We utilized R package “ggplot2” to analysis the differential expression of TMEM176B in various malignancies or cancer subtypes. Extraction of SKCM in TCGA and corresponding normal tissue data in GTEx has been analyzed for differential expression of TMEM176B in cutaneous melanoma. Click “repository” in TCGA database and make the following settings: under File directory (select “Gene Expression Quantification” under Data Category module, select “HTseq-FPKM” under Workflow Type module); under Cases (under Primary Site module, select “skin,” and under Project, select “TCGA-SKCM”); under Cases (select “skin” under the Primary Site module and “TCGA-SKCM” under Project) to get the clinical information of SKCM.

R package “survminer” and “survival” were utilized to analyze the prognostic value of the expression level of TMEM176B in different tumors. We collected the data from the TCGA database (Liu et al., 2018) and compared the overall survival (OS) of cancer patients separated by the median expression level of TMEM176B. The “pROC” package and the “ggplot2” package were used to extract and analyze the SKCM data of TCGA and the corresponding normal tissue data in GTEx to obtain the receiver operating characteristic (ROC) curves.

We performed Cox regression analysis to identify the association between SKCM disease characteristics and the expression level of TMEM176B.

In PrognoScan, the association between TMEM176B expression and prognosis of cancer patients was analyzed, including distant metastasis-free survival (DMFS), disease-free survival (DFS), relapse-free survival (RFS), OS, and distant recurrence-free survival (DRFS).

TIMER is a fully functional public platform for analyzing of immune infiltration in tumors. Correlations between TMEM176B expression and immune infiltration levels in more than 30 different cancer types in TCGA were obtained in the gene module of TIMER. We also investigated the connection between tumor purity and TMEM176B expression. And we utilized R package “GSVA” to obtain the link among TMEM176B and other immunocytes (Bindea et al., 2013). Next “CIBERSORT algorithm” was used to explore the immune infiltration between high- and low-expression with TMEM176B to obtain enriched immunocytes between the two groups.

We acquired the top 100 genes associated with TMEM176B in SKCM cases using the “Similar Gene Detection” module of GEPIA2 and performed correlation analysis on the top 5 genes using “Correlation Analysis” in GEPIA2. We then obtained protein networks that may interact with TMEM176B in the STRING database. In addition, we subjected the 100 genes obtained in GEPIA2 to GO and KEGG pathway analysis using the “clusterProfiler” and “ggplot2” R packages.

R package “ggplot2” was used to determine the link between the expression level of TMEM176B in SKCM and other genes, including CD274, CTLA4, GZMA, GAMB, IFNG, PRF1, and so on.

Aged 6–8 weeks Tmem176b ^−/− mice and C57BL/6J mice were purchased from GemPharmatech Co. Ltd. and Shanghai Jie Si Jie Laboratory Animal Co. Ltd., respectively.

B16-OVA melanoma cells were cultured in DMEM medium, which contains 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin.

We injected 1 × 10⁶ B16-OVA cells into the right flank of WT or Tmem176b ^−/− mice subcutaneously and used vernier calipers to measure tumor volume every 1–2 days. On the 14th day, the mice were euthanized and their tumor tissues, spleen, lymph nodes, and peripheral blood were taken for further research.

Cells were stained with CD3 (BioLegend, clone:17A2), CD4 (BioLegend, clone: GK1.5), CD8 (BioLegend, clone:53-6.7), GZMB (Biolegend, clone: QA16A02), IFN-γ (Biolegend, clone: XMG1.2), Ki-67 (BioLegend, clone: 16A8), PD-1(Biolegend, clone:29F.1A12), PD-L1 (BioLegend, clone:10F.9G2), CD44 (Biolegend, clone: IM7), CD62L (BioLegend, clone: MEL-14), CD25(BioLegend, clone: 3C7), CD69(BioLegend, H1.2F3), BCL2 (Cell Signaling Technology, clone: 124), PRF(BioLegend, clone: S16009A). For intracellular staining, cells were given PMA/ionomycin (BioLegend, Cat. No. 423303) re-stimulation for 4–5 h, fixed, permeabilized, and then stained. FACS analysis was performed on a BD FACS Aria III flow cytometer and analyzed by FlowJo V.10 software. BD FACS Aria III flow cytometer was used to perform FACS analysis and data analysis was performed in FlowJo V.10 software.

Graphpad Prism (v8) and R language version (3.6.3) were utilized for designing figures and statistical analysis. Wilcoxon rank sum test and Student’s t tests were used to compare the expression of TMEM176B in different groups. When p values were less than 0.05, differences were considered significant.

We extracted clinical information from the TCGA, including TNM stage, pathologic stage, age, race, weight, gender and whether or not they had received radiation. Supplementary Table S1 summarizes all the information. The table manifested that the expression of TMEM176B is closely related to the TNM stage of SKCM, the pathological stage and radiation therapy.

We demonstrated the expression status of TMEM176B across various types of cancer. TMEM176B expression was considerably enhanced in some malignancies, as demonstrated in Figure 1A, which further demonstrates that heterogeneity is an important attribute of cancer and a major contributor to tumor progression. Compared to normal tissue, TMEM176B expression was higher in SKCM (p <0.001) (Figure 1B). The correlation analysis revealed that there was a statistical difference between TMEM176B and T stage, pathologic stage (p <0.05), and melanoma ulceration, radiation therapy, breslow depth (p <0.001) (Figures 1C–G).

Since the significant differential expression within tumor and normal tissues, we investigated the effect of TMEM176B expression on the prognosis of tumor patients. Cancer cases were separated into high- and low-expression groups based on TMEM176B median expression levels to evaluate how TMEM176B expression influences prognosis of patients in SKCM. The area under the curve can be derived from the ROC curve as 0.668 (Figure 2A). High TMEM176B level is linked with a better prognosis, according to the Kaplan-Meier survival analysis (Figure 2B). We also summarized in Supplementary Figures 1, 2 that the expression level of TMEM176B is closely associated with the prognosis of patients with various malignancies. We next combined clinical and pathological data, and analyzed the association between TMEM176B expression and a number of clinical parameters in SKCM patients. For OS, univariate Cox regression showed that TMEM176B played a detrimental role in patients with SKCM with the following characteristics: age >60 (n = 210, HR = 1.656, 95% CI from 1.251 to 2.192, p < 0.001), T stage 3 and 4 (n = 243, HR = 2.085, 95% CI from 1.501 to 2.895, p < 0.001), N stage 2 and 3 (n = 105, HR = 1.818, 95% CI from 1.310 to 2.523, p < 0.001), M stage 1 (n = 24, HR = 1.897, 95% CI from 1.029 to 3.496, p < 0.05), and Pathologic stage III and IV (n = 193, HR = 1.617, 95% CI from 1.207 to 2.165, p = 0.001) (Figure 2C). It is clear from this that TMEM176B plays an important role and serves as a reliable prognostic marker in SKCM.

We screened for proteins or related genes that might bind to TMEM176B in order to learn more about how it plays a role in carcinogenesis. GEPIA2 was used to identify the top 100 genes with the strongest relationship to TMEM176B in SKCM. The expression level of TMEM176B is positively linked with TMEM176A (Transmembrane Protein 176A) (R = 0.94), LAIR1 (Leukocyte Associated Immunoglobulin Like Receptor 1) (R = 0.79), SLC7A7 (Solute Carrier Family 7 Member 7) (R = 0.78), ITGB2 (Integrin Subunit Beta 2) (R = 0.77), HCK (HCK Proto-Oncogene, Src Family Tyrosine Kinase) (R = 0.77), as illustrated in Figures 3A–E. The STRING program was also used to generate a network of proteins that interact with TMEM176B (Figure 3F). We subjected the top 100 genes obtained from GEPIA2 to GO/KEGG analysis, Figures 3G,H suggest these gene enriched in “regulation of lymphocyte activation” and “phagocytosis” pathway.

As can be seen from the above, TMEM176B was highly expressed in tumor tissues of SKCM and that higher expression correlated with better prognosis of SKCM patients. A variety of immune cells with different functions are present in the tumor microenvironment, including immunosuppressive cells and cells that promote tumor immunity (Quail and Joyce, 2013). As a result, it is necessary to continually explore the relationship between TMEM176B and the infiltration of immune cells. Data from TIMER suggests statistical significance between TMEM176B expression and immune cells, with B cells, CD4⁺ T cells, CD8⁺ T cells, macrophages, neutrophils, and dendritic cells were significantly positively correlated with the expression level of TMEM176B in SKCM (Figure 4A). TMEM176B expression levels, on the other hand, were significantly adversely associated with tumor purity (Figure 4A). We can also derive the association among TMEM176B expression and other immune cells from Figure 4B. Supplementary Figures S3A–C demonstrated correlation of TMEM176B expression and immune infiltration in different cancers of TCGA except SKCM. Among them, high expression of TMEM176B is associated with a higher degree of immune infiltration in most tumors. However, in LIHC, COAD, MESO and DLBC, the correlations were negative.

Using the CIBERSORT method, we investigated TMEM176B expression profiles to assess the level of 22 different immunocytes (Figure 4C). Our findings revealed that 9 kinds of immunocytes, including CD8⁺ T cells, activated CD4 memory T cells, resting and activated NK cells, Monocytes, Macrophages, resting dendritic cells and eosinophils were influenced by TMEM176B expression, among these cells CD8⁺ T cells and macrophages were significantly different between the TMEM176B high and low groups. The results of the high expression group compared to the low expression group showed that CD8⁺ T cells increased, whereas M0 Macrophages decreased. These findings imply that TMEM176B modulates patient survival via immunological infiltrative interactions in the tumor microenvironment. CD8⁺ cytotoxic T lymphocytes (CTLs) are essential in antitumor immunity. Therefore, the impact of TMEM176B on CD8⁺ T cell infiltration level and the prognosis of tumor patients deserves close attention.

We would like to further study the regulatory relationship between TMEM176B and CD8⁺ T cells in the tumor microenvironment (TME). We investigated the role of Tmem176b in a mouse melanoma model and sought to continue to explore the role of Tmem176b in the regulation of CD8⁺ T cells in the tumor microenvironment. We constructed a mouse tumor model (Figure 5A) to figure out how Tmem176b regulates the differentiation, function, and immune infiltration levels of CD8⁺ T cells in the tumor microenvironment. After implanting the tumor subcutaneously for 14 days, we euthanized the mice for flow cytometry analysis. The tumor volume of mice following Tmem176b knockout was similarly greater and grew quicker than that of WT mice (p < 0.05) (Figure 5B).

Our findings revealed that the proportion of CD3⁺ CD8⁺ T cells in Tumor-infiltrating T lymphocytes (TILS), is dramatically reduced after Tmem176b knockout (Figures 5C,D), the same trend can be seen in CD3⁺ CD4⁺ T cells (Figure 5H). This downward trend was also found in peripheral blood, spleen, and lymph nodes, although we did not obtain statistical differences in all tissues (Supplementary Figure S4A), and this difference is more pronounced in CD8⁺ T cells, consistent with our previous results that higher expression of TMEM176B in SKCM was associated with higher levels of infiltration of immune cells (Figure 4). CD8⁺ T cells are subdivided into naive T cells (Tnaive), effector T cells (Teff) and memory T cells (Tmem) subsets. In our study, compared to wild type mice, more CD8⁺ Tnaive and less CD8⁺ Teff presented in tumor of Tmem176b ^−/− mice (Figures 5E,F), and the same change had also been observed in tumor-infiltrated CD4⁺ T cells (Figure 5H). However, Tmem remained unchanged between groups (Figures 5G,H). The influence of Tmem176b on the differentiation of T cells was not significant in other non-tumor organs (Supplementary Figures S4B,C). It suggested that T cell populations were normal in non-tumor organs and T cell homeostasis was altered in response to immunogenic stress exhibited by the facts that increased Tnaive cell number and decreased Teff cell number were presented in the local tumor of Tmem176b ^−/− mice.

We also looked at the expression of some immunosuppressive molecules following tumor transplantation. Comparing to WT mice, our findings revealed that the expression of multiple activation or effector function-related molecules were inhibited in tumor-infiltrated CD8⁺ T cells of Tmem176b ^−/− mice (Figure 6) although the trend was not consistent in blood, spleen or lymph node (Supplementary Figures S4D, S5). Our findings showed that after tumor implantation, knocking down Tmem176b has effect on T cell activation rather than anti-apoptosis, or proliferation (Supplementary Figure S5).

We could conclude from Figure 6 that the expression levels of cytotoxic molecules such as IFN-γ, PRF1, and GZMB in tumor-infiltrating CD8⁺ T cells was reduced following Tmem176b deletion, indicating that the function of tumor-infiltrating T cells was significantly compromised. It is the possible reason for the larger tumor size in Tmem176b knockout mice. To verify our results, we researched the correlation between the expression of TMEM176B and some molecules in SKCM. As shown in Supplementary Figure S6, the expression of TMEM176B was significantly positive related to the expression levels of PD-L1 (CD274), CTLA4, LAG3, GZMK, GZMB, GZMA, GZMM, PRF1, TNF and IFNG in SKCM. PD1 is a hallmark of T-cell exhaustion related to co-expression of other checkpoint molecules such as LAG-3 and TIM-3, as well as an activation marker for T-cells including antigen specific responses. The results of this part indicate that Tmem176b positively regulates the effector function and activation of tumor-infiltrating CD8⁺ T cells.