spleen of the mice (Supplementary Fig. S3D). Besides, SARD did not affect the protein expression of PD-L1 in the orthotopic tumors but significantly downregulated the protein expression of CTLA4 in the spleen (Supplementary Fig. S3E).

Figure 4.

SARD promoted responses to ICIs in both MSI and MSS cecal orthotopic tumors. A, Treatment overview of C57/BL 6 mice inoculated with cecal orthotopic MC-38 tumors. B, Tumor growth curves for mice that were fed an ND or SARD with or without anti–PD-L1 antibody treatment. C, Representative MRI images of MC-38 tumors captured at day 25 after inoculation. D, Representative images of polychromatic IHC. E, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. F, Analysis of Treg infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. G, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. H, Treatment overview of BALB/c mice inoculated with cecal orthotopic CT-26 tumors. I, Tumor growth curves for mice fed an ND or SARD with or without combined anti–PD-L1 and anti–CTLA4 antibody treatment. J, Representative MRI images of CT-26 tumors captured at day 25 after inoculation. K, Representative images of polychromatic IHC. L, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. M, Analysis of Treg infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. N, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, nonsignificant.

Figure 4.

SARD promoted responses to ICIs in both MSI and MSS cecal orthotopic tumors. A, Treatment overview of C57/BL 6 mice inoculated with cecal orthotopic MC-38 tumors. B, Tumor growth curves for mice that were fed an ND or SARD with or without anti–PD-L1 antibody treatment. C, Representative MRI images of MC-38 tumors captured at day 25 after inoculation. D, Representative images of polychromatic IHC. E, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. F, Analysis of Treg infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. G, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. H, Treatment overview of BALB/c mice inoculated with cecal orthotopic CT-26 tumors. I, Tumor growth curves for mice fed an ND or SARD with or without combined anti–PD-L1 and anti–CTLA4 antibody treatment. J, Representative MRI images of CT-26 tumors captured at day 25 after inoculation. K, Representative images of polychromatic IHC. L, Analysis of CD8+ T-cell infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. M, Analysis of Treg infiltration in polychromatic IHC images of tumors collected from both WT and CBS−/+ mice. N, CD8+ T-cell/Treg ratios in tumors based on polychromatic IHC analysis. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, nonsignificant.

Close modal

SARD induced large-scale alterations in the tumor transcriptome

Tumor bulk sequencing was performed in cecal orthotopic CT-26 tumors collected from treatment-free mice that received a normal diet or SARD. The results indicated that SARD induced large-scale alterations in the tumor transcriptome, characterized by multiple differentially expressed genes (Fig. 5A). Notably, expression of Aak-1, a newly identified suppressor of T-cell migration, was significantly decreased by SARD. GSEA indicated that SARD increased the expression of chemokines and chemokine receptors, immune checkpoints and TME-associated gene signatures, and a more activated immune microenvironment was observed in tumors collected from the SARD-treated mice (Fig. 5B). Gene Ontology (GO) analysis indicated that peptidyl-cysteine S-nitrosylation was significantly decreased in the tumors of mice fed with a normal diet (ND), possibly related to the decreased tumor H2S levels in the SARD-fed mice (Fig. 5C). In addition, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis suggested that SARD decreased activation of the Wnt signaling pathway, which is a classic driver of proliferation and metastasis in colon cancer (Fig. 5D). CIBERSORT analysis based on bulk sequencing indicated that SARD induced the increased infiltration of multiple types of lymphocytes in tumors, including CD8+ T cells, activated CD4+ memory T cells, and resting mast cells (Fig. 5E). Besides, SARD also decreased the infiltration of resting dendritic cells (Supplementary Fig. S3F).

Figure 5.

SARD induced large-scale transcriptional alterations in orthotopic tumors. A, Top 100 differentially expressed genes in CT-26 tumors collected from mice fed an ND or SARD for 2 weeks. B, GSEA of the TME, chemokine, and chemokine receptors and immune checkpoints of CT-26 tumors. C, GO enrichment analysis of differentially expressed genes in CT-26 tumors. D, KEGG enrichment analysis of differentially expressed genes in CT-26 tumors. E, CIBERSORT analysis of tumor-infiltrating lymphocytes in CT-26 tumors. *, P < 0.05 vs. ND.

Figure 5.

SARD induced large-scale transcriptional alterations in orthotopic tumors. A, Top 100 differentially expressed genes in CT-26 tumors collected from mice fed an ND or SARD for 2 weeks. B, GSEA of the TME, chemokine, and chemokine receptors and immune checkpoints of CT-26 tumors. C, GO enrichment analysis of differentially expressed genes in CT-26 tumors. D, KEGG enrichment analysis of differentially expressed genes in CT-26 tumors. E, CIBERSORT analysis of tumor-infiltrating lymphocytes in CT-26 tumors. *, P < 0.05 vs. ND.

Close modal

H2S upregulates the differentiation of CD4+ T cells toward Tregs and inhibits the migration of CD8+ T cells

Based on these results, we investigated the regulatory effect of H2S on the activation of Tregs. Remarkable colocalization between FOXP3 and CBS was revealed by IF analysis of colon cancer tumor tissues (Supplementary Fig. S2B). Analysis based on sequencing data from CD4+ T cells (GSE59241) collected from WT and CBS−/− mice indicated that compared with that in WT mice, the mRNA expression of CD4 in CBS−/− mice was significantly downregulated (Supplementary Fig. S2C). In our study, the mRNA expression of CD4, CD25, and FOXP3 was significantly downregulated in the spleens of CBS−/+ mice (Supplementary Fig. S2D). GYY4137, a slow-releasing donor of H2S, dose dependently increased the activation of Tregs (Fig. 6A). Sulfur amino acid restriction (SAR) medium decreased the activation of Tregs, and this effect was significantly abrogated by GYY4137 (Fig. 6B). In addition, GYY4137 significantly upregulated the secretion of IFNγ in Tregs (Fig. 6C), whereas the expression of the proliferative marker Ki67 was not affected (Fig. 6D). GO enrichment analysis of transcriptome sequencing data indicated that H2S increased MHC class II receptor activity and the immune response in CD4+ T cells sorted from PBMCs (Fig. 6E). KEGG enrichment analysis indicated that H2S increased Th1 and Th2 cell differentiation (Fig. 6F). GSEA further validated that H2S promoted the differentiation of CD4+ T cells toward Th1, Th2, and Th17 cells and especially toward Tregs by activating multiple signature genes (Fig. 6G). Analysis based on sequencing data from CD4+ T cells (GSE59241) collected from both WT and CBS−/− mice also indicated that CBS knockout decreased the differentiation of CD4+ T cells toward Tregs (Supplementary Fig. S3G and S3H). We further investigated the effect of H2S on the migration of CD8+ T cells toward tumor beds by utilizing a coculture system composed of tumor spheres and human-derived CD8+ T cells. The results indicated that the SAR medium significantly increased the number of tumor-infiltrated CD8+ T cells, and this effect was inhibited by H2S (Fig. 6H). In addition, GYY4137 significantly enhanced the function of CD8+ T cells. The secretion of GZMA and PRF1 was significantly upregulated after GYY4137 treatment (Fig. 6I and J). These results revealed the cell type–dependent effect of SAR and H2S on immune cells, which might be the mechanism underlying the immunosuppressive effect of H2S in the colon cancer TME.

Figure 6.

H2S promoted Treg differentiation and inhibited CD8+ T-cell migration. A, Flow cytometry analysis of CD4+ T cells sorted from human PBMCs that were activated with anti-CD3 and anti-CD28 antibodies and treated with increasing doses of GYY4137 for 48 hours. B, Flow cytometry analysis of CD4+ T cells sorted from human PBMCs that were activated with anti-CD3 and anti-CD28 antibodies and incubated with SAR medium with or without 200 μmol/L GYY4137 for 48 hours. C and D, The influence of GYY4137 on Treg function (secretion of IFNγ) and proliferation (Ki67 expression). E, CD4+ T cells activated with anti-CD3 and anti-CD28 antibodies for 48 hours with or without 200 μmol/L GYY4137 treatment were collected for transcriptome sequencing, and GO enrichment analysis was subsequently performed. F, KEGG enrichment analysis of the sequencing results collected as described in C. G, GSEA of Th1, Th2, Treg, and Th17 signatures in CD4+ T cells activated with anti-CD3 and CD28 antibodies for 48 hours with or without 200 μmol/L GYY4137 treatment. H, Tumor spheres generated with HT-29 cell suspensions. CD8+ T cells sorted from PBMCs were activated with anti-CD3 and anti-CD28 antibodies for 48 hours, subsequently labeled with CFSE and added to tumor spheres incubated with control medium, SAR medium, or SAR medium with 200 μmol/L GYY4137. The spheres were fixed 48 hours after the addition of CD8+ T cells and stained with DAPI for the collection of IF images. I and J, The influence of GYY4137 on CD8+ T-cell function (secretion of GZMA and PRF1). *, P < 0.05 vs. control; **, P < 0.05 vs. control; #, P < 0.05 vs. SAR; ##, P < 0.01 vs. SAR; ***, P < 0.001; ns, nonsignificant.

Figure 6.

H2S promoted Treg differentiation and inhibited CD8+ T-cell migration. A, Flow cytometry analysis of CD4+ T cells sorted from human PBMCs that were activated with anti-CD3 and anti-CD28 antibodies and treated with increasing doses of GYY4137 for 48 hours. B, Flow cytometry analysis of CD4+ T cells sorted from human PBMCs that were activated with anti-CD3 and anti-CD28 antibodies and incubated with SAR medium with or without 200 μmol/L GYY4137 for 48 hours. C and D, The influence of GYY4137 on Treg function (secretion of IFNγ) and proliferation (Ki67 expression). E, CD4+ T cells activated with anti-CD3 and anti-CD28 antibodies for 48 hours with or without 200 μmol/L GYY4137 treatment were collected for transcriptome sequencing, and GO enrichment analysis was subsequently performed. F, KEGG enrichment analysis of the sequencing results collected as described in C. G, GSEA of Th1, Th2, Treg, and Th17 signatures in CD4+ T cells activated with anti-CD3 and CD28 antibodies for 48 hours with or without 200 μmol/L GYY4137 treatment. H, Tumor spheres generated with HT-29 cell suspensions. CD8+ T cells sorted from PBMCs were activated with anti-CD3 and anti-CD28 antibodies for 48 hours, subsequently labeled with CFSE and added to tumor spheres incubated with control medium, SAR medium, or SAR medium with 200 μmol/L GYY4137. The spheres were fixed 48 hours after the addition of CD8+ T cells and stained with DAPI for the collection of IF images. I and J, The influence of GYY4137 on CD8+ T-cell function (secretion of GZMA and PRF1). *, P < 0.05 vs. control; **, P < 0.05 vs. control; #, P < 0.05 vs. SAR; ##, P < 0.01 vs. SAR; ***, P < 0.001; ns, nonsignificant.

Close modal

H2S promotes the activation of Tregs by inducing ENO1 persulfidation at cysteine 119

Proteins with altered persulfidation were investigated in Tregs after incubation with SAR medium with or without GYY4137 utilizing a modified biotin switch assay and liquid chromatography with tandem mass spectrometry (Fig. 7A). α-Enolase (ENO1) was identified as among the proteins whose persulfidation was altered to the greatest extent (Fig. 7B). The modified biotin switch assay further confirmed that the SAR medium significantly decreased the persulfidation of ENO1, and this effect could be ameliorated by GYY4137 (Fig. 7C). DTT was added to rule out nonspecific labeling to indicate persulfidation (Fig. 7D). ENO1 has been validated as a master regulator of Treg metabolism and development through its ability to upregulate FOXP3 at the transcriptional level (44). Six cysteine residues are contained in the amino acid sequence of ENO1, and we screened the functional cysteine residue responsible for the persulfidation-mediated effect by sequential point mutations to alanine. The results suggested that mutating one cysteine residue had no significant effect on the decreased persulfidation status of ENO1 induced by SAR, indicating that ENO1 might be persulfidated at multiple cysteine residues (Fig. 7E). However, ChIP assays with the FOXP3 promoter indicated that the C119A mutation abrogated the increased binding of ENO1 with the FOXP3 promoter induced by GYY4137 (Fig. 7F). Besides, we investigated the effect of ENO1 C119A on the function of Tregs induced by GYY4137. Suppression assays indicated that C119A abrogated the increased inhibitory effect of Tregs induced by GYY4137 on the proliferation of CD8+ T cells (Fig. 7G). ENO1 C119A expressing plasmids also abolished the increased expression of Treg function genes (IFNγ, FOXP3, and IL10) induced by GYY4137 (Fig. 7H). Together, these results indicated that H2S could promote the activation of Tregs by inducing ENO1 persulfidation at cysteine 119.

Figure 7.

H2S promoted the activation of Tregs by inducing ENO1 persulfidation at cysteine 119. A, Jurkat cells were incubated with SAR medium with or without 200 μmol/L GYY4137 for 48 hours before modified biotin switch assays. Bands in the respective lanes in SDS-PAGE gels were collected for mass spectrometry. B, Volcano plots showing the results of mass spectrometry analysis of proteins with an altered persulfidation status. Top, PBS-treated control vs. SAR; bottom, PBS-treated control vs. SAR+GYY. C, Modified biotin switch assays showing the decreased persulfidation of ENO1 after incubation with SAR medium, which was ameliorated by GYY4137. D, DTT-treated samples served as negative controls. E, Modified biotin switch assays showing that the expression plasmids had no effect on persulfidation mediated by GYY4137 in Jurkat cells. DTT-treated samples served as negative controls. F, ChIP assays show that C119A expression plasmids abrogated the increased binding of ENO1 with the FOXP3 promoter induced by GYY4137. WT, C337A, C357A, C389A, and C399A expression plasmids had no significant effect on the increased binding of ENO1 with the FOXP3 promoter induced by GYY4137. G, Flow cytometry analysis showing that C119A expression plasmids abolished the inhibitory effect of Tregs on the proliferation of CD8+T cells elicited by GYY4137. H, mRNA levels of Treg function genes (IFNγ, FOXP3, and IL10) indicated that C119A-expressing plasmids ameliorated the increased expression of these genes induced by GYY4137 in Tregs. **, P < 0.01 vs. control; ***, P < 0.001 vs. control; #, P < 0.05 vs. WT; ##, P < 0.01 vs. WT; ###, P < 0.001 vs. WT; ns, nonsignificant.

Figure 7.

H2S promoted the activation of Tregs by inducing ENO1 persulfidation at cysteine 119. A, Jurkat cells were incubated with SAR medium with or without 200 μmol/L GYY4137 for 48 hours before modified biotin switch assays. Bands in the respective lanes in SDS-PAGE gels were collected for mass spectrometry. B, Volcano plots showing the results of mass spectrometry analysis of proteins with an altered persulfidation status. Top, PBS-treated control vs. SAR; bottom, PBS-treated control vs. SAR+GYY. C, Modified biotin switch assays showing the decreased persulfidation of ENO1 after incubation with SAR medium, which was ameliorated by GYY4137. D, DTT-treated samples served as negative controls. E, Modified biotin switch assays showing that the expression plasmids had no effect on persulfidation mediated by GYY4137 in Jurkat cells. DTT-treated samples served as negative controls. F, ChIP assays show that C119A expression plasmids abrogated the increased binding of ENO1 with the FOXP3 promoter induced by GYY4137. WT, C337A, C357A, C389A, and C399A expression plasmids had no significant effect on the increased binding of ENO1 with the FOXP3 promoter induced by GYY4137. G, Flow cytometry analysis showing that C119A expression plasmids abolished the inhibitory effect of Tregs on the proliferation of CD8+T cells elicited by GYY4137. H, mRNA levels of Treg function genes (IFNγ, FOXP3, and IL10) indicated that C119A-expressing plasmids ameliorated the increased expression of these genes induced by GYY4137 in Tregs. **, P < 0.01 vs. control; ***, P < 0.001 vs. control; #, P < 0.05 vs. WT; ##, P < 0.01 vs. WT; ###, P < 0.001 vs. WT; ns, nonsignificant.

Close modal

SAR inhibited the expression of AAK-1 by decreasing ELK4 persulfidation at cysteine 25

The above-mentioned results in Fig. 6F show that SAR induced the increased infiltration of CD8+ T cells in tumor spheres, which could be ameliorated by H2S. Recently, AAK-1 was validated as a suppressor of T-cell migration in a large-scale Sleeping Beauty screen (47). Sequencing of bulk tumors collected from SARD-fed and ND-fed mice indicated that the SARD significantly decreased the expression of AAK-1, which was also validated at the mRNA and protein levels (Fig. 8AC). We screened the promoter region of AAK-1, and two potential binding sites for ELK4 were identified (Fig. 8D). Modified biotin switch assays indicated that the SAR medium significantly decreased the persulfidation of ELK4 in CD8+ T cells, which could be ameliorated by H2S (Fig. 8E). We further performed point mutation of both cysteine residues in ELK4, and a single mutation could not abolish the increased persulfidation induced by H2S (Fig. 8F). Intriguingly, the C25A mutation largely abrogated the increased expression of AAK-1 induced by H2S (Fig. 8G). ChIP assays indicated that H2S significantly increased the binding of ELK4 with AAK-1 promoter regions, which was abolished to some extent by the C25A mutation in ELK4 (Fig. 8H). Together, these results indicated that SAR could promote the migration of T cells by inhibiting the expression of AAK-1 through decreasing ELK4 persulfidation at C25.

Figure 8.

A SARD promoted the migration of CD8+ T cells by upregulating the expression of AAK-1 by decreasing ELK4 persulfidation at cysteine 25. A, Volcano plot of differentially expressed genes in CT-26 tumors collected from mice that were fed an ND or SARD for 2 weeks. AAK-1 was among the genes with decreased expression in the tumors from SARD-fed mice. B, Real-time PCR results showing AAK-1 mRNA levels in tumors from mice fed an ND or SARD. C, Western blotting results of AAK-1 protein levels in tumors from mice fed an ND or SARD. D, Two ELK4-binding sites in the promoter region of AAK-1 were predicted by the JASPAR database. E, Modified biotin switch assays showing the decreased persulfidation of ELK4 after incubation with SAR medium, which was ameliorated by GYY4137. F, Modified biotin switch assays showing that the expressing plasmids had no effect on persulfidation mediated by GYY4137 in Jurkat cells. DTT-treated samples served as negative controls. G, ELK4 C25A expression plasmids attenuated the increased expression of AAK-1 induced by GYY4137. H, ChIP assays showing that ELK4 C25A expression plasmids attenuated the increased binding of ELK4 with the AAK-1 promoter induced by GYY4137. *, P < 0.05; **, P < 0.01.

Figure 8.

A SARD promoted the migration of CD8+ T cells by upregulating the expression of AAK-1 by decreasing ELK4 persulfidation at cysteine 25. A, Volcano plot of differentially expressed genes in CT-26 tumors collected from mice that were fed an ND or SARD for 2 weeks. AAK-1 was among the genes with decreased expression in the tumors from SARD-fed mice. B, Real-time PCR results showing AAK-1 mRNA levels in tumors from mice fed an ND or SARD. C, Western blotting results of AAK-1 protein levels in tumors from mice fed an ND or SARD. D, Two ELK4-binding sites in the promoter region of AAK-1 were predicted by the JASPAR database. E, Modified biotin switch assays showing the decreased persulfidation of ELK4 after incubation with SAR medium, which was ameliorated by GYY4137. F, Modified biotin switch assays showing that the expressing plasmids had no effect on persulfidation mediated by GYY4137 in Jurkat cells. DTT-treated samples served as negative controls. G, ELK4 C25A expression plasmids attenuated the increased expression of AAK-1 induced by GYY4137. H, ChIP assays showing that ELK4 C25A expression plasmids attenuated the increased binding of ELK4 with the AAK-1 promoter induced by GYY4137. *, P < 0.05; **, P < 0.01.

Close modal

Revealing molecules that control Treg differentiation and function is a promising approach to enhance antitumor immunity and improve the responses to ICIs in patients with advanced-stage colon cancer. Previous studies have confirmed that H2S-derived CBS promotes energy synthesis, cell proliferation, and angiogenesis in colon cancer (27). In addition, inhibition of the CBS–H2S axis can enhance the chemosensitivity (5-FU and oxaliplatin) of colon cancer cells (29, 48). An increasing number of studies have shown that H2S is an endogenous regulator of the immune system (49). Through sulfhydrating LKB1 or promoting TET1- and TET2-mediated FOXP3 demethylation, H2S promotes regulatory T-cell differentiation and proliferation and maintains human immune homeostasis (18,19). However, the specific role of H2S in the immunotherapy of colon cancer remains to be illustrated.

In our study, both in vitro and in vivo experiments confirmed that the endogenous CBS–H2S axis promoted the differentiation of Tregs, and CBS deficiency could increase the efficacy of anti–PD-L1 treatment in a colon cancer mouse model. In vivo experiments also confirmed that exogenous H2S promoted the differentiation of Tregs, and H2S depletion through SARD increased the efficacy of anti–PD-L1 treatment in an orthotopic transplantation mouse model of colon cancer.

Previous studies have shown that the SOP is responsible for the catabolism of H2S (50). In our study, we found that the SOP might shield colon cancer cells from increased H2S derived from luminal bacterial activity in colon cancer patients. Interestingly, the increased expression of SOP enzymes (SQR, TST, and ETHE1) in colon cancer corresponds to the proliferation-promoting effect of H2S, whereas their expression was significantly decreased in normal colon epithelial cells and CD8+ T cells, corresponding to the growth-inhibiting effect of H2S in these cells. We concluded that in addition to endogenous H2S-producing enzymes (CBS, CSE, MPST), these SOP enzymes may also serve as potential targets for colon cancer therapy.

Our results showed that CBS depletion increased the CD8+ T-cell/Treg ratio and the response to anti–PD-L1 therapy. In contrast, Ying-Hongshi and colleagues proved that in hepatocellular carcinoma (HCC), activation of the CBS–H2S axis could reduce the abundance of Tregs, and CBS deficiency promoted immune evasion and tumor growth in CBS heterozygous knockout mice (51). One possible reason for this discrepancy might be that the expression of CBS is downregulated in HCC tissues and upregulated in colon cancer tissues compared with corresponding normal tissues. In addition, in colon cancer tissues, H2S is produced both endogenously in tumor cells and exogenously in luminal bacteria (52,53), whereas in HCC tissues, H2S is mainly produced endogenously.

The molecular mechanisms by which H2S promotes tumor growth in the colon cancer TME have remained elusive. The latest research shows that H2S activates intracellular signaling pathways through the persulfidation of specific cysteine residues in target proteins, which functions in a manner analogous to nitrosylation (54). Specifically, this modification occurs when sulfur in H2S attaches to the cysteine residue (-SH) in proteins and converts it to -SSH (55). In our study, we concluded that H2S could upregulate FOXP3 expression and promote the differentiation of Tregs through sulfhydrating ENO1. ENO1 functions both as a glycolytic enzyme and as a master regulator of Treg differentiation. Silencing ENO1 in the absence of treatment with 2DG (a glycolytic inhibitor) led to a reduction in FOXP3 expression, whereas silencing ENO1 restored FOXP3 mRNA expression in 2DG-treated cells (44). In addition, alternative splicing of ENO1 results in a shorter isoform that has been shown to bind the c-myc promoter and function as a tumor suppressor (56). Previous studies have indicated that the activation of Th1 immune responses promotes cytotoxic immune cell function. However, our data indicated that H2S increased Th1 cell differentiation in CD4+ T cells and inhibited the function of CD8+ T cells, which might be related to the cell-specific functional persulfidation sites in Tregs and CD8+ T cells, respectively.

CD8+ cytotoxic T cells are the major effectors of the anticancer immune response and the mainstay of current successful cancer immunotherapy (57). We found that H2S could upregulate AAK-1 expression and inhibit the migration of CD8+ T cells by sulfhydrating ETS transcription factor ELK4. A previous study showed that AAK-1 localizes to the leading edge of migrating HeLa cells, suggesting that AAK-1 may regulate the endocytosis of chemokine receptors (58). A small-molecule inhibitor of AAK-1 increased the expression of CXCR3 on activated T cells and promoted T-cell migration toward the chemokine CXCL10, which is frequently produced in the TME (47, 59). Our research further clarified the close correlation between AAK-1 and the immune microenvironment in colon cancer.

Although our study illustrates the potential value of restricting luminal-driven H2S as a novel strategy of activating antitumor immunity in colon cancer, it should be noted that immunotherapy has been used in a limited number of patients with metastatic MSI-high tumors; therefore, the efficacy of this strategy in metastatic tumors requires further investigation. Nevertheless, the recently published striking effect of immunotherapy in achieving clinical complete remission in locally advanced rectal cancer with an MSI-high genotype sheds light on the adoption of immunotherapy in an earlier stage in colorectal cancer (60). Studies including our previous work have revealed the vital role of CBS overexpression in promoting proliferation and metastasis in colon cancer cells; however, the data in this study based on CBS−/+ mice and the SAR diet indicated that the elevated H2S levels caused by luminal sulfur amino acid metabolism exerted an immunosuppressive effect in colorectal cancer and that CBS activity in tumor cells may not be relevant for immunotherapy.

In summary, H2S depletion could create a favorable immune microenvironment in colon cancer by decreasing the persulfidation of ENO1 in Tregs and ELK4 in CD8+ T cells. Our results further indicate that a SARD may be exploited as a novel approach to promote responses to immunotherapies in both MSI and MSS colon cancer.

No disclosures were reported.

T. Yue: Resources, data curation, formal analysis, validation, investigation, methodology, writing–review and editing. J. Li: Resources, data curation, software, formal analysis, methodology, writing–original draft, writing–review and editing. J. Zhu: Supervision, validation, investigation, project administration. S. Zuo: Supervision, validation, investigation, project administration, writing–review and editing. X. Wang: Supervision, funding acquisition, project administration, writing–review and editing. Y. Liu: Data curation, supervision, funding acquisition, methodology, writing–original draft, project administration, writing–review and editing. J. Liu: Data curation, supervision, funding acquisition, methodology, writing–original draft, project administration, writing–review and editing. X. Liu: Data curation, supervision, funding acquisition, methodology, writing–original draft, project administration, writing–review and editing. P. Wang: Data curation, supervision, funding acquisition, methodology, writing–original draft, project administration, writing–review and editing. S. Chen: Data curation, supervision, funding acquisition, methodology, writing–original draft, project administration, writing–review and editing.

This study was supported by grants from the National Natural Science Foundation of China (No. 81902384). The authors wish to thank Professor Dingfang Bu from the Central Laboratory in Peking University First Hospital for her assistance in the preparation of the manuscript as well as the healthy donors and patients enrolled in our study.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

1.
Siegel
RL
,
Miller
KD
,
Fuchs
HE
,
Jemal
A
.
Cancer statistics, 2021
.
CA Cancer J Clin
2021
;
71
:
7
33
.
2.
Zaborowski
AM
,
Winter
DC
,
Lynch
L
.
The therapeutic and prognostic implications of immunobiology in colorectal cancer: a review
.
Br J Cancer
2021
;
125
:
1341
9
.
3.
Liu
C
,
Liu
R
,
Wang
B
,
Lian
J
,
Yao
Y
,
Sun
H
, et al
.
Blocking IL-17A enhances tumor response to anti-PD-1 immunotherapy in microsatellite stable colorectal cancer
.
J Immunother Cancer
2021
;
9
:
e001895
.
4.
Zhang
B
,
Yao
K
,
Zhou
E
,
Zhang
L
,
Cheng
C
.
Chr20q amplification defines a distinct molecular subtype of microsatellite stable colorectal cancer
.
Cancer Res
2021
;
81
:
1977
87
.
5.
Ghonim
MA
,
Ibba
SV
,
Tarhuni
AF
,
Errami
Y
,
Luu
HH
,
Dean
MJ
, et al
.
Targeting PARP-1 with metronomic therapy modulates MDSC suppressive function and enhances anti-PD-1 immunotherapy in colon cancer
.
J Immunother Cancer
2021
;
9
:
e001643
.
6.
Ganesh
K
,
Stadler
ZK
,
Cercek
A
,
Mendelsohn
RB
,
Shia
J
,
Segal
NH
, et al
.
Immunotherapy in colorectal cancer: rationale, challenges and potential
.
Nat Rev Gastroenterol Hepatol
2019
;
16
:
361
75
.
7.
Schmitt
M
,
Greten
FR
.
The inflammatory pathogenesis of colorectal cancer
.
Nat Rev Immunol
2021
;
21
:
653
67
.
8.
Tan
AC
,
Bagley
SJ
,
Wen
PY
,
Lim
M
,
Platten
M
,
Colman
H
, et al
.
Systematic review of combinations of targeted or immunotherapy in advanced solid tumors
.
J Immunother Cancer
2021
;
9
:
e002459
.
9.
Nishikawa
H
,
Koyama
S
.
Mechanisms of regulatory T cell infiltration in tumors: implications for innovative immune precision therapies
.
J Immunother Cancer
2021
;
9
:
e002591
.
10.
Gerard
CL
,
Delyon
J
,
Wicky
A
,
Homicsko
K
,
Cuendet
MA
,
Michielin
O
.
Turning tumors from cold to inflamed to improve immunotherapy response
.
Cancer Treat Rev
2021
;
101
:
102227
.
11.
Lu
Y
,
Li
Y
,
Liu
Q
,
Tian
N
,
Du
P
,
Zhu
F
, et al
.
MondoA-thioredoxin-interacting protein axis maintains regulatory T-cell identity and function in colorectal cancer microenvironment
.
Gastroenterology
2021
;
161
:
575
91
.
12.
Kroemer
M
,
Turco
C
,
Spehner
L
,
Viot
J
,
Idirène
I
,
Bouard
A
, et al
.
Investigation of the prognostic value of CD4 T cell subsets expanded from tumor-infiltrating lymphocytes of colorectal cancer liver metastases
.
J Immunother Cancer
2020
;
8
:
e001478
.
13.
Adamczyk
A
,
Pastille
E
,
Kehrmann
J
,
Vu
VP
,
Geffers
R
,
Wasmer
MH
, et al
.
GPR15 facilitates recruitment of regulatory T cells to promote colorectal cancer
.
Cancer Res
2021
;
81
:
2970
82
.
14.
Villarreal
DO
,
L'Huillier
A
,
Mottershead
C
,
Filippova
EV
,
Coder
BD
, et al
.
Targeting CCR8 induces protective antitumor immunity and enhances vaccine-induced responses in colon cancer
.
Cancer Res
2018
;
78
:
5340
8
.
15.
Watson
MJ
,
Vignali
PDA
,
Mullett
SJ
,
Overacre-Delgoffe
AE
,
Peralta
RM
,
Grebinoski
S
, et al
.
Metabolic support of tumour-infiltrating regulatory T cells by lactic acid
.
Nature
2021
;
591
:
645
51
.
16.
Wang
H
,
Franco
F
,
Tsui
YC
,
Xie
X
,
Trefny
MP
,
Zappasodi
R
, et al
.
CD36-mediated metabolic adaptation supports regulatory T cell survival and function in tumors
.
Nat Immunol
2020
;
21
:
298
308
.
17.
Simpson
TR
,
Li
F
,
Montalvo-Ortiz
W
,
Sepulveda
MA
,
Bergerhoff
K
,
Arce
F
, et al
.
Fc-dependent depletion of tumor-infiltrating regulatory T cells co-defines the efficacy of anti-CTLA-4 therapy against melanoma
.
J Exp Med
2013
;
210
:
1695
710
.
18.
Cui
C
,
Fan
J
,
Zeng
Q
,
Cai
J
,
Chen
Y
,
Chen
Z
, et al
.
CD4(+) T-cell endogenous cystathionine γ lyase-hydrogen sulfide attenuates hypertension by sulfhydrating liver kinase B1 to promote T regulatory cell differentiation and proliferation
.
Circulation
2020
;
142
:
1752
69
.
19.
Yang
R
,
Qu
C
,
Zhou
Y
,
Konkel
JE
,
Shi
S
,
Liu
Y
, et al
.
Hydrogen sulfide promotes Tet1- and Tet2-mediated FOXP3 demethylation to drive regulatory T cell differentiation and maintain immune homeostasis
.
Immunity
2015
;
43
:
251
63
.
20.
Nguyen
LH
,
Cao
Y
,
Hur
J
,
Mehta
RS
,
Sikavi
DR
,
Wang
Y
, et al
.
The sulfur microbial diet is associated with increased risk of early-onset colorectal cancer precursors
.
Gastroenterology
2021
;
161
:
1423
32
.
21.
Sikavi
DR
,
Nguyen
LH
,
Haruki
K
,
Ugai
T
,
Ma
W
,
Wang
DD
, et al
.
The sulfur microbial diet and risk of colorectal cancer by molecular subtypes and intratumoral microbial species in adult men
.
Clin Translat Gastroenterol
2021
;
12
:
e00338
.
22.
Green
CL
,
Lamming
DW
,
Fontana
L
.
Molecular mechanisms of dietary restriction promoting health and longevity
.
Nat Rev Mol Cell Biol
2022
;
23
:
56
73
.
23.
Roy
DG
,
Chen
J
,
Mamane
V
,
Ma
EH
,
Muhire
BM
,
Sheldon
RD
, et al
.
Methionine metabolism shapes T helper cell responses through regulation of epigenetic reprogramming
.
Cell Metab
2020
;
31
:
250
66
.
24.
Cramer
SL
,
Saha
A
,
Liu
J
,
Tadi
S
,
Tiziani
S
,
Yan
W
, et al
.
Systemic depletion of L-cyst(e)ine with cyst(e)inase increases reactive oxygen species and suppresses tumor growth
.
Nat Med
2017
;
23
:
120
7
.
25.
Badgley
MA
,
Kremer
DM
,
Maurer
HC
,
DelGiorno
KE
,
Lee
HJ
,
Purohit
V
, et al
.
Cysteine depletion induces pancreatic tumor ferroptosis in mice
.
Science
2020
;
368
:
85
9
.
26.
Hine
C
,
Harputlugil
E
,
Zhang
Y
,
Ruckenstuhl
C
,
Lee
BC
,
Brace
L
, et al
.
Endogenous hydrogen sulfide production is essential for dietary restriction benefits
.
Cell
2015
;
160
:
132
44
.
27.
Szabo
C
,
Coletta
C
,
Chao
C
,
Módis
K
,
Szczesny
B
,
Papapetropoulos
A
, et al
.
Tumor-derived hydrogen sulfide, produced by cystathionine-β-synthase, stimulates bioenergetics, cell proliferation, and angiogenesis in colon cancer
.
Proc Natl Acad Sci U S A
2013
;
110
:
12474
9
.
28.
Chen
S
,
Zuo
S
,
Zhu
J
,
Yue
T
,
Bu
D
,
Wang
X
, et al
.
Decreased expression of cystathionine β-synthase exacerbates intestinal barrier injury in ulcerative colitis
.
J Crohns Colitis
2019
;
13
:
1067
80
.
29.
Chen
S
,
Yue
T
,
Huang
Z
,
Zhu
J
,
Bu
D
,
Wang
X
, et al
.
Inhibition of hydrogen sulfide synthesis reverses acquired resistance to 5-FU through miR-215–5p-EREG/TYMS axis in colon cancer cells
.
Cancer Lett
2019
;
466
:
49
60
.
30.
Chen
S
,
Bu
D
,
Zhu
J
,
Yue
T
,
Guo
S
,
Wang
X
, et al
.
Endogenous hydrogen sulfide regulates xCT stability through persulfidation of OTUB1 at cysteine 91 in colon cancer cells
.
Neoplasia
2021
;
23
:
461
72
.
31.
Geng
B
,
Du
JB
,
Tang
CS
. [
Application of sensitive sulphur electrode assay to measure and analyze cystathionine-gamma-lyase/hydrogen sulfide pathway of cardiovascular tissues, cells and plasma in rats]
.
Beijing Da Xue Xue Bao Yi Xue Ban
2005
;
37
:
545
8
.
32.
Hoffmann
P
,
Eder
R
,
Boeld
TJ
,
Doser
K
,
Piseshka
B
,
Andreesen
R
, et al
.
Only the CD45RA+ subpopulation of CD4+CD25high T cells gives rise to homogeneous regulatory T-cell lines upon in vitro expansion
.
Blood
2006
;
108
:
4260
7
.
33.
Naito
S
,
von Eschenbach
AC
,
Giavazzi
R
,
Fidler
IJ
.
Growth and metastasis of tumor cells isolated from a human renal cell carcinoma implanted into different organs of nude mice
.
Cancer Res
1986
;
46
:
4109
15
.
34.
Ahern
E
,
Harjunpää
H
,
O'Donnell
JS
,
Allen
S
,
Dougall
WC
,
Teng
MWL
, et al
.
RANKL blockade improves efficacy of PD1-PD-L1 blockade or dual PD1-PD-L1 and CTLA4 blockade in mouse models of cancer
.
Oncoimmunology
2018
;
7
:
e1431088
.
35.
Bohn
T
,
Rapp
S
,
Luther
N
,
Klein
M
,
Bruehl
TJ
,
Kojima
N
, et al
.
Tumor immunoevasion via acidosis-dependent induction of regulatory tumor-associated macrophages
.
Nat Immunol
2018
;
19
:
1319
29
.
36.
Onda
M
,
Kobayashi
K
,
Pastan
I
.
Depletion of regulatory T cells in tumors with an anti-CD25 immunotoxin induces CD8 T cell-mediated systemic antitumor immunity
.
Proc Nat Acad Sci USA
2019
;
116
:
4575
82
.
37.
Krechetova
LV
,
Vanko
LV
,
Vtorushina
VV
,
Nikolaeva
MA
,
Inviyaeva
EV
,
Tetruashvili
NK
.
Lymphocyte activation in the development of immune tolerance in women with recurrent pregnancy loss
.
Biochemistry Biokhimiia
2020
;
85
:
583
93
.
38.
Lazarus
J
,
Maj
T
,
Smith
JJ
,
Perusina Lanfranca
M
,
Rao
A
,
D'Angelica
MI
, et al
.
Spatial and phenotypic immune profiling of metastatic colon cancer
.
JCI Insight
2018
;
3
:
e121932
.
39.
Sun
Y
,
Wu
L
,
Zhong
Y
,
Zhou
K
,
Hou
Y
,
Wang
Z
, et al
.
Single-cell landscape of the ecosystem in early-relapse hepatocellular carcinoma
.
Cell
2021
;
184
:
404
21
.
40.
Jin
S
,
Li
R
,
Chen
MY
,
Yu
C
,
Tang
LQ
,
Liu
YM
, et al
.
Single-cell transcriptomic analysis defines the interplay between tumor cells, viral infection, and the microenvironment in nasopharyngeal carcinoma
.
Cell Res
2020
;
30
:
950
65
.
41.
Fiegle
E
,
Doleschel
D
,
Koletnik
S
,
Rix
A
,
Weiskirchen
R
,
Borkham-Kamphorst
E
, et al
.
Dual CTLA-4 and PD-L1 blockade inhibits tumor growth and liver metastasis in a highly aggressive orthotopic mouse model of colon cancer
.
Neoplasia
2019
;
21
:
932
44
.
42.
Courau
T
,
Bonnereau
J
,
Chicoteau
J
,
Bottois
H
,
Remark
R
,
Assante Miranda
L
, et al
.
Cocultures of human colorectal tumor spheroids with immune cells reveal the therapeutic potential of MICA/B and NKG2A targeting for cancer treatment
.
J Immunother Cancer
2019
;
7
:
74
.
43.
Remoli
AL
,
Marsili
G
,
Perrotti
E
,
Gallerani
E
,
Ilari
R
,
Nappi
F
, et al
.
Intracellular HIV-1 Tat protein represses constitutive LMP2 transcription increasing proteasome activity by interfering with the binding of IRF-1 to STAT1
.
Biochem J
2006
;
396
:
371
80
.
44.
De Rosa
V
,
Galgani
M
,
Porcellini
A
,
Colamatteo
A
,
Santopaolo
M
,
Zuchegna
C
, et al
.
Glycolysis controls the induction of human regulatory T cells by modulating the expression of FOXP3 exon 2 splicing variants
.
Nat Immunol
2015
;
16
:
1174
84
.
45.
Covello
KL
,
Kehler
J
,
Yu
H
,
Gordan
JD
,
Arsham
AM
,
Hu
CJ
, et al
.
HIF-2alpha regulates Oct-4: effects of hypoxia on stem cell function, embryonic development, and tumor growth
.
Genes Dev
2006
;
20
:
557
70
.
46.
Dey
A
,
Prabhudesai
S
,
Zhang
Y
,
Rao
G
,
Thirugnanam
K
,
Hossen
MN
, et al
.
Cystathione β-synthase regulates HIF-1α stability through persulfidation of PHD2
.
Sci Adv
2020
;
6
:
eaaz8534
.
47.
Rogers
LM
,
Wang
Z
,
Mott
SL
,
Dupuy
AJ
,
Weiner
GJ
.
A genetic screen to identify gain- and loss-of-function modifications that enhance T-cell infiltration into tumors
.
Cancer Immunol Res
2020
;
8
:
1206
14
.
48.
Yue
T
,
Zuo
S
,
Bu
D
,
Zhu
J
,
Chen
S
,
Ma
Y
, et al
.
Aminooxyacetic acid (AOAA) sensitizes colon cancer cells to oxaliplatin via exaggerating apoptosis induced by ROS
.
J Cancer
2020
;
11
:
1828
38
.
49.
Dilek
N
,
Papapetropoulos
A
,
Toliver-Kinsky
T
,
Szabo
C
.
Hydrogen sulfide: an endogenous regulator of the immune system
.
Pharmacol Res
2020
;
161
:
105119
.
50.
Hildebrandt
TM
,
Grieshaber
MK
.
Three enzymatic activities catalyze the oxidation of sulfide to thiosulfate in mammalian and invertebrate mitochondria
.
FEBS J
2008
;
275
:
3352
61
.
51.
Zhou
YF
,
Song
SS
,
Tian
MX
,
Tang
Z
,
Wang
H
,
Fang
Y
, et al
.
Cystathionine beta-synthase mediated PRRX2/IL-6/STAT3 inactivation suppresses Tregs infiltration and induces apoptosis to inhibit HCC carcinogenesis
.
J Immunother Cancer
2021
;
9
:
e003031
.
52.
Peck
SC
,
Denger
K
,
Burrichter
A
,
Irwin
SM
,
Balskus
EP
,
Schleheck
D
.
A glycyl radical enzyme enables hydrogen sulfide production by the human intestinal bacterium Bilophila wadsworthia
.
Proc Nat Acad Sci USA
2019
;
116
:
3171
6
.
53.
Xiao
A
,
Liu
C
,
Li
J
.
The role of H(2)S in the gastrointestinal tract and microbiota
.
Adv Exp Med Biol
2021
;
1315
:
67
98
.
54.
Stamler
JS
,
Simon
DI
,
Osborne
JA
,
Mullins
ME
,
Jaraki
O
,
Michel
T
, et al
.
S-nitrosylation of proteins with nitric oxide: synthesis and characterization of biologically active compounds
.
Proc Nat Acad Sci USA
1992
;
89
:
444
8
.
55.
Mustafa
AK
,
Gadalla
MM
,
Sen
N
,
Kim
S
,
Mu
W
,
Gazi
SK
, et al
.
H2S signals through protein S-sulfhydration
.
Sci Signal
2009
;
2
:
ra72
.
56.
Cancemi
P
,
Buttacavoli
M
,
Roz
E
,
Feo
S
.
Expression of alpha-enolase (ENO1), myc promoter-binding protein-1 (MBP-1) and matrix metalloproteinases (MMP-2 and MMP-9) reflect the nature and aggressiveness of breast tumors
.
Int J Mol Sci
2019
;
20
:
3952
.
57.
Raskov
H
,
Orhan
A
,
Christensen
JP
,
Gögenur
I
.
Cytotoxic CD8(+) T cells in cancer and cancer immunotherapy
.
Br J Cancer
2021
;
124
:
359
67
.
58.
Conner
SD
,
Schmid
SL
.
Identification of an adaptor-associated kinase, AAK1, as a regulator of clathrin-mediated endocytosis
.
J Cell Biol
2002
;
156
:
921
9
.
59.
Nagarsheth
N
,
Wicha
MS
,
Zou
W
.
Chemokines in the cancer microenvironment and their relevance in cancer immunotherapy
.
Nat Rev Immunol
2017
;
17
:
559
72
.
60.
Cercek
A
,
Lumish
M
,
Sinopoli
J
,
Weiss
J
,
Shia
J
,
Lamendola-Essel
M
, et al
.
PD-1 blockade in mismatch repair-deficient, locally advanced rectal cancer
.
N Engl J Med
2022
;
386
:
2363
76
.