Upregulation of ST3GAL5 predicts suppression of malignant biological behavior in bladder cancer

Abstract

Background: Bladder cancer (BLCA) is the most prevalent malignancy in the urinary tract system, while ST3GAL5 is a protein-coding gene that catalyzes the formation of ganglioside monosialodihexosylganglioside 3 (GM3) synthase. GM3 synthase has been reported to significantly influence the malignant process of various cancers. However, the expression profile and functional role of ST3GAL5 specifucally in BLCA remain to be elucidated.

Methods: In order to investigate the relationship between ST3GAL5 expression and malignant biological behavior and prognosis in BLCA. The mRNA expression of ST3GAL5 and clinicopathological characteristics in BLCA were firstly evaluated by public databases. Next, immunohistochemical staining was performed to analyze the protein expression of ST3GAL5 in BLCA and paracancerous tissues, as well as the expression of various types of malignant biological behavior. Subsequently, the gene set enrichment analyses (GSEA) were performed for ST3GAL5 and all correlated genes in BLCA by sorting Pearson's Correlation Coefficient. GSEA was also used to validate the pathways affected by the different expression levels of ST3GAL5 in BLCA patients.

Results: The mRNA and protein expression of ST3GAL5 in BLCA were significantly higher in low-grade and non-muscle-invasive BLCA (p < 0.05). The results from bioinformatics databases indicated that upregulation of ST3GAL5 have a lower grade, lower pathological stage, less susceptibility to lymphatic metastasis, and lower mortality rates. Kaplan–Meier survival analysis demonstrated that upregulation of ST3GAL5 was associated with better survival in BLCA (p < 0.05).

Conclusion: Upregulation of ST3GAL5 may be related to tumor suppression in BLCA, and may be a potential prognostic and therapeutic marker for BLCA.

Graphical abstract

Keywords

bladder cancer / immunohistochemistry / malignant biological behavior / ST3GAL5

1 INTRODUCTION

Bladder cancer (BLCA) is the most common malignancy of the urinary tract system[1], with high morbidity and mortality rates and about 3.4 million people worldwide were affected in 2015[2]. In the United States, 80 470 new cases of BLCA and 17 670 cases of BLCA-associated deaths were expected in 2019[1]. Based on the histological grading[3], BLCA can be classified as high-grade and low-grade. Low-grade BLCA rarely invades the bladder muscle, recurs or metastasis; however, most BLCA-associated deaths are attributed to high-grade disease. BLCA can also be divided into muscle-invasive BLCA and non-muscle-invasive BLCA according to its invasiveness[4]. Compared to non-muscle-invasive BLCA, muscle-invasive disease has a poorer prognosis and a higher cancer-specific mortality rate. Although early diagnosis and treatment of BLCA can significantly improve patient survival and prognosis status, the mortality associated with BLCA remains high. Therefore, there is a current need to explore a novel clinical and therapeutic biomarker to improve the diagnosis and treatment of BLCA.

ST3GAL5 is a protein-coding gene that catalyzes the formation of ganglioside monosialodihexosylganglioside 3 (GM3) synthase[5]. The ganglioside GM3 is known to be involved in the induction of cell differentiation, regulation of cell proliferation, maintenance of fibroblast morphology, signal transduction, and integrin-mediated cell proliferation[6]. In addition, GM3 synthase is associated with a variety of cancers, including brain cancer, lung cancer, and melanomas, and has been reported to significantly influence the malignant process of cancer[6–8]. However, the expression profile and functional role of ST3GAL5 in BLCA remain unclear. The expression of ST3GAL5 in BLCA and its functional role have been less studied. In a previous study[9], we have confirmed from a comprehensive bioinformatics analysis through publicly available databases that the downregulation of ST3GAL5 at the mRNA level is associated with invasiveness, grade, and prognosis in BLCA, but no validation at the protein level.

In this study, the protein expression of ST3GAL5 in BLCA was investigated via immunohistochemical staining. In addition, the relationship between the clinicopathological features and gene expression of ST3GAL5 in BLCA were analyzed through publicly bioinformatical databases. Furthermore, Gene Set Enrichment Analyses (GSEA) were performed on genes that were co-expressed with ST3GAL5 BLCA, and GSEA verification on the different expression level of ST3GAL5 in BLCA were investigated again.

2 MATERIALS AND METHODS

2.1 Data sources

The mRNA expression of ST3GAL5 in BLCA was obtained from The Cancer Genome Atlas (TCGA)[10] and Gene Expression Omnibus (GEO)[11]. The data from the GEO accession number GSE13507[12] and GSE120736 were downloaded to analyze the mRNA expression of ST3GAL5. Data from TCGA and GTEx databases were downloaded using the UCSC Xena browser tool[13]. The protein expression of ST3GAL5 in BLCA was obtained from patients with histologically confirmed bladder cancer in Sun Yat-sen University Cancer Center from May 2019 to April 2020.

2.2 Immunohistochemistry (IHC) validation

Patients with confirmed bladder cancer were selected to evaluate IHC on their cancer tissues and paracancerous tissues to verify the protein expression of ST3GAL5 between BLCA and normal bladder tissues or different malignant biological types of BLCA. IHC was performed on paraffin sections using an EnVision method, ST3GAL5 Rabbit Polyclonal antibody (Catalog number: 14614-1-AP, Proteintech Group, Inc, Wuhan, China) was selected at dilution of 1:50 (under 40× lens). Cancer staging was assessed using the 8th edition of the American Joint Committee on Cancer (AJCC) cancer staging system[4]. Histological grade was evaluated according to the 2016 World Health Organization (WHO) classification[3]. All patients have obtained approval from the ethics committee (Approval number: KJ2022-062-01).

2.3 Histological explanation

The positive staining was defined as cytoplasm staining of yellow or dark brown. Non-tumor cells (such as lymphocytes, stromal cells, endothelial cells, and urothelial cells) were served as internal positive controls for all tissue sections. The semi-quantitative grading criteria[14] for IHC are as follows: (−) indicated negative staining in tumor cells; (±) < 20% tumor cells were positive staining; (+) ≥ 20% but < 50% tumor cells were positive staining; (++) ≥ 50% but < 75% tumor cells were positive staining; (+++) ≥ 75% tumor cells were positive staining. In addition, we defined tumor cells < 20% staining as reduced or loss expression, and otherwise defined as positive.

2.4 Clinicopathological characteristics and survival analysis

The clinicopathological characteristics and survival analysis were firstly applied with the GEO accession GSE13507 data set to explore the relationship between different ST3GAL5 expressions and clinical outcomes of BLCA patients, respectively, and then further validated with the TCGA-BLCA data set.

2.5 Enrichment analyses

First, the Pearson Correlation Coefficient (PCC) of ST3GAL5 and all genes in bladder cancer from TCGA-BLCA data set were calculated using the cor.test function of the R language, sort all the genes according to the PCC value. Next, the GSEA of the Gene Ontology (GO, c5.all.v7.1.entrez.gmt) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG, c2.cp.kegg.v7.0.entrez.gmt) signaling pathway were performed using the R package “clusterProfiler” v3.17.0[15]. In addition, the ST3GAL5 expression value from the TCGA-BLCA data set was divided into high- and low-group based on the mean expression level, then GSEA of the KEGG signaling pathway was performed again to verify signal pathways using the GSEA software v4.0.3[16]. Gene set permutations were performed 1000 times for each analysis. The false discovery rate < 0.25, |normalized enrichment score| > 1, and p < 0.05 were considered significant differences.

2.6 Statistical analysis

All statistical analyses were performed under R v4.01 and R-studio v1.3.959[17]. The box plots and scatter plots were constructed using the R package “ggplot2” v3.3.2[18]. χ² test and Fisher's exact were used to compare the categorical variables. Student's t-test was used to compare the means of two independent samples. Kaplan–Meier analysis and log-rank test was performed by using R package “survival” v3.2.3 and “survminer” v0.4.7. Multivariate Cox proportional hazards regression model was conducted to adjust for covariate effects and stratified analysis. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated. Missing data were excluded from the analysis. p < 0.05 was considered statistically significant.

3 RESULTS

3.1 The mRNA expression of ST3GAL5 in patients with BLCA

Two independent BLCA data sets (GSE120736 and GSE13507) from the GEO database were analyzed for the mRNA expression level of ST3GAL5, the results both revealed that high expression of ST3GAL5 was significantly positively associated with low-grade BLCA and non-muscle invasive bladder cancer (Figure 1A,B).

3.2 The protein expression of ST3GAL5 in patients with BLCA

A total of nine patients with BLCA were selected in the present study, of the nine cases of bladder cancer, eight were male and one was female, with a mean age of 62.3 years. Among them, seven cases of high grade, two cases of low grade, five patients were muscle-invasive bladder cancer, four patients were non-muscle invasive bladder cancer, one case of stage Ta, four cases of stage T1, three cases of stage T2, and one case of stage T4 (see Table 1 for details). According to the immunohistochemical results, the protein expression of ST3GAL5 was strongly positive staining in normal bladder tissue, moderately positive staining in atypical hyperplasia bladder tissue, and significantly reduced in bladder urothelial carcinoma tissue (Figure 2A–C). Similarly, for the different grades of bladder cancer, ST3GAL5 protein expression was moderate positive staining in low-grade bladder cancer and weakly positive staining or reduced expression in high-grade bladder cancer tissue (Figure 2D–F). Similar results were also seen in the muscle-invasive bladder, where ST3GAL5 protein expression was moderate positive staining in non-muscle-invasive bladder cancer and weakly positive staining or reduced expression in muscle-invasive bladder cancer tissue (Figure 2G–I).

3.3 The relationship between different ST3GAL5 expression and clinicopathological characteristics of patients with BLCA

This study investigated the relationship between the different ST3GAL5 mRNA expression and the clinicopathological characteristics of patients with BLCA from GSE13507 data set. Then, the clinical features were re-validated using the TCGA-BLCA data set. Compared with high expression of ST3GAL5, the low expression of ST3GAL5 was significant differences in stage, grade, lymph node metastasis and vital status (Table 2, p < 0.05). Similarly, the same results are found in the TCGA-BLCA data set (Table 3, p < 0.05). These results indicated that lower expression levels of STGAL5 have a higher grade, higher pathological stage, greater susceptibility to lymphatic metastasis and higher mortality rates.

3.4 The relationship between ST3GAL5 expression and survival prognosis in patients with BLCA

We first performed a Kaplan-Meier survival analysis on patients with BLCA using the GES13507 data set, the result indicted that low expression of ST3GAL5 were significantly negatively associated with overall survival (OS) and disease-specific survival (DSS) (Figure 3A,B). Then, TCGA-BLCA data set was used to verify the prognosis results. Similarly, the results showed that low expression of ST3GAL5 were both negatively associated with OS and DSS in TCGA-BLCA data set (Figure 3C,D). These results all indicated the adverse prognostic relevance of ST3GAL5 expression in patients with BLCA.

3.5 The GSEA with ST3GAL5 expression in BLCA

Firstly, a batch correlation analysis of 20 530 genes from 411 case of BLCA patients in TCG-BLCA data set were performed using Pearson's method, and after excluding the invalid and duplicate genes, finally 17 541 correlated genes were selected for GSEA analysis after sorting them by PCC value. Subsequently, GO functional and KEGG pathway enrichment analyses were performed with ST3GAL5 and all the correlated genes, in which GO analyzes functions in biological processes, molecular functions, and cellular components. GO function enrichment analysis results clustered 884 items, of which 119 were activated and 765 were suppressed, and predominantly activated in “cofactor binding,” “small molecule catabolic process,” “cellular glucuronidation,” “phosphatidylglycerol acyl chain remodeling,” “positive regulation of glycogen metabolic process,” “phospholipid metabolic process,” “protein targeting,” “short chain fatty acid metabolic process,” “cofactor metabolic process” and “Acyl-CoA ligase activity,” and suppressed in “cartilage development,” “peptidyl threonine modification,” “ADP metabolic process,” “cell adhesion mediated by integrin,” “cellular response to ionizing radiation,” “cerebral cortex development,” “chromatin assembly or disassembly,” “collagen binding,” “formation of primary germ layer,” and “kinetochore” (Figure 4A).

Furthermore, the KEGG pathways analysis for ST3GAL5 and its correlateded genes results clustered 74 items, of which 21 were activated and 50 were suppressed, which mainly activated in “peroxisome,” “drug metabolism cytochrome P450,” “metabolism of xenobiotics by cytochrome P450,” “PPAR signaling pathway,” “ribosome,” “steroid hormone biosynthesis,” “valine leucine and isoleucine degradation,” “glycerophospholipid metabolism,” “retinol metabolism,” and “fatty acid metabolism,” and suppressed in “mismatch repair,” “homologous recombination,” “pyrimidine metabolism,” “base excision repair,” “ECM receptor interaction,” “allograft rejection,” “prion diseases,” “glioma,” “graft versus host disease,” and “autoimmune thyroid disease” (Figure 4B).

Finally, in order to verify the KEGG signal pathway, we next divided the ST3GAL5 expression level into high- and low-groups according to the mean value of ST3GAL5 expression from TCGA-BLCA data set, and then GSEA analysis was performed again.

The results showed that 13 signaling pathways were significantly clustered, of which 11 items were activated and 2 items suppressed, and all items were included in the enrichment analysis for correlation analysis. The results demonstrated the KEGG activated in “ascorbate and aldarate metabolism,” “steroid hormone biosynthesis,” “arachidonic acid metabolism,” “retinol metabolism,” “drug metabolism cytochrome P450,” “metabolism of xenobiotics by cytochrome P450,” “linoleic acid metabolism,” “pentose and glucuronate interconversions,” “starch and sucrose metabolism,” “fatty acid metabolism,” and “glycosylphosphatidylinositol anchor biosynthesis,” and suppressed in “nucleotide excision repair,” and “pancreatic cancer” (Table 4).

4 DISCUSSION

About 75% of primary BLCAs are superficial tumors[19] that can be easily treated with transurethral resection of bladder tumors (TURBT), but in the about 70% of these treated patients still relapsed after a period of time[20], and of these patients, about 30% progressed to high-grade and higher stage disease[20]. Patients with BLCA to higher stages have a poor prognosis outcome for recurrence and progression[21]. Although lots of new tumor biomarkers have been proposed, they are limited in predicting the prognosis and treatment of BLCA. Therefore, novel biomarkers are urgently needed to predict the prognosis and evaluation of BLCA.

At present, the expression of ST3GAL5 and its clinicopathological characteristics in patients with BLCA have less been comprehensive investigated. In a previous study[9], we confirmed from a comprehensive bioinformatics analysis that downregulation of ST3GAL5 from mRNA levels is associated with muscle-invasiveness, high-grade, and poor-prognosis in BLCA, but lack of protein level validation. In this study, we first explored the difference in the expression of ST3GAL5 from the mRNA level in different malignant biological behaviors of bladder cancer through two independent BLCA data sets. The results were consistent with the previous study. Then immunohistochemistry was performed on nine cases of histologically confirmed BLCA patients with different grades and stages. Experimental validation of the results revealed that bladder cancer patients with up-regulation of ST3GAL5 in patients with BLCA have significantly lower histological grade and fewer muscle invasion.

Next, a clinicopathological analysis using data from TCGA-BLCA data set was conducted. The results indicated that up-regulation of ST3GAL5 may be associated with lower grade, lower stage, less lymph node metastasis and less distant metastasis, Following, we explored the relationship of survival analysis between the different ST3GAL5 expression in BLCA. Firstly, the Kaplan–Meier analysis result of OS and DSS using data from the GEO accession GSE13507 data set suggested that high ST3GAL5 expression had a better prognosis in BLCA. Meanwhile, Kaplan-Meier analysis using the data from TCGA-BLCA data set was performed for validation again. The results also showed similar result. Although crossover was found in the results of the Kaplan-Meier analysis, the crossover points were all over 100 months. Study have shown that patients with BLCA who undergo cystectomy have a 5%–15% probability of pelvic recurrence, which usually occurs during the first 24 months after surgery[22], and patients usually have a poor prognosis after pelvic recurrence, even with treatment, median survival ranged from 4 months to 8 months after diagnosis[23]. The median survival time for patients with progressive disease receiving platinum-based chemotherapy is 9 months to 26 months[24]. In addition, for patients with upper urinary tract recurrence, the median survival time was 10 months to 55 months[23], and in this study, the crossover point is greater than eight years and, therefore, remains significantly different in relation to clinical outcomes. Taken together, these findings indicated that ST3GAL5 may be considered as a cancer suppressor gene in BLCA, and may therefore suppress the progression to high grade and high stage of BLCA. However, the mechanism of the role of ST3GAL5 in the occurrence and prognosis of BLCA needs to be validated by more basic experimental researches.

ST3GAL5 is an encoding gene that catalyzes the formation of the ganglioside GM3 synthase. GM3 synthase is a kind of glycosyltransferase, and changes in the expression of this protein can affect the level of glycosylation of downstream proteins, thereby affecting biological function[25]. ST3GAL5 is found up-regulated in various cancers, such as in brain, lung and melanoma, it may serve as a cancer-associated carbohydrate antigen in cancer immunotherapy[7,8]. In addition, ST3GAL5 can also suppress cancer cell proliferation by angiogenesis suppression or decreasing cell motility[7]. ST3GAL5 has been reported to have anti-proliferative effects on a variety of cancers, including colon[26], breast[27,28], liver[29], and other types of cancer[7,8]. Although some studies[30,31] have shown that ST3GAL5 has an anti-cancer effect on human bladder cancer, its mechanism of action is still unclear. In addition, it has been reported that ST3GAL5 effects may be related to tumor cell apoptosis and angiogenesis inhibition. However, the expression profile and functional role of ST3GAL5 in BLCA are unclear.

In this study, the GO functional and KEGG signaling pathway of ST3GAL5 in BLCA was investigated by using data from TCGA-BLCA data set. First, all genes expressed in BLCA were analyzed by batch correlation analysis, and the correlation coefficient related to ST3GAL5 expression was calculated. Then, sorted all the genes according to PCC value, and finally gene set enrichment analysis was performed using R package “clusterProfiler.” In the results from GO functional enrichment analysis, the most significantly activated enrichment numbers associated with tumors were protein targeting, cofactor binding, and cofactor metabolic processes, the most significantly suppressed enrichment numbers associated with tumors were collagen binding, chromatin assembly or disassembly, and ADP metabolic process.

Subsequently, GSEA between high- and low-expression of ST3GAL5 in TCGA-BLCA data set was performed to verify the signal pathway. All KEGG signal pathway gene set enrichment analysis of co-expressed genes was verified in the results of GSEA of different expression levels of ST3GAL5 in TCGA-BLCA data set.

5 CONCLUSION

In the present study, we validated the expression of ST3GAL5 in various malignant biological behaviors of BLCA from mRNA and protein expression levels, and analyzed the clinical, pathological characteristics and prognosis of ST3GAL5 expression in BLCA from public available database. All results suggest that up-regulation of ST3GAL5 may inhibit the malignant biological behavior of BLCA progression. This funding provides a new direction to explore the clinical prognosis biomarkers for patients with BLCA.

References

[1] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69:7–34.

[2] Vos T, Allen C, Arora M, Barber RM, Bhutta ZA, Brown A, et al. Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016;388:1545–1602.

[3] Compérat EM, Burger M, Gontero P, Mostafid AH, Palou J, Rouprêt M, et al. Grading of urothelial carcinoma and the new “World Health Organisation classification of tumours of the urinary system and male genital organs 2016”. Eur Urol Focus. 2019;5:457–466.

[4] Magers MJ, Lopez-Beltran A, Montironi R, Williamson SR, Kaimakliotis HZ, Cheng L. Staging of bladder cancer. Histopathology. 2019;74:112–134.

[5] Berselli P, Zava S, Sottocornola E, Milani S, Berra B, Colombo I. Human GM3 synthase: a new mRNA variant encodes an NH2-terminal extended form of the protein. Biochim Biophys Acta. 2006;1759:348–358.

[6] Inokuchi JI, Inamori KI, Kabayama K, Nagafuku M, Uemura S, Go S, et al. Biology of GM3 Ganglioside. Prog Mol Biol Transl Sci. 2018;156:151–195.

[7] Zheng C, Terreni M, Sollogoub M, Zhang Y. Ganglioside GM3 and Its Role in Cancer. Curr Med Chem. 2019;26:2933–2947.

[8] Hakomori SI, Handa K. GM3 and cancer. Glycoconj J. 2015;32:1–8.

[9] Ouyang S, Liu JH, Ni Z, Ding GF, Wang QZ. Downregulation of ST3GAL5 is associated with muscle invasion, high grade and a poor prognosis in patients with bladder cancer. Oncol Lett. 2020;20:828–840.

[10] Hutter C, Zenklusen JC. The cancer genome atlas: creating lasting value beyond its data. Cell. 2018;173:283–285.

[11] Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res. 2012;41:D991–D995.

[12] Kim WJ, Kim EJ, Kim SK, Kim YJ, Ha YS, Jeong P, et al. Predictive value of progression-related gene classifier in primary non-muscle invasive bladder cancer. Mol Cancer. 2010;9:3.

[13] Goldman MJ, Craft B, Hastie M, Repečka K, McDade F, Kamath A, et al. Visualizing and interpreting cancer genomics data via the Xena platform. Nat Biotechnol. 2020;38:675–678.

[14] Liu B, Tang LH, Liu Z, Mei M, Yu R, Dhall D, et al. α-Internexin: a novel biomarker for pancreatic neuroendocrine tumor aggressiveness. J Clin Endocrinol Metab. 2014;99:E786–E795.

[15] Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS. 2012;16:284–287.

[16] Subramanian A, Kuehn H, Gould J, Tamayo P, Mesirov JP. GSEA-P: a desktop application for Gene Set Enrichment Analysis. Bioinformatics. 2007;23:3251–3253.

[17] R Core Team. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2020.

[18] Wickham H. ggplot2: elegant graphics for data analysis. 2nd ed. New York: Springer-Verlag; 2016.

[19] Miyazaki J, Nishiyama H. Epidemiology of urothelial carcinoma. Int J Urol. 2017;24:730–734.

[20] Kaufman DS, Shipley WU, Feldman AS. Bladder cancer. Lancet. 2009;374:239–249.

[21] van der Heijden AG, Witjes JA. Recurrence, progression, and follow-up in non–muscle-invasive bladder cancer. Eur Urol Suppl. 2009;8:556–562.

[22] Huguet J. Seguimiento oncológico después de cistectomía radical basado en patrones de recidiva tumoral y sus factores de riesgo. Actas Urol Esp. 2013;37:376–382.

[23] Soukup V, Babjuk M, Bellmunt J, Dalbagni G, Giannarini G, Hakenberg OW, et al. Follow-up after surgical treatment of bladder cancer: a critical analysis of the literature. Eur Urol. 2012;62:290–302.

[24] Vrooman OP, Witjes JA. Follow-up of patients after curative bladder cancer treatment: guidelines vs. practice. Curr Opin Urol. 2010;20:437–442.

[25] Bowser LE, Young M, Wenger OK, Ammous Z, Brigatti KW, Carson VJ, et al. Recessive GM3 synthase deficiency: Natural history, biochemistry, and therapeutic frontier. Mol Gen Metab. 2019;126:475–488.

[26] Chung TW, Choi HJ, Kim SJ, Kwak CH, Song KH, Jin UH, et al. The ganglioside GM3 is associated with cisplatin-induced apoptosis in human colon cancer cells. PLoS One. 2014;9:e92786.

[27] Li Q, Sun M, Yu M, Fu Q, Jiang H, Yu G, et al. Gangliosides profiling in serum of breast cancer patient: GM3 as a potential diagnostic biomarker. Glycoconj J. 2019;36:419–428.

[28] Guthmann MD, Castro MA, Cinat G, Venier C, Koliren L, Bitton RJ, et al. Cellular and humoral immune response to N-Glycolyl-GM3 elicited by prolonged immunotherapy with an anti-idiotypic vaccine in high-risk and metastatic breast cancer patients. J Immunother. 2006;29:215–223.

[29] Cai H, Zhou H, Miao Y, Li N, Zhao L, Jia L. MiRNA expression profiles reveal the involvement of miR-26a, miR-548l and miR-34a in hepatocellular carcinoma progression through regulation of ST3GAL5. Lab Invest. 2017;97:530–542.

[30] Satoh M, Ito A, Nojiri H, Handa K, Numahata K, Ohyama C, et al. Enhanced GM3 expression, associated with decreased invasiveness, is induced by brefeldin A in bladder cancer cells. Int J Oncol. 2001;19:723–731.

[31] Wang H, Isaji T, Satoh M, Li D, Arai Y, Gu J. Antitumor effects of exogenous ganglioside GM3 on bladder cancer in an orthotopic cancer model. Urology. 2013;81:210.e11–210.e15.

RIGHTS & PERMISSIONS

2025 The Author(s). UroPrecision published by John Wiley & Sons Australia, Ltd on behalf of Higher Education Press.