Abstract

Background

   T-cell exhaustion is a major barrier to effective antitumor immunity
   and limits the efficacy of cancer immunotherapies. This study
   investigates the role of suppressor of T-cell signaling 2 (STS2, also
   known as UBASH3A) in regulating CD8^+ T-cell exhaustion within the
   tumor microenvironment.

Methods

   We used genetic ablation of STS2 in mouse models to assess tumor
   control and responses to anti-programmed cell death protein 1 (PD-1)
   checkpoint blockade therapy. CD8^+ tumor-infiltrating lymphocytes
   (TILs) were characterized through flow cytometry, mass cytometry, and
   single-cell transcriptomics. Mechanistic studies included
   co-immunoprecipitation, protein degradation assays, and endocytosis
   measurements to elucidate the interplay between STS2 and PD-1.

Results

   STS2 expression progressively increased with T-cell exhaustion. Genetic
   deletion of STS2 enhanced tumor control and improved responses to
   anti-PD-1 therapy. STS2-deficient CD8^+ TILs maintained a more
   functional state, exhibiting enhanced effector activity, proliferation,
   and antitumor efficacy while resisting terminal exhaustion.
   Mechanistically, we discovered that STS2 physically interacts with PD-1
   and modulates its expression, endocytosis, and degradation at the
   protein level.

Conclusions

   Our findings establish STS2 as a multifaceted regulator of T-cell
   exhaustion and highlight its potential as a therapeutic target for
   enhancing antitumor immunity and improving cancer immunotherapy
   outcomes.

   Keywords: Immunosuppression, Immunotherapy
     __________________________________________________________________

WHAT IS ALREADY KNOWN ON THIS TOPIC

     * T-cell exhaustion is a major barrier to effective antitumor
       immunity and limits the efficacy of cancer immunotherapies. The
       molecular mechanisms underlying T-cell exhaustion in the tumor
       microenvironment are not fully understood. Suppressor of T-cell
       signaling 2 (STS2) is known to negatively regulate T-cell receptor
       signaling, but its role in T-cell exhaustion and antitumor immunity
       has not been explored.

WHAT THIS STUDY ADDS

     * This study identifies STS2 as a critical regulator of CD8^+ T-cell
       differentiation in the tumor microenvironment. STS2 deficiency
       enhances tumor control and improves responses to anti-programmed
       cell death protein 1 (PD-1) checkpoint blockade therapy.
       STS2-deficient CD8^+ tumor-infiltrating lymphocytes maintain a more
       effective state, exhibiting enhanced functionality, proliferation,
       and antitumor activity while resisting terminal exhaustion. We
       uncover a novel mechanism by which STS2 regulates PD-1 expression
       at both transcriptional and post-transcriptional levels.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

     * These findings highlight STS2 as a potential therapeutic target for
       enhancing antitumor immunity and improving cancer immunotherapy
       outcomes. The study provides new insights into the molecular
       mechanisms of T-cell differentiation, which could inform the
       development of novel strategies to overcome resistance to
       checkpoint blockade therapies. Future research may focus on
       developing specific STS2 inhibitors and evaluating their efficacy
       in combination with existing immunotherapies.

Introduction

   Cancer immunotherapy, particularly immune checkpoint blockade (ICB),
   has revolutionized cancer treatment in recent years, demonstrating
   unprecedented efficacy in various malignancies.[49]^1 CD8^+ T cells
   play a pivotal role in antitumor immunity, recognizing and eliminating
   cancer cells through their cytotoxic activity.[50]^2 However, in the
   tumor microenvironment (TME), these cells often become exhausted,
   limiting their ability to control tumor growth.[51]^3 Recent studies
   have revealed that T-cell exhaustion is a gradual process, with cells
   transitioning through various states from effector to terminally
   exhausted phenotypes.[52]^4 Importantly, cells in intermediate states
   of exhaustion, often referred to as “progenitor exhausted” T cells,
   retain some proliferative capacity and responsiveness to checkpoint
   blockade, making them a key target for therapeutic interventions.[53]^5
   Understanding the molecular mechanisms underlying T-cell exhaustion is
   crucial for developing strategies to enhance the efficacy of cancer
   immunotherapies.

   Suppressor of T-cell receptor signaling 2 (STS2), also known as
   Ubiquitin-Associated And SH3 Domain-Containing Protein A (UBASH3A) and
   T-Cell Ubiquitin Ligand Protein,[54]^6 is a member of the STS family of
   proteins known to negatively regulate T-cell receptor (TCR)
   signaling.[55]^7 STS2 has been implicated in various autoimmune
   diseases[56]^8 9 and is highly expressed in lymphocytes.[57]^10 11
   While its role in modulating TCR signaling has been established, the
   potential involvement of STS2 in T-cell exhaustion and antitumor
   immunity remains unexplored.

   Programmed cell death protein 1 (PD-1) is a key inhibitory receptor
   expressed on exhausted T cells and a major target for checkpoint
   blockade therapies.[58]^12 The regulation of PD-1 involves complex
   mechanisms at both transcriptional and post-translational
   levels.[59]^13 Recent studies have highlighted the importance of
   protein trafficking, endocytosis, and degradation in modulating PD-1
   function.[60]^14 15 However, the potential role of STS2 in regulating
   PD-1 expression and function has not been investigated.

   In this study, we sought to elucidate the role of STS2 in CD8^+ T-cell
   exhaustion and antitumor immunity. We hypothesized that STS2, given its
   known inhibitory effects on T-cell signaling, might play a crucial role
   in promoting T-cell exhaustion in the TME. Using a combination of
   genetic mouse models, transcriptomic analyses, and functional assays,
   we investigated the impact of STS2 deficiency on CD8^+ T-cell function,
   exhaustion states, and antitumor responses. Furthermore, we explored
   the molecular mechanisms by which STS2 regulates T-cell function, with
   a particular focus on its interaction with PD-1 and its effects on
   protein metabolic process.

Results

STS2 is upregulated in tumor-infiltrating CD8^+ Tex cells

   While STS2 (gene symbol UBASH3A) is known to be highly expressed in
   peripheral blood and splenic lymphocytes, particularly T cells, under
   normal physiological conditions, its role in tumor development and its
   expression pattern within the TME remain unclear. To elucidate the
   immune cell expression of STS2 within the TME, we analyzed publicly
   available single-cell RNA-sequencing (scRNA-seq) data from human breast
   cancer tissue.[61]^16 Our analysis revealed high STS2 expression in T
   cells ([62]figure 1A). To validate these findings at the protein level,
   we performed CyTOF (Cytometry by time of flight) analysis of
   CD45-positive tumor-infiltrating cells in endometrial cancer tissues
   and found that STS2 is predominantly expressed in CD8^+ T cells, with
   detectable expression also observed in CD4^+ T cells ([63]online
   supplemental figure S1C,D). For further validation, we employed
   multiplex immunofluorescence on tissue microarray (TMA) of endometrial
   and colorectal cancers labeled with panCK/CD4/CD8/STS2 ([64]online
   supplemental figure S1E). In the endometrial cancer samples, STS2
   expression was predominantly observed in CD8^+ T cells compared with
   CD4^+ T cells ([65]figure 1B). Similarly, in colorectal cancer TMAs,
   the proportion of STS2^+CD8^+ T cells was significantly higher than
   that of STS2^+CD4^+ T cells in both tumor and adjacent tissues
   ([66]figure 1C, [67]online supplemental figure S1F). Notably, the
   proportion of STS2^+CD8^+ T cells among all CD8^+ T cells was
   significantly higher in tumors compared with adjacent tissues, while
   STS2^+ cells among CD4^+ T cells showed only an upward trend in tumors
   ([68]online supplemental figure S1G).

Figure 1. STS2 is upregulated in tumor-infiltrating CD8^+ Tex cells. (A) STS2
(gene symbol UBASH3A) mRNA levels in human breast cancer cohorts. (B–C)
Immunofluorescent staining of indicated molecules and quantification of
STS2^+CD4^+ or CD8^+ T cells of all CD4^+ or CD8^+ T cells in endometrial
cancer (B, n=189) and colon cancer (C, n=156) tissues. The scale bar of
higher magnification is 5 µm and the lower is 50 µm, as indicated in the
images. Statistical analysis was done by paired Student’s t-test. (D) STS2
mRNA levels in CD8^+TILs as classified by ProjecTILs, across human (left) and
mice (right) cohorts. (E) TCGA database RNA sequencing analysis of the
Pearson correlation between the expression of STS2 and exhausted T-cell or
effector T-cell signature genes in people with colon cancer (COAD),
endometrial cancer (UCEC), breast cancer (BRCA), or skin cutaneous melanoma
(SKCM). (F) t-SNE plots of CyTOF data depicting STS2 distribution across
PhenoGraph-defined CD8^+ TIL populations in human endometrial cancer
specimens. (G) STS2 expression per mouse (right) and representative
histograms (left) in Tim3^−CD39^−, Tim3^+CD39^+ and all CD8^+ TILs (n=4
mice). Statistical analysis was done by paired Student’s t-test. (H-I) Effect
of IL-2 on STS2 mRNA and protein expression after stimulation for 3 days in
mice CD8^+T cells. Statistical analysis was done by Student’s t-test. (J)
ATAC-seq and Cut&Run (H3K27ac)[69]^23 tracks of Stat5CA overexpression P14
cells at the STS2 locus. Blue highlights indicate ATAC peaks increased in P14
STAT5CA cells compared with P14 Empty. CM, central memory; CyTOF, cytometry
by time of flight; EM, effector memory; IL, interleukin; MFI, mean
fluorescent intensity; mRNA, messenger RNA; STS2, suppressor of T-cell
receptor signaling 2 ;TCGA, The Cancer Genome Atlas; TEMRA, effector
memory-expressing CD45RA T cells; Tex, exhausted T cells; TIL,
tumor-infiltrating lymphocyte; Tpex, progenitor exhausted T cells; TPM,
transcripts per million; t-SNE, t-distributed stochastic neighbor embedding.

   [70]Figure 1
   [71]Open in a new tab

   To further characterize STS2 expression in various subtypes of
   tumor-infiltrating CD8^+ T cells, we analyzed publicly available
   scRNA-seq data from human and mouse tumor-infiltrating lymphocytes
   (TILs) atlas using ProjecTILs classification.[72]17,[73]19 We observed
   that STS2 expression gradually increased with CD8^+ T-cell
   differentiation state, with the highest expression in exhausted T cells
   (Tex), followed by progenitor exhausted T cells (Tpex) ([74]figure 1D).
   Transcriptomic data from TCGA (The Cancer Genome Atlas) cohorts (colon
   cancer (COAD), endometrial cancer (UCEC), breast cancer (BRCA), and
   skin cutaneous melanoma (SKCM)) showed significant positive
   correlations between STS2 expression and exhausted T-cell signatures
   (HAVCR2, TIGIT, LAG3, PDCD1, LAYN), stronger than those with effector
   T-cell signatures (CX3CR1, FGFBP2, FCGR3A) ([75]figure 1E). To validate
   STS2 expression in CD8^+ TILs, we performed CyTOF analysis on CD8^+
   TILs from human endometrial cancer tissues and mouse MC38 tumors. In
   human endometrial cancer, STS2 expression was particularly enriched in
   exhausted CD8^+ T cells ([76]figure 1F, [77]online supplemental figure
   S2A). Analysis of CD8^+ TILs from murine MC38 tumors confirmed higher
   STS2 expression in terminally exhausted CD8^+ T cells (Tim3^+CD39^+)
   compared with Tim3^−CD39^− cells ([78]figure 1G).

   In transcriptional profiles from patients with melanoma treated with
   pembrolizumab,[79]^20 STS2 expression in tumors was higher than in
   PBMCs in non-naïve CD8^+ T cells ([80]online supplemental figure S2B).
   Analysis of tumor-infiltrating CD8^+ T cells transcriptome data from
   [81]GSE123235[82]^5 revealed significantly higher STS2 expression in
   terminally exhausted (Slamf6^−Tim3^+) cells compared with progenitor
   exhausted (Slamf6^+Tim3^−) cells ([83]online supplemental figure S2C,
   left). Similarly, in [84]GSE149876 data comparing four CD8^+ T-cell
   exhaustion states,[85]^21 STS2 expression was significantly higher in
   the terminal exhaustion group compared with progenitor and intermediate
   exhaustion groups ([86]online supplemental figure S2C, right). We
   further validated the high expression of STS2 in Tex cells using TISCH
   analysis of 14 public pan-cancer single-cell transcriptome datasets
   ([87]online supplemental figure S2D).

   To investigate the mechanism underlying STS2 upregulation, we found
   that interleukin (IL)-2 promoted STS2 expression in a dose-dependent
   manner at messenger RNA (mRNA) and protein levels ([88]figure 1H,I).
   Furthermore, STAT5 is a key downstream effector molecule of IL-2
   signaling and plays an important regulatory role in T cells.[89]^22
   Analysis of STAT5 CA overexpression CUT&RUN-seq and ATAC-seq
   data[90]^23 revealed enhanced chromatin accessibility and
   transcriptional binding at the STS2 locus ([91]figure 1J), suggesting
   that IL-2-mediated STAT5 activation may contribute to STS2 upregulation
   in CD8^+ T cells.

   Collectively, these results demonstrate that STS2 expression in the TME
   progressively increases with CD8^+ T-cell exhaustion, reaching its
   highest levels in terminally exhausted CD8^+ T cells.

STS2 deficiency enhanced the antitumor activity and immunotherapy effect

   Analysis using the Tumor Immune Dysfunction and Exclusion score,[92]^24
   25 which assesses gene function in T cells, revealed that patients with
   high CTL (cytotoxic T lymphocyte) infiltration in the STS2
   high-expression group had poorer prognosis than in the STS2
   low-expression group, suggesting STS2 association with T-cell
   dysfunction ([93]online supplemental figure S2E,F). Given the high
   expression of STS2 in CD8^+ terminally Tex cells and its association
   with T-cell inactivation, we hypothesized that STS2 loss might
   contribute to improved tumor control. To test this hypothesis, we
   generated STS2 knockout mice ([94]online supplemental figure S3A–C) and
   confirmed that STS2 deficiency did not affect lymphocyte numbers or
   CD3^+ cell numbers and CD4/CD8 cell proportions in peripheral blood and
   spleen ([95]online supplemental figure S3D–F).

   In vitro co-culture experiments with MC38 tumor cells and lymphocytes
   demonstrated that STS2-deficient (STS2^−/−) CD8^+ T cells induced
   higher rates of late-stage and overall apoptosis in tumor cells
   ([96]figure 2A and [97]online supplemental figure S3G). To investigate
   the immune-mediated role of STS2 in tumor control in vivo, we
   subcutaneously injected MC38 colorectal carcinoma cells into wild-type
   (WT) and STS2^−/− mice. STS2^−/− mice exhibited enhanced tumor control
   compared with WT mice ([98]figure 2B). This enhanced tumor control in
   STS2^−/− mice was also observed in MMTV-PyMT breast tumors ([99]figure
   2C) and minimally immunogenic B16F10 melanomas ([100]figure 2D),
   indicating that STS2 deficiency enables control of diverse tumor types.

Figure 2. STS2 deficiency enhanced the antitumor activity and immunotherapy
effect. (A) Apoptosis of MC38 tumor cells co-cultured with WT or STS2^−/−
CD8^+ T cells at a 3:1 effector:target ratio for 3 hours, measured by annexin
V and 7-AAD staining. The CD8^+ T cells were pre-activated in vitro for
3 days with anti-CD3/CD28 and IL-2. n=3 biologically independent samples.
Statistical analysis was done by Student’s t-test. (B–C) Temporal analysis of
tumor progression (mean size±SEM) in wild-type (WT) and STS2^−/− hosts
bearing MC38 (B, n=5) or MMTV-PyMT (C, n=7) neoplasms. Statistical analysis
was done by two-way ANOVA. (D) Comparative B16F10 melanoma burden between WT
and STS2^−/− mice received B16F10 tumor initiation on day 7 and day 14. n=8
mice per group. Statistical analysis was done by Student’s t-test. (E) MC38
tumor progression dynamics in WT or STS2^−/− mice that received isotype
control or anti-CD8-depleting antibody. n=6 mice per group. Statistical
analysis was done by two-way ANOVA. (F) MC38 tumor size after
NOD/Shi-Prkdc^scid Il2rg^em1/Cyagen mice received 2×10^6 pre-activated WT or
STS2^−/− CD8^+T cells (intravenous) on day 8 after tumor initiation. n=8 mice
for WT and STS2^−/− groups and 5 mice for no transfer group. Statistical
analysis was done by two-way ANOVA. (G–H) Tumor growth in WT or STS2^−/− mice
with MC38 (G) and B16F10 (H) that were treated with isotype or
anti-programmed cell death protein 1 blocking antibody beginning on day 15
(G) or day 6 (H) after tumor implantation. n=8 mice for MC38 groups and 7
mice for B16F10 groups. Statistical analysis was done by two-way ANOVA. (I)
Significantly higher expression of STS2 in CD8^+ T cells from Post versus
Pre-ICB treatment at NR patients with melanoma in [101]GSE120575. Statistical
analysis was done by Student’s t-test. 7-AAD, 7-aminoactinomycin D; ANOVA,
analysis of variance; ICB, immune checkpoint blockade; NR, non-responder;
Pre, pre-ICB; Post, post-ICB treatment; R, responder; STS2, suppressor of
T-cell receptor signaling 2.

   [102]Figure 2
   [103]Open in a new tab

   To identify the specific cell type mediating tumor control in STS2^−/−
   mice, we depleted CD8^+ T cells prior to MC38 tumor initiation. This
   depletion abolished the tumor control observed in isotype
   antibody-treated STS2^−/− mice, underscoring the crucial role of CD8^+
   T cells in this phenotype ([104]figure 2E and [105]online supplemental
   figure S3H). To further validate that STS2-mediated tumor control is
   CD8^+ T cell-intrinsic, we employed an adoptive transfer model. CD8^+ T
   cells from either WT or STS2^−/− mice were intravenously transferred
   into NOD/Shi-Prkdc^scid Il2rg^em1/Cyagen mice (which lack mature T, B,
   and natural killer immune cells) bearing MC38 tumors. The results
   demonstrated that STS2^−/− CD8^+ T cells exhibited enhanced tumor
   growth suppression compared with WT CD8^+ T cells ([106]figure 2F and
   [107]online supplemental figure S3I).

   The improved CD8^+ T cell-mediated tumor control in STS2^−/− mice
   suggested that these cells might be more responsive to CD8^+ T
   cell-enhancing immunotherapies. To test this hypothesis, we treated
   mice bearing established MC38 and B16F10 tumors with anti-PD-1 blocking
   antibodies ([108]figure 2G–H and [109]online supplemental figure
   S3J–K). While B16F10 tumors were better controlled in STS2^−/− mice
   compared with WT mice, they still progressed ([110]figure 2H and
   [111]online supplemental figure S2K), allowing for therapeutic
   intervention to control established tumors. These results indicate that
   STS2 deficiency in CD8^+ T cells enhanced the efficacy of anti-PD-1
   immunotherapy.

   To validate the antitumor immune and immunotherapy effects of STS2 in
   patients, we analyzed public datasets. Analysis of single-cell
   transcriptomics data from patient with melanoma tumor samples treated
   with checkpoint inhibitors ([112]GSE120575)[113]^26 revealed
   significant upregulation of STS2 expression in CD8^+ T cells from NR
   (non-responder) tumors compared with R (responder) tumors following ICB
   treatment, as well as in post-ICB samples compared with pre-ICB samples
   ([114]figure 2I).

   Collectively, these results demonstrate that STS2-deficient CD8^+ T
   cells contribute to enhanced antitumor immunity and exhibit synergistic
   effects with immunotherapy.

STS2-deficient CD8^+ T cells resist exhaustion and maintain functionality in
the TME

   To elucidate the mechanism by which STS2 deficiency in CD8^+ T cells
   facilitates long-term tumor control, we analyzed MC38-infiltrating
   CD8^+ T cells from WT and STS2^−/− mice 12 days post-MC38 implantation
   using CyTOF based on established methodologies[115]^27 ([116]online
   supplemental table S1, [117]online supplemental figure S4A).
   PhenoGraph-based clustering of WT and STS2^−/− CD8^+ T cells yielded
   eight distinct clusters ([118]figure 3A). In WT CD8^+ T cells, clusters
   1 and 2 comprised a larger proportion of WT cells (40% WT vs 17%
   STS2^−/−), whereas clusters 3 and 4 (24% WT vs 40% STS2^−/−)
   predominated in STS2^−/− CD8^+ T cells ([119]figure 3B). Protein
   expression analysis revealed that the dominant WT CD8^+ T-cell clusters
   (c1 and c2) expressed high levels of TOX, CD103, Lag3, Tim3, CTLA-4,
   and CD39, the high expression of these exhaustion markers and
   co-inhibitory receptors indicating a terminally
   differentiated/exhausted phenotype ([120]figure 3C). Conversely,
   STS2^−/− CD8^+ T cells showed a higher proportion of TCF1^+PD-1^+CD8^+
   T cells (c3) ([121]figure 3C, [122]online supplemental figure S4B),
   known for their self-renewal capacity and ability to generate
   terminally differentiated cytotoxic T cells. The predominant clusters
   in STS2^−/− CD8^+ T cells (c4) expressed lower levels of inhibitory
   receptors and increased levels of effector-induced protein interferon
   (IFN)-γ, suggesting an effective and activated state ([123]figure 3C).
   Notably, the proportions of non-activated, naïve/central memory
   TCF1^+PD-1^−CD39^− cells (c7) were comparable between WT and STS2^−/−
   CD8^+ TILs, indicating that STS2 deficiency does not deplete the TCF1^+
   stem-like population.

Figure 3. STS2-deficient CD8^+T cells resist exhaustion and maintain
functionality in the tumor microenvironment. Comprehensive mass cytometry
analysis of pooled MC38 tumor-infiltrating CD8^+TILs from WT and STS2^−/−
mice (n=5 per genotype). (A–B) High-dimensional visualization (t-SNE) of
CD8^+ TIL populations (A) with corresponding cluster frequency analysis (B).
(C–D) Expression profiling of functional markers across identified clusters
(C) and individual samples (D), presented as normalized Z-scores. (E–G)
Analysis of human endometrial cancer specimens (n=3), stratified by STS2
expression levels, including cluster identification (E), frequency
distribution (F), and protein expression patterns (G). (H–I) Flow cytometric
quantification of CD39^+PD-1^+ (n=5) or PD-1^+ (n=5) (H) or IFN-γ^+ (n=4) (I)
of CD8^+ T cells in MC38 tumor. Statistical analysis was done by Student’s
t-test. IFN, interferon; PBMC, peripheral blood mononuclear cells; PD-1,
programmed cell death protein 1; STS2, suppressor of T-cell receptor; t-SNE,
t-distributed stochastic neighbor embedding; tdLN, tumor-draining lymph
nodes; TILs, tumor-infiltrating lymphocytes; WT, wild-type.

   [124]Figure 3
   [125]Open in a new tab

   We also analyzed CD8^+ T cells from peripheral blood and tumor-draining
   lymph nodes (tdLN). t-SNE (t-distributed stochastic neighbor embedding)
   dimensionality reduction clustering showed that CD8^+ T cells in
   peripheral blood and tdLN clustered together and were distinct from
   tumor-infiltrating CD8^+ T cells ([126]online supplemental figure S4C).
   Protein expression analysis revealed higher IFN-γ expression in
   STS2^–/– CD8^+ T cells across all three compartments, PBMC (peripheral
   blood mononuclear cells), tdLN, and tumor ([127]figure 3D). In tdLN,
   CD62L and Tcf1 were higher in the STS2^–/– group, while in tumors,
   expression of immunosuppressive receptors (CTLA-4, PD-1, CD39, Tim3,
   and Lag3) and Tox was higher in the WT group ([128]figure 3D).

   To investigate the relationship between STS2 expression and cellular
   activation in human tumor samples, we analyzed STS2 distribution in
   CD8^+ TILs from three human endometrial tumor tissues. Clustering CD8^+
   TILs based on the upper and lower thirds of STS2 expression revealed
   distinct enrichment patterns ([129]figure 3E,F, [130]online
   supplemental figure S4C). Compared with the STS2-high group, cluster 3
   showed the largest decrease in STS2-low cells, while cluster 4 showed
   the largest increase ([131]online supplemental figure S4D–F).
   Consistently, increased levels of exhaustion markers and co-inhibitory
   receptors (Tox, PD-1, ICOS, Tim3, Lag3, CTLA4) were observed in the
   STS2-high subset, whereas activation marker (CD25) was increased in the
   STS2-low subset in human endometrial tumor tissues ([132]figure 3G). We
   used spectral flow cytometry to verify the decreased proportion of
   CD39^+PD-1^+ and PD-1^+CD8^+ cells ([133]figure 3H), and the increased
   proportion of IFN-γ^+CD8^+ cells ([134]figure 3I) in STS2^–/– mice MC38
   tumors ([135]online supplemental figure S4G,H).

   To further investigate whether STS2 deficiency affects the functional
   trajectory of CD8^+ T cells under persistent stimulation in vitro, we
   conducted a temporal analysis of CD8^+ T cells stimulated with
   anti-CD3/CD28 and IL-2 over 10 days. CyTOF analysis revealed that
   STS2^–/– CD8^+ T cells maintained higher expression of the effector
   cytokine IFN-γ at day 10 compared with WT cells ([136]online
   supplemental figure S5). Similarly, STS2^–/– cells sustained higher
   levels of proliferation marker Ki67 and activation marker CD25 by day
   10, while expressing lower levels of exhaustion-associated markers PD-1
   and CD39 ([137]online supplemental figure S5). These in vitro findings
   support our in vivo observations that STS2 deficiency delays functional
   decline and acquisition of exhaustion markers under persistent
   stimulation.

   In summary, both mouse and human CD8^+ TILs demonstrated that STS2
   high-expressing cells were enriched for co-expression of exhaustion
   markers and co-inhibitory receptors, with STS2 most highly expressed in
   the most terminally exhausted subsets. STS2-deficient CD8^+ TILs
   exhibited resistance to exhaustion and maintained functionality within
   the TME.

Transcriptional profiling revealed STS2-deficient CD8^+TILs sustained
effector programming and resistance to exhaustion

   Single-cell transcriptomic data revealed no significant differences in
   T-cell states between the spleens of tumor-bearing WT and STS2^–/– mice
   ([138]online supplemental figure S6A). To decipher how STS2
   transcriptionally programs CD8^+ TILs, we performed combined scRNA-seq
   and TCR sequencing plus antibody staining on CD45-enriched
   tumor-infiltrating cells from WT and STS2^–/– mice, with two samples
   per group ([139]online supplemental figure S6B,C). Consistent with
   [140]figure 1, STS2 expression in the immune microenvironment was
   predominantly in CD8^+ T cells ([141]online supplemental figure S6C,D).
   Seurat-based clustering of the WT and STS2^–/– CD8^+ TILs resolved six
   distinct clusters ([142]figure 4A, [143]online supplemental table S2),
   with notable differences in cluster proportions between the two groups
   ([144]figure 4B, [145]online supplemental figure S6E).

Figure 4. STS2-deficient CD8^+TILs sustained effector programming and
resistance to exhaustion. (A–B) Single-cell transcriptomic profiling of WT
and STS2^–/– CD8^+TILs (A) with quantitative distribution analysis across
identified clusters (B). (C) Bubble plot showing relative expression of
cell-state-associated genes. Bubble size is proportional to the percentage of
cells expressing a gene and color intensity is proportional to average scaled
gene expression. (D) Heatmap illustrating expression of 19 curated gene
signatures across CD8^+TILs cluster. Heatmap was generated based on the
scaled gene signature scores. (E) The terminal exhausted signature score of
WT and STS2^–/– CD8^+TILs. Statistical analysis was done by Student’s t-test.
The trend lines connected the median values between groups. (F) RNA velocity
analysis from WT and STS2^–/– CD8^+TILs in figure 4A for clusters 1–6. Arrows
show directions of cell state transitions. (G) Pseudotime analysis of the
indicated states from [146]online supplemental figure S4G. (H) The single
cell counts of each clone type in each cluster. (I) Frequency of CD8^+TILs
cells within paired TCR sequences. Clonotype frequency (X) indicates the
relative abundance of each T-cell clone within the analyzable TCR repertoire:
medium (1–10%), large (10–100%), and hyper-expanded (>100%, theoretical
maximum). The frequency (X) represents the proportion of cells belonging to
each clonotype among the CD8^+ TILs that have paired TCR sequences. IFN,
interferon; STS2, suppressor of T-cell receptor signaling 2; TCR, T-cell
receptor; TILs, tumor-infiltrating lymphocytes; UMAP, uniform manifold
approximation and projection; WT, wild-type.

   [147]Figure 4
   [148]Open in a new tab

   The naïve-like cluster (c1) expressed stemness-associated markers Tcf7
   (encoding TCF1), Sell (encoding CD62L), and Slamf6 (encoding Ly108) and
   displayed a naïve-like phenotype with high expression of a naïve gene
   signature ([149]online supplemental table S3), and its abundance was
   comparable between WT and STS2^−/− cells (15% WT vs 13% STS2^−/−)
   ([150]figure 4B–D). Notably, cluster 2 (IFN−responsive effector
   memory), which highly expressed the effector-associated chemokine Ccl5,
   effector memory marker Ly6c2, and signatures for IFN response (Isg15,
   Bst2, Ifi27l2a), cytotoxicity, and TCR signaling, along with the
   activated cluster (c3) expressing NF-κB signaling pathway gene Nfkbia,
   AP-1 family member Junb, NR4A family transcription factor Nr4a1, and
   survival factor receptor Il7r, together accounted for 39% of STS2^−/−
   CD8^+ TILs but only 18% in WT CD8^+ TILs ([151]figure 4B–D).

   Clusters 4, 5, and 6 expressed Tox, a key regulator of exhaustion, and
   showed lower proportions in the STS2^−/− group compared with WT (47% vs
   68% total) ([152]figure 4B,C). Among these, the metabolically active
   cluster (c4) exhibited high expression of glycolysis (Ldha), DNA
   replication (MCM family), and protein synthesis (Eif5a) genes, along
   with elevated oxidative phosphorylation and glycolysis signatures. The
   proliferating cluster (c5) exhibited high expression of cell
   cycle-related markers (Mki67, Top2a, Stmn1) and showed the MAPK
   signaling signature. Clusters 6 showed abundant expression of Pdcd1
   (encoding PD-1) and Havcr2 (encoding TIM3), with the highest expression
   of co-inhibitory receptors Entpd1 (encoding CD39), Cd38, and Cd244a
   ([153]figure 4B,C). Remarkably, STS2^−/− CD8^+ TILs downregulated
   terminally exhausted gene signatures ([154]online supplemental table
   S3), while maintaining progenitor and intermediate exhausted gene
   signatures[155]^21 ([156]figure 4E and [157]online supplemental figure
   S6F).

   To infer directionality of CD8^+T cell development, we assessed RNA
   velocity[158]^28 overlaid on the UMAP (uniform manifold approximation
   and projection) of cells from all identified clusters ([159]figure 4F).
   We identified continuous trajectories aligning CD8^+ T-cell
   subpopulations, particularly evident from clusters 2 and 3 to 6,
   supporting differentiation from activated and/or IFN−responsive
   effector memory cells to terminal exhausted cells ([160]figure 4F, up).
   Consistently, this trajectory was observed when visualizing WT or
   STS2^−/− cell clusters separately ([161]figure 4F, down). To further
   understand the differentiation trajectories of CD8^+ T cells, we
   projected the expression of naïve, activation/effector, cytotoxicity,
   and exhaustion genes and signatures onto the UMAP ([162]online
   supplemental figure S6G,H). Building on this, Monocle3 analysis
   revealed the differentiation trajectories ([163]online supplemental
   figure S6I), and pseudotime analysis indicated that STS2^−/− cells
   mainly accumulated in the intermediate differentiation cell state
   ([164]figure 4G), further supporting the idea that STS2 deficiency
   delays terminal differentiation.

   Single-cell TCR sequencing provided insights into the clonal dynamics
   of tumor-infiltrating T cells ([165]online supplemental figure S6J). We
   observed hyper-expanded clones were predominantly enriched in the
   terminal exhausted CD8^+ T-cell cluster, suggesting that extensive
   proliferation may be associated with T-cell dysfunction rather than
   enhanced antitumor activity ([166]figure 4H). Moreover, we analyzed the
   expression of effector function, cytotoxicity, and TCR signaling
   signatures across different clone size categories (medium, large, and
   hyper-expanded), which revealed that hyper-expanded clones exhibited
   lower signature scores for these functional pathways compared with
   medium and large clones ([167]online supplemental figure S6K).
   Furthermore, the proportion of hyper-expanded clones was lower in
   STS2^−/− mice compared with WT mice (17% vs 23%) ([168]figure 4I),
   suggesting that STS2 deficiency inhibits T-cell differentiation toward
   a highly expanded but functionally compromised state under persistent
   antigen stimulation.

   Finally, to validate our analysis in the context of established T-cell
   states, we used the R package ProjecTILs[169]^17 to project CD8^+ TILs
   onto the mouse CD8^+ TIL reference atlas. Consistent with our previous
   observations, the proportion of CD8_Tex was higher in the WT group,
   while the proportion of effector memory CD8^+ T cells was higher in
   STS2^−/− cells ([170]online supplemental figure S6L).

   These findings collectively demonstrate that STS2 deficiency promotes a
   more functional CD8^+ T-cell repertoire within the TME, characterized
   by sustained effector programming and resistance to terminal
   exhaustion.

STS2 broadly regulates T-cell function and modulates PD-1 expression at the
post-transcriptional level

   To investigate the regulatory mechanisms of STS2 in lymphocytes, we
   performed RNA sequencing (RNA-seq) and ATAC-seq (assay for
   transposase-accessible chromatin using sequencing) on lymphocytes
   isolated from the spleens of WT and STS2^−/− mice ([171]figure 5A–D),
   which were stimulated in vitro for 3 days with anti-CD3/CD28 and IL-2.
   The RNA-seq results revealed significant changes at the mRNA level in
   lymphocytes due to STS2 deficiency. Among the most significantly
   downregulated genes in the STS2^−/− group were stress-related heat
   shock genes (such as HSPA1A and HSPA1B) ([172]figure 5A), which have
   been implicated in resistance to immunotherapy.[173]^29 In terms of
   immune regulatory pathways, STS2 deficiency led to transcriptional
   changes in the expression of surface molecules (Btla, Ctla4, Cd247,
   Lag3, Cd3e), cytokines (Il4, Il5, Il10, Ccl3, Ccl4), effector molecules
   (Tnf, Ifng, Gzmb, Prf1), and transcription factors (Tox, Jun, Tbx21)
   ([174]figure 5B), as well as alterations in their chromatin
   accessibility ([175]figure 5D), indicating that the TCR and CD28
   pathways were affected by the absence of STS2. Pathway enrichment
   analysis of differentially expressed genes showed that the genes
   regulated by STS2 were primarily enriched in processes related to the
   inflammatory response, regulation of intracellular signal transduction,
   and regulation of protein metabolic processes ([176]figure 5C),
   consistent with previous reports.[177]1030,[178]32

Figure 5. STS2 broadly regulates T-cell function. (A) Volcano plot showing
DEGs in RNA-seq of WT and STS2^−/− lymphocytes stimulated in vitro for 3 days
with anti-CD3/CD28 and IL-2. (B) Heatmap showing relative expression (Z score
of the normalized counts) of surface molecules, cytokines and effector
molecules and transcriptional factors in RNA-seq of WT and STS2^−/−
lymphocytes. (C) Bubble plot of GO pathways enriched in DEGs between WT and
STS2^−/− lymphocytes. The size of the dots represents the counts of enriched
genes, and the color represents the enrichment score. (D) ATAC-seq tracks of
the indicated gene locus in activated WT and STS2^−/− lymphocytes. (E) Bubble
plot of GO pathways enriched in STS2-interacting proteins. The size of the
dots represents the counts of enriched genes, and the color represents the
enrichment score. (F) Apoptosis of WT or STS2^−/− lymphocytes co-cultured
with MC38 tumor cells measured by annexin V and propidium iodide (PI)
staining. n=3 biologically independent samples. Statistical analysis was done
by Student’s t-test. (G) In vitro proliferation of CD8^+T cells from WT and
STS2^−/− mice stimulated by anti-CD3, anti-CD28 and IL-2 for 4 days and
analyzed by CellTiter 96 AQueous One Solution Cell Proliferation Assay.
Statistical analysis was done by two-way analysis of variance. (H) RT-qPCR
analysis of Mki67 expression in WT and STS2^−/− CD8^+ TILs. Statistical
analysis was done by Student’s t-test. (I) Co-IP detection of the interaction
between STS2 with CD3ε and c-Cbl. ATAC-seq, assay for transposase-accessible
chromatin using sequencing; Co-IP, co-immunoprecipitation; DEG,
differentially expressed gene; GO, gene ontology; IL, interleukin; mRNA,
messenger RNA; RNA-seq, RNA sequencing; RT-qPCR, reverse transcription
quantitative polymerase chain reaction; STS2, suppressor of T-cell receptor
signaling 2; WT, wild-type.

   [179]Figure 5
   [180]Open in a new tab

   To further explore the regulatory mechanism of STS2 in lymphocytes, we
   analyzed the interacting proteins of STS2 reported by a previous
   study[181]^10 ([182]figure 5E). Interestingly, proteins interacting
   with STS2 were enriched in the regulation of cell death pathways. This
   led us to investigate the impact of STS2 on lymphocyte apoptosis. After
   co-culturing lymphocytes and MC38 tumor cells in vitro, we observed
   that both late and total apoptosis in the STS2^−/− lymphocytes were
   significantly lower than those in the control group ([183]figure 5F and
   [184]online supplemental figure S7A). Additionally, we found that the
   proliferation rate of STS2-deficient lymphocytes in vitro was
   significantly higher than that of the WT group ([185]figure 5G), as
   well as the level of Mki67 mRNA expression was higher in STS2^−/−
   CD8^+TILs ([186]figure 5H).

   Previous studies have demonstrated that STS2 regulates T-cell function
   through multiple pathways, particularly impacting TCR signaling.[187]^7
   33 34 Unlike STS1, another member of the STS family, STS2 exerts its
   influence not through phosphatase activity, but rather by interacting
   with key nodes in the TCR pathway.[188]35,[189]37 To further elucidate
   this mechanism, we performed in vitro experiments to detect the
   phosphorylation of Zap70 and Erk1/2, key nodes in the TCR signaling
   process, following anti-CD3/CD28 stimulation. There was no significant
   difference between the STS2^−/− and WT groups ([190]online supplemental
   figure S7B). Additionally, we confirmed the interaction of STS2 with
   Cbl-b and CD3ε through co-immunoprecipitation assays ([191]figure 5I).

   Notably, the transcriptional level of Pdcd1 was increased in STS2^−/−
   CD8^+TILs ([192]figure 6A, right), but cell-surface levels of PD-1 were
   downregulated in STS2^−/− CD8^+T cells in tumors ([193]figure 3H),
   peripheral blood, and spleen ([194]figure 6A, left). We speculate that
   regulation at the transcriptional level may be due to the interaction
   of STS2 with the IκB complex (TAK1 and NEMO),[195]^30 which inhibits
   TCR-induced NFκB signaling through ubiquitin-dependent inhibition,
   thereby suppressing PD-1 transcription. STS2’s interacting proteins and
   regulated genes were enriched in pathways related to protein
   localization, synthesis, and degradation, such as cellular protein
   localization, cellular component biogenesis, and regulation of protein
   metabolic processes ([196]figure 5C,E). Given that STS2 is involved in
   the regulation of protein metabolic processes, we hypothesize that STS2
   may interact with PD-1 at the protein level. Co-immunoprecipitation
   experiments confirmed the interaction between ectopically expressed
   Flag-tagged PD-1 and Myc-STS2 in 293T cells ([197]figure 6B), and
   endogenously in Jurkat cells ([198]figure 6C). These data suggest that
   STS2 interacts with PD-1 and mediates its downregulation by protein
   metabolic processes.

Figure 6. STS2 modulates PD-1 expression at the post-transcriptional level.
(A) The ratio of PD-1^+CD8^+ of CD8^+ cells in PBMC and spleen cells from
MC38 tumor-bearing mice in WT and STS2^−/− mice (left). RT-qPCR analysis of
Pdcd1 expression in WT and STS2^−/− CD8^+ tumor-infiltrating lymphocytes
(right). Statistical analysis was done by Student’s t-test. (B–C) Validation
of the interaction between STS2 and PD-1 by Co-IP in 293T (B) and Jurkat
cells (C). (D) Immunoblotting showing the construction of STS2 knockout cells
in Jurkat cells. (E–H) Change in the levels of PD-1 on the cell surface after
treatment with the proteasome inhibitor MG132 (E–F) or the lysosome inhibitor
CQ (G–H) in Jurkat (E, G) or mice lymphocytes (F, H). n=4 for Jurkat cells
per group and n=3 for mice lymphocytes per group. Statistical analysis was
done by Student’s t-test. (I) GSEA plots showing enrichment in PROTEASOME and
ENDOCYTOSIS pathways of DEGs between WT and STS2^−/− lymphocytes RNA-seq.
(J–K) The remaining antibody-labeled surface PD-1 was detected with an
APC-conjugated secondary antibody in Jurkat (J) or mice lymphocytes (K).
Statistical analysis was done by two-way analysis of variance. Co-IP,
co-immunoprecipitation; GSEA, gene set enrichment analysis; MFI, mean
fluorescent intensity; NES, normalized enrichment score; PBMC, peripheral
blood mononuclear cells; PD-1, programmed cell death protein 1; RNA-seq, RNA
sequencing; RT-qPCR, reverse transcription quantitative polymerase chain
reaction; STS2, suppressor of T-cell receptor signaling 2 ;t-SNE,
t-distributed stochastic neighbor embedding; WT, wild-type.

   [199]Figure 6
   [200]Open in a new tab

   To elucidate the mechanism of STS2-mediated PD-1 regulation, we
   generated STS2-knockout (sgSTS2) Jurkat cells ([201]figure 6D) and
   examined whether STS2 affects PD-1 degradation via the proteasome or
   lysosome pathways. Treatment with MG132, a specific proteasome
   inhibitor, protected the PD-1 protein from degradation in both sgSTS2
   Jurkat cells and mouse STS2^–/– lymphocytes ([202]figure 6E,F),
   suggesting the involvement of the proteasome in STS2 regulation, which
   is consistent with the enrichment in the proteasome pathway of DEGs
   (differentially expressed genes) in RNA-seq of STS2^–/– lymphocytes
   ([203]figure 6I, top). Moreover, we found that knockdown of STS2 caused
   increased PD-1 ubiquitination in Jurkat cells ([204]online supplemental
   figure S7C). Interestingly, lysosome inhibitors also protected PD-1
   protein from degradation in these cells ([205]figure 6G,H), indicating
   that both proteasomal and lysosomal pathways may be involved in
   STS2-mediated PD-1 regulation.

   Given that STS2 can affect the endocytosis pathway ([206]figure 6I,
   down), as demonstrated by its ability to facilitate the endocytosis of
   cell-surface TCR–CD3 complexes and/or inhibit their recycling back to
   the plasma membrane,[207]^10 we explored whether STS2 influences PD-1
   endocytosis. We employed a surface labeling assay to track PD-1
   internalization ([208]online supplemental figure S7D). Surprisingly,
   our results showed that in the STS2-deficient group, although the
   initial PD-1 content on the membrane surface was relatively low, the
   proportion of PD-1 remaining on the membrane surface after endocytosis
   was higher compared with the control group ([209]figure 6J,K). This
   suggests that STS2 deficiency inhibits the endocytosis of PD-1,
   potentially contributing to its altered expression and function in
   STS2^–/– T cells.

   Collectively, these findings demonstrate that STS2 plays a multifaceted
   role in regulating T-cell function, including modulation of cell death,
   proliferation, and protein metabolic process. Importantly, our results
   reveal a novel mechanism by which STS2 regulates PD-1 expression
   through both transcriptional and post-transcriptional processes,
   including protein–protein interactions, degradation pathways, and
   endocytosis. This intricate regulation of PD-1 by STS2 may contribute
   to the enhanced antitumor immunity observed in STS2-deficient T cells.

Discussion

   In this study, we have uncovered a critical role for STS2 in regulating
   CD8^+ T-cell exhaustion and antitumor immunity. Our findings reveal
   that STS2 deficiency enhances T cell-mediated tumor control and
   improves the efficacy of checkpoint blockade immunotherapy ([210]figure
   7). This effect is mediated through multiple mechanisms, including the
   maintenance of effector function, resistance to terminal exhaustion,
   and modulation of PD-1 expression.

Figure 7. STS2 deficiency enhances T cell-mediated tumor control and improves
the efficacy of checkpoint blockade immunotherapy. ICB, immune checkpoint
blockade; PD-1, programmed cell death protein 1; STS2, suppressor of T-cell
receptor signaling 2; WT, wild-type.

   [211]Figure 7
   [212]Open in a new tab

   The upregulation of STS2 in tumor-infiltrating CD8^+ Tex cells, as
   demonstrated by our scRNA-seq and protein-level analyses, suggests that
   STS2 may serve as a marker of T-cell exhaustion in the TME. This
   observation is consistent across multiple tumor types and species,
   indicating a conserved role for STS2 in T-cell biology. The gradual
   increase in STS2 expression with T-cell exhaustion progression implies
   that STS2 may be involved in the transition from effector to exhausted
   states.

   Our transcriptional profiling revealed that STS2 deficiency promotes a
   more functional CD8^+ T-cell state within tumors. The increased
   proportion of activated and effector-like cells and the maintenance of
   IFN-responsive states in STS2^−/− CD8^+ TILs suggest that STS2 may act
   as a molecular brake on T-cell function. By removing this brake,
   STS2-deficient T cells resist terminal exhaustion and maintain
   antitumor potential through enhanced TCR signaling and decreased PD-1
   expression. This finding aligns with the current understanding of
   T-cell exhaustion as a plastic manipulable for therapeutic
   benefit.[213]^38 39

   The broad regulatory effects of STS2 on T-cell function, including its
   impact on cell death, proliferation, and protein metabolic process,
   highlight its potential as a multifaceted regulator of T-cell
   responses. STS2 enhances endocytosis and downregulation of TCR-CD3
   complexes after TCR stimulation in Jurkat cells.[214]^10 Moreover, STS2
   inhibits TCR signaling by interacting with c-Cbl and Cbl-b of the Cbl
   family.[215]^40 41 Our discovery of STS2’s interaction with PD-1 and
   its involvement in PD-1 regulation at both transcriptional and
   post-transcriptional levels is particularly intriguing.

   Notably, our results suggest STS2’s role in regulating PD-1 through a
   ubiquitin-dependent mechanism. This finding is consistent with previous
   studies showing that STS2 interacts with EGFR and regulates its
   expression through Cbl-mediated ubiquitination and subsequent
   degradation,[216]^41 and that Nrdp1 promotes Zap70 ubiquitin-dependent
   degradation in an STS2-dependent manner, thereby negatively regulating
   TCR signaling pathways.[217]^42 Interestingly, this finding aligns with
   recent studies on post-translational modifications of PD-1. For
   instance, TOX has been shown to affect the endocytic recycling pathway
   of PD-1,[218]^14 FBXO38 can influence PD-1 ubiquitination,[219]^15 and
   UFL1 affects the UFMylation of PD-1.[220]^43 Our discovery that STS2
   can affect PD-1 post-translational modification through multiple
   pathways adds to this growing body of evidence and underscores the
   complex regulation of this crucial immune checkpoint molecule.

   From a therapeutic perspective, our results suggest that targeting STS2
   could be a promising strategy to enhance antitumor immunity. The
   synergistic effect observed between STS2 deficiency and anti-PD-1
   immunotherapy is particularly encouraging. This combination approach
   could potentially overcome resistance to checkpoint blockade therapies,
   a significant challenge in cancer immunotherapy.[221]^44 45

   However, several questions remain to be addressed. It is important to
   note that our study primarily examines polyclonal rather than
   antigen-specific T-cell responses. Future studies employing
   antigen-specific models, such as TCR transgenic T cells or defined
   tumor antigens, would further refine our understanding of how STS2
   influences antigen-specific T-cell exhaustion and antitumor immunity.
   The precise molecular mechanisms by which STS2 promotes T-cell
   exhaustion and PD-1 expression need further elucidation. Additionally,
   the potential off-target effects of STS2 inhibition on other immune
   cell populations and normal tissues should be carefully evaluated. The
   translational potential of these findings will depend on the
   development of specific STS2 inhibitors and their evaluation in
   preclinical models.

   In conclusion, our study identifies STS2 as a key regulator of CD8^+
   T-cell exhaustion and antitumor immunity. By maintaining T-cell
   functionality, modulating PD-1 expression, and influencing the balance
   between effector and exhausted states, STS2 deficiency enhances tumor
   control and improves responses to immunotherapy. These findings not
   only advance our understanding of T-cell biology in the TME but also
   pave the way for novel therapeutic strategies in cancer immunotherapy.
   Future studies should focus on developing targeted approaches to
   modulate STS2 function in T cells, potentially opening new avenues for
   enhancing antitumor immune responses in patients with cancer.

Materials and methods

Cell culture and isolation

   Cell lines MC38, B16F10, Jurkat, and 293T were obtained from the
   National Infrastructure of Cell Line Resources (Beijing, China). All
   cell lines were maintained in Dulbecco’s Modified Eagle Medium
   supplemented with 10% fetal bovine serum and 1%
   penicillin/streptomycin. Cells were cultured at 37°C in a humidified
   incubator with 5% CO[2].

   CD8^+ T cells were isolated from mice splenocytes and then purified by
   negative selection using the EasySep Mouse CD8^+ T Cell Enrichment Kit
   (STEMCELL) according to the manufacturer’s instructions.

Animal studies

   C57BL/6N mice were purchased from Beijing Vital River Laboratory Animal
   Technology (Beijing, China). Animals were housed under specific
   pathogen-free conditions with a 12-hour light/dark cycle and ad libitum
   access to standard chow and water. All experiments used age-matched and
   sex-matched mice between 7 and 10 weeks old at the start of each study.

Generation of STS2 knockout mice

   STS2 knockout (STS2^−/−) mice were generated on the C57BL/6N background
   by Cyagen Biotechnology using CRISPR-Cas9 technology. Exons 2–5 of the
   STS2 gene were targeted for deletion. Cas9 mRNA and guide RNAs (gRNAs)
   were co-injected into fertilized eggs. Resulting pups were genotyped by
   PCR and confirmed by sequencing analysis. The usability of the STS2
   antibody in FC (flow cytometry) and CyTOF was validated by staining
   immune cells derived from WT and STS2^−/− mice ([222]online
   supplemental figure S1A,B).

   The following sgRNA sequences for mice STS2 gene knockout were used:

   gRNA1 (matching forward strand of gene): 5'-
   TTACTGCTTAAGCCGTAGGGTGG-3';

   gRNA2 (matching reverse strand of gene): 5'-
   TCCCTCATGCTTATTCCCTGTGG-3'.

Human tissue samples

   Surgical specimens were obtained from the patients with endometrial
   cancer at Beijing Obstetrics and Gynecology Hospital (Approval ID:
   2022-KY-037–01). TMAs were constructed by harvesting 400 µm tissue
   cores from paraffin-wax embedded samples collected from 180 patients
   with colorectal cancer and 62 patients with endometrial cancer.

Statistical analysis

   Statistical analyses and generation of graphics were performed using
   GraphPad Prism V.8. All data were expressed as mean±SEM. For
   statistical results, when only two groups were applied, Student’s
   t-test was used; when more than two groups were applied, one-way
   analysis of variance (ANOVA) was used. For the tumor growth data,
   two-way ANOVA was used. Details of statistical tests used are indicated
   in the respective figure legends. P value<0.05 (two-tailed) was
   considered statistically significant, with *p<0.05, **p<0.01, and
   ***p<0.001, respectively.

   Other materials and methods were provided in the [223]online
   supplemental file 2.

Supplementary material

   online supplemental file 1
   [224]jitc-13-6-s001.pdf^ (3.2MB, pdf)
   DOI: 10.1136/jitc-2024-010735
   online supplemental file 2
   [225]jitc-13-6-s002.pdf^ (285.1KB, pdf)
   DOI: 10.1136/jitc-2024-010735
   online supplemental file 3
   [226]jitc-13-6-s003.pdf^ (231.2KB, pdf)
   DOI: 10.1136/jitc-2024-010735

Acknowledgements