This article is mentioned in:

Introduction

Hepatocellular carcinoma (HCC) is the sixth most common malignant tumor globally and >700,000 people die from this disease every year (1). Patients with HCC do not present with obvious symptoms until advanced stages of the disease, making treatment difficult and ineffective (2). Hence, it is vital to identify some potential biomarkers for the diagnosis or prognosis of HCC.

The extracellular environment of solid tumors is acidic, with a pH between 6.5 and 6.9, whereas normal tissues is alkaline, with a pH between 7.2 and 7.5 (3). Tissue acidosis can result in tumor progression (4). Therefore, interventions targeting the acidic microenvironment of tumors may provide new therapeutic opportunities. However, the mechanism of action behind the generation of the acidic microenvironment in tumorigenesis is still in the initial stage, with further research urgently needed.

In order to maintain the stability of the pH, tumor cells transport acidic substances to the outside or transport extracellular basic substances into the cells through a series of transport proteins, such as the H+-ATPase and the Na+/H+ exchanger 1 (5). The transmembrane protein (TMEM) family contains proteins that span the entire width of the lipid bilayer and are permanently immobilized (6). A number of TMEMs act as channels, allowing specific substances to be transported between the intracellular and extracellular environment (7). However, the functions of TMEMs are unclear, and investigations into their functions are urgently required.

TMEMs are present in various cell types and are involved in a number of important physiological processes. For example, TMEM16A acts as a calcium-activated chloride channel (8,9), whereas TMEM132A may be involved in brain development during the embryonic stages (10). Furthermore, studies have confirmed that TMEMs may serve an important role in tumor growth and development (11,12). For instance, TMEM88 binds to disheveled proteins and promotes the invasion and metastasis of non-small cell lung cancer by activating the p38/glycogen synthase kinase 3β/Snail pathway (12).

Recently, Yang et al demonstrated that one type of CI− channels was activated by an acidic extracellular pH, and it participated in the proton-activated CI− (PAC) currents; this channel protein was confirmed to be TMEM206, also termed PAC or pacc1 (13). However, little is known about TMEM206. Based on the results of the aforementioned study, it may be speculated that TMEM206 also serve a fundamental role in tumorigenesis and progression. To improve the understanding of TMEM206, the present study aimed to examine the expression profiling and prognostic values of TMEM206 in human cancers using multiple public databases, and potential co-expression genes that may be associated with the dysregulation of TMEM206 in hepatocellular carcinoma (HCC) were explored, which may be beneficial for the further study of TMEM206.

Materials and methods

Expression profiling of TMEM206 in human cancers

Gene expression profiling interactive analysis (GEPIA, http://gepia.cancer-pku.cn) and Oncomine (http://www.oncomine.org) databases were used to explore the potential features of TMEM206 in human tumors and normal tissues. GEPIA software is a web tool that analyzes the RNA expression data from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) projects (https://www.cancer.gov/) (14). The Oncomine database is a web server used for bioinformatics services containing 715 datasets that provides a large-scale, high quality and consistent analytical method for gene expression profile analysis (15).

Pan-cancer survival analysis

The Kaplan-Meier plotter (http://kmplot.com//analysis) (16) was used to assess the prognostic roles of TMEM206 in 10,461 cancer samples. The forest plot was constructed using the R programming language (17). The Human Protein Atlas (HPA) dataset (https://www.proteinatlas.org/) (18,19) is based on TCGA and was used to analyze the association between TMEM206 RNA and protein expression and overall survival rate of patients with HCC. Additionally, the UCSC Xena browser (http://xena.ucsc.edu) (20) was used to evaluate the association between TMEM206 mRNA expression levels and the prognosis of patients with HCC. Due to the crossing of survival curves, Simon's two-stage test was used rather than the log-rank test (21,22).

Potential transcription regulatory mechanisms of TMEM206

The cBioPortal for Cancer Genomics (http://www.cbioportal.org) was used as a web resource to analyze and visualize the genomics data. The cBioPortal was also used to obtain the potential co-expression genes of TMEM206 in HCC (|Spearman's correlation coefficient |>0.4) (23). Additionally, the potential co-expression genes of TMEM206 were extracted from Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/) datasets GSE36376 and GSE76427, separately, which have previously been used to analyze gene expression for tumor and adjacent non-tumor tissues in HCC (24,25). To identify the genes co-expressed with TMEM206, a Spearman's correlation test was used by the R platform (|Spearman's correlation coefficient|>0.4) (26). Finally, using online Venn software (http://bioinformatics.psb.ugent.be/webtools/Venn/), the intersection of these gene lists was determined as the co-expression genes of TMEM206 in HCC.

The miRWalk (27,28), miRDB (29) and TargetScan (30) databases were used to identify potential interacting microRNAs (miRNAs/miRs) (TargetScan, context++ score percentile >80; miRDB, score >80 and miRWalk: Score >0.8). The shared miRNAs among the three sets of prediction results were regarded as potential miRNAs that target TMEM206 mRNA.

Functional annotation of a co-expression gene network of TMEM206

Kyoto Encyclopedia of Genes and Genomes (KEGG) (31) and gene ontology (GO) (32,33) are commonly used for the functional annotations of genes. MetaScape (https://metascape.org/gp/index.html#/main/step1) is an integrated analytics platform that integrates multiple annotation datasets. The functional enrichment analysis was performed using these platforms to analyze the function of TMEM206 co-expression genes (34).

Results

Expression of TMEM206 in cancer

The distribution of TMEM206 expression in various tissues of the human body based on TCGA and GTEx projects was obtained from the GEPIA website (Fig. 1A). TMEM206 was observed to be abundant in human tissues, especially in the brain and blood (Fig. 1B). Subsequently, the differential expression of TMEM206 in human tumor types and normal control tissues was investigated using the Oncomine database; the results demonstrated that 30 out of 150 tumors exhibited TMEM206 upregulation, and eight exhibited downregulation compared with that in normal tissues (Fig. 1C). Although, TMEM206 was abundant in normal and malignant tissues, significant increases in TMEM206 expression level were observed in colorectal, breast and liver cancer and lymphoma, and decreased levels were observed in kidney cancer compared with those in the corresponding normal tissues.

Figure 1. Pan-cancer expression profiling analysis of TMEM206. (A) Body map of TMEM206 in GEPIA; a darker color indicates a higher gene expression level. Red represents the expression of TMEM206 in tumor specimens, and green represents that in normal specimens. (B) Bar chart of the number of studies in which TMEM206 expression was significantly different in normal tissues compared with tumor tissues. (C) Boxplot of the pan-cancer expression profiling of TMEM206 in human cancers. Cell color is determined by the best gene rank percentile for the analyses within the cell; an analysis may be counted in more than one type of cancer. The color refers to the tumor (red) or normal tissue (blue). GEPIA, Gene expression profiling interactive analysis; TMEM, transmembrane protein.

Pan-cancer survival analysis of TMEM206 mRNA expression suggests a prognostic role of TMEM206 in HCC

The Kaplan-Meier plotter datasets were used for pan-cancer survival analysis. The results demonstrated that TMEM206 exhibited notable prognostic significance in 5 of the 21 analyzed types of tumors (Table SI). Heterogeneity among various types of cancer was observed (Fig. 2). A higher expression level of TMEM206 was associated with a poor prognosis of head and neck squamous cell carcinoma, renal papillary cell carcinoma, HCC and sarcoma, while a lower expression level of TMEM206 was associated with a poor prognosis of thymoma (Fig. 2). Subsequently using the HPA dataset, which is based on experimental data, it was found that high protein expression level of TMEM206 was related to a poor survival rate of HCC (Fig. 3). Therefore, the present study focused on TMEM206 in HCC for the subsequent investigation.

Figure 2. Forest plot of the pan-cancer survival analysis.

Figure 3. The association between the survival rate (%) and transmembrane protein 206 protein expression levels. Data were analyzed using a Simon's two-stage test.

The UCSC Xena browser was used to evaluate the association between the DNA variants and upregulation of TMEM206 in HCC. Expression was colored red to blue for high to low expression (Fig. 4A). Survival analysis using the Simon's two-stage test (low expression, n=212 and high expression, n=206, respectively) demonstrated that high expression of TMEM206 mRNA was an unfavorable prognostic marker for patients with HCC (Fig. 4B). However, the methylation status and copy number variant of TMEM206 DNA were not associated with survival in HCC (Fig. 4C and D). As there was no significant difference, the Simon's two-stage test was not used inspite of the crossover of survival curves.

Figure 4. Association between the expression level, methylation status and copy number variant of TMEM206 and the prognosis of patients with HCC. (A) The heat map of the expression level, methylation status and copy number variant of TMEM206. (B-D) Kaplan Meier plots of the association between the (B) expression level, (C) methylation status or (D) copy number variant of TMEM206 and the prognosis of patients with HCC were analyzed by a two-stage test. HCC, hepatocellular carcinoma; TMEM, transmembrane protein.

Functional annotation of the TMEM206 co-expression gene network

Co-expression gene analysis was used to further investigate the possible functions of TMEM206 in HCC. Data were acquired from cBioPortal and the GEO datasets GSE36376 and GSE76427 (|Spearman's correlation coefficient |>0.4|). The intersections of the three gene lists in the Venn diagram were considered to be potential co-expression genes for TMEM206 (Fig. 5A; Table SII). MetaScape was used to analyze functional gene enrichment; the most significantly enriched GO term was the ‘U2-type spliceosomal complex’, which suggested that may regulate the expression of TMEM206 and the co-expression genes associated with it (Fig. 5B).

Figure 5. Venn plot and Metascape enrichment results of co-expression genes for TMEM206. (A) The intersection of gene lists (TCGA, GSE76427 and GSE36376 HCC) identified as co-expressed genes for TMEM206. (B) The enriched terms of the co-expressed genes for TMEM206. TCGA, The Cancer Genome Atlas; HCC, hepatocellular carcinoma; TMEM, transmembrane protein; COPI, coat protein complex I.

In addition, the miRWalk, miRDB and TargetScan databases were used to identify the potential miRNAs involved in the regulation of TMEM206. With strict screening criteria, two microRNAs, hsa-miR-325 and hsa-miR-510-5p (also termed has-miR-510) that may bind to TMEM206 mRNA were identified (Fig. 6A). Kaplan-Meier survival analysis demonstrated that hsa-miR-325 and hsa-miR-510-5p were associated with the prognosis of patients with HCC. High expression levels of hsa-miR-325 and hsa-miR-510-5p were favorable for overall survival (P=2.3×10−10 and P=7.2×10−10, respectively; Fig. 6B and C), suggesting that hsa-miR-325 and hsa-miR-510-5p expression levels were negatively associated with TMEM206 in HCC.

Figure 6. Potential microRNAs involved in the regulation of TMEM206. (A) Potential microRNAs involved in the regulation of TMEM206. (B and C) Association between (B) hsa-miR-325 and (C) hsa-miR-510-5p and overall survival time in patients with hepatocellular carcinoma (P=2.3×10-10, P=7.2 ×10-10; respectively). miR, microRNA; TMEM, transmembrane protein.

Discussion

Numerous studies have demonstrated that TMEM expression is up- or downregulated in tumor tissues compared with that in adjacent healthy or benign tissues (35,36). For example, TMEM97 acts as prognostic biomarker for non-small cell lung cancer (35). Furthermore, a previous study has reported that TMEM proteins are involved in tumorigenesis and cancer development, such as TMEM45A, which participates in the proliferation and invasion of ovarian cancer cells (37). Therefore, an improved understanding of the TMEM family may help identify their functions in various types of cancer and improve therapeutic strategies.

TMEM206 is a newly identified transmembrane protein, Yang et al (13) performed unbiased RNA interference screening and demonstrated that TMEM206 was essential for the widely observed PAC currents (ICl,H), which are involved in acid-sensing ion channels. Little is known about the role of TMEM206 in tumors. Zhao et al (38) reported that TMEM206 mRNA and protein expression levels were higher in colorectal cancer (CRC) tissues compared with adjacent non-tumor tissues, overexpression of TMEM206 promoted the proliferation and invasion of CRC cells and positively regulated the levels of phospho-AKT and its downstream signaling pathway components, suggesting that TMEM206 promoted the development and progression of CRC by enhancing the interactions between the AKT and ERK signaling pathways.

A previous study has reported that the TMEM family is widely expressed in mammalian cells (39), which is in accordance with the present study. In the present study, the results of the expression profiling analysis demonstrated that TMEM206 expression was significantly increased in colorectal cancer, breast cancer, lymphoma and liver cancer, and reduced in the kidney cancer datasets. In terms of a prognostic role for TMEM206, the present study demonstrated that TMEM206 was a significant prognostic factor for certain types of tumor and served either protective or unfavorable roles depending on the tumor type. These contrary results make the role of TMEM206 dubious.

Further analysis of data from patients with HCC demonstrated that the upregulation of TMEM206 was an unfavorable prognostic factor for HCC. TMEM206 was associated with the overall survival probability and 5-year survival rate of patients with liver cancer in The Kaplan Meier plotter and HPA. In addition, in the present study, the UCSC Xena browser was used to evaluate the association between the DNA variants and upregulation of TMEM206 in HCC. However, no significant associations were observed between the copy number variant or the extent of methylation of TMEM206 and the prognosis of patients with HCC. GEO and cBioPortal were used to determine genes co-expressed with TMEM206 and 31 genes were discovered. Using MetaScape, the signalling pathways that co-expression genes of TMEM206 may participate in were identified and the term U2-type spliceosomal complex was most significant. Finally, the present study identified two potential miRNAs, hsa-miR-325 and hsa-miR-510-5p, that may target TMEM206.

Hsa-miR-325 and hsa-miR-510-5p participate in tumor development (40–43). Hsa-miR-325 has been identified as a potential biomarker in human bladder cancer and low expression of hsa-miR-325 was associated with poor overall survival in patients with bladder cancer (40). In addition, downregulation of hsa-miR-325 has been demonstrated to promote the progression of non-small cell lung cancer and HCC (41,42). Chen et al (43) found that hsa-miR-510-5p acted as a tumor suppressor in renal cell carcinoma by reducing cell proliferation, migration and inducing apoptosis.

In conclusion, the results of the present study demonstrated that upregulation of TMEM206 is a potential prognostic indicator for HCC. However, studies into the role of TMEM206 are still at the early stage, and further research is required to clarify its role in HCC and other types of tumors.

Supplementary Material

Supporting Data

Supporting Data

Acknowledgements

Not applicable.

Funding

This study was supported by the National Natural Science Foundation of China (grant no. 81860276).

Availability of data and materials

The datasets used/and or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

LZ, SYL, XY, YQW and YXC contributed toward data analysis, drafting and revising the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References