In Silico Analysis for Sphingolipid Metabolism-related Genes in Human Kidney Clear Cell Carcinoma Using The Cancer Genome Atlas

Article information

Keimyung Med J. 2020;39(1):14-22
Publication date (electronic) : 2020 June 2
doi : https://doi.org/10.46308/kmj.2020.00101
1Department of Biochemistry, College of Medicine, Gachon University, Incheon, Korea
2Department of Immunology, Keimyung University School of Medicine, Daegu, Korea
3Institute of Medical Science, Keimyung University, Daegu, Korea
Corresponding Author: Shin Kim, M.D., Ph.D. Department of Immunology, Keimyung University School of Medicine, 1095 Dalgubeol-daero, Dalseo-gu, Daegu, 42601, Korea Tel: +82-53-258-7359, Fax: +82-53-258-7355 E-mail: god98005@dsmc.or.kr
Received 2020 March 25; Revised 2020 May 12; Accepted 2020 May 15.

Abstract

The sphingolipid rheostat concept states that the cellular fate is largely determined by various sphingolipid metabolites and the associated signaling pathways. Aberrant regulation of the sphingolipid metabolism-related components is closely associated with cancer survival and death, including aspects like cancer development, proliferation, progression, and response to anticancer drugs. In the present study, we investigated the expression and prognostic significance of the sphingolipid metabolism-related genes in clear cell renal cell carcinoma (ccRCC), the most common pathological subtype of kidney cancer, using an RNA-sequencing dataset of The Cancer Genome Atlas Kidney Clear Cell Carcinoma (TCGA KIRC) cohort. Expression levels of various sphingolipid metabolism-related genes were significantly altered in ccRCC tissues compared with those of normal solid tissues. Notably, the expression of B4GALNT1, BNIP3, DEGS1, GAL3ST1, S1PR4, SLC26A10, SMPDL3A, and SPHK1 was significantly upregulated, whereas the expression of B4GALT6, HPGD, LPAR1, SFTPB, ST6GALNAC5, and UGT8 was significantly downregulated in ccRCC tissues. Notably, among these significantly-altered sphingolipid metabolism-related genes, the Kaplan-Meier survival analyses showed that high expression levels of B4GALNT1, SLC26A10, and SPHK1 were associated with a poor prognosis of patients with ccRCC, whereas high expression levels of BNIP3, HPGD, and SMPDL3A were associated with a better prognosis. Taken together, our study suggests that B4GALNT1, SLC26A10, SPHK1, BNIP3, HPGD, and SMPDL3A may be novel prognostic biomarkers and targets for a therapeutic strategy to improve the treatment of ccRCC.

Introduction

Renal cell carcinoma (RCC), derived from the kidney parenchyma accounts for almost 90% of all kidney tumors [1]. According to histology of the RCC subtypes recommended by the World Health Organization, RCC is divided into three main subtypes, including clear cell renal cell carcinoma (ccRCC), papillary renal cell carcinoma, and chromophobe renal cell carcinoma, among which ccRCC is the most common subtype (80%–90%) [2]. Despite an adequate surgical resection, recurrence and metastasis occur in 25% of the patients due to the aggressiveness of ccRCC [3]. To improve this clinical limitation, there has been extensive research to find potential biomarkers that can predict the prognosis for patients with ccRCC using genomic profiling approaches such as microarray analysis, mutation analysis, and gene expression analysis [4-6]. Moreover, recent advances in biotechnology such as RNA sequencing (RNAseq) have revolutionized various fields of research, providing new insight into complex diseases such as cancer with respect to identification of the causes, establishment of targets for therapeutic strategies, and exploration of new prognostic biomarkers.

The sphingolipid rheostat is a biochemical concept that involves regulation of the interconversion of sphingolipid metabolites, including ceramide, sphingosine, and sphingosine-1-phosphate (S1P), and their related signaling pathways, which play an important role in cell fate determination [7]. Accumulating evidences indicate that the dysregulated sphingolipid metabolism-related genes have clinicopathological relevance in various human cancers [8,9]. For example, a high expression of sphingosine kinase 1 (SPHK1) was significantly associated with a worse outcome in survival analysis of patients with breast cancer [10], and a high expression of ceramide synthase (CERS) 5 was significantly associated with a poor prognosis in colorectal cancer (CRC) patients [11]. However, research is scarce on the expression and prognostic potential of various sphingolipid metabolism-related genes in ccRCC. Thus, in the present study, we investigated the mRNA expression levels of the dysregulated sphingolipid metabolism-related genes in ccRCC using gene expression RNAseq data from The Cancer Genome Atlas Kidney Clear Cell Carcinoma (TCGA KIRC) cohort and their prognostic relevance.

Materials and Methods

Gene expression databases and cluster analysis

The gene expression RNAseq dataset (level 3, dataset ID: TCGA.KIRC.sampleMap/HiSeqV2) and clinical characteristic datasets (TCGA.KIRC.sampleMap/KIRC_clinicalMatrix and survival/KIRC_survival.txt) for the TCGA KIRC cohort were downloaded from the UCSC Xena database (https://xena.ucsc.edu). The RNAseq dataset provides gene-level transcript estimates with log2(x+1)-transformed counts normalized using the RNA-Seq by Expectation-Maximization (RSEM) normalization package. The RNAseq dataset of TCGA KIRC comprises 606 samples, including 533 primary tumor tissues, one additional new primary tumor tissue, and 72 normal solid tissues (NST). The mRNA expression levels of the sphingolipid metabolism-related genes [12] were sorted from the TCGA KIRC data. We calculated the |log2FoldChange (FC)| to screen for genes with more than 2-fold differences in expression levels between groups. This study met the publication guidelines provided by TCGA (http://www.cancer.gov/about-nci/organization/ccg/research/structrual-genomics/tcga/ using-tcga/citing-tcga). Cluster analysis was performed using the Cluster 3.0 software [13] to classify the samples into statistically similar groups, and the resulting heatmaps were visualized in TreeView [14] (version 1.6). Expression levels in the heatmaps have been quantile normalized [15], median-centered, and scaled for visualization.

Kaplan-Meier (KM) survival analysis with log rank test

Prior to a survival analysis, we calculated the median gene expression values of the selected sphingolipid metabolism-related genes from TCGA KIRC and determined the values as the cutoff values for survival analysis. Based on the median value of each gene, ccRCC samples of TCGA KIRC was divided into 2 groups. The survival of the two groups was analyzed using the KM method, and statistical significance was assessed using the log-rank test.

Statistical analysis

Statistical analysis was performed with the SPSS software (version 25.0; IBM SPSS, Armonk, NY, USA). The Kolmogorov-Smirnov test was used for testing normality of the data. Differences between groups were then statistically analyzed using the Student’s t-test. The association between inter-individual mRNA levels of the sphingolipid metabolism-related genes was assessed using the Pearson’s correlation coefficients for continuous variables. In general, p < 0.05 was considered to denote significance in all statistical analyses performed in the study.

Results

Altered expression levels of the sphingolipid metabolism-related genes in TCGA KIRC cohort.

The heatmap revealed that various sphingolipid metabolism-related genes were dysregulated in ccRCC compared with NST of patients with ccRCC (Fig. 1). The Student’s t-test confirmed that various sphingolipid metabolism-related genes were significantly dysregulated in ccRCC tissues compared with NST (p < 0.05; Fig. 2A). Specifically, beta-1,4-N-acetyl-galactosaminyltransferase 1 (B4GALNT1), the BCL2 interacting protein 3 (BNIP3), delta 4-desaturase, sphingolipid 1 (DEGS1), galactose-3-O-sulforansferase 1 (GAL3ST1), the sphingosine-1-phosphate receptor 4 (S1PR4), the solute carrier family 26 member 10 (SLC26A10), sphingomyelin phosphodiesterase acid like 3A (SMPDL3A), and SPHK1 were significantly upregulated in ccRCC, whereas beta-1,4-galactosyltransferase 6 (B4GALT6), 15-hydroxyprostaglandin dehydrogenase (HPGD), the lysophosphatidic acid receptor 1 (LPAR1), the surfactant protein B (SFTPB), ST6 N-acetylgalactosaminide alpha-2,6-sialyltransferase 5 (ST6GALNAC5), and UDP glycosyltransferase 8 (UGT8) were significantly downregulated in ccRCC compared to their respective expression in NST (the Student’s t-test, p < 0.05, |log2FC| ≥ 1.0; Fig. 2B). These 14 dysregulated sphingolipid metabolism-related genes may therefore play an important role in the ccRCC pathophysiology.

Fig. 1.

Heatmap presenting the relative mRNA expression levels of the sphingolipid metabolism-related genes in the TCGA KIRC cohort. The data are presented in matrix format in which each row represents an individual gene and each column represents a single tissue. Each cell in the matrix represents the expression level of a gene feature in an individual tissue. The red and green colors in cells reflect relative high and low expression levels, respectively, as indicated in the scale bar. The samples were ordered from NSTs to ccRCC samples according to the standardized expression level of each gene as indicated. TCGA KIRC, The Cancer Genome Atlas kidney clear cell carcinoma; ccRCC, clear cell renal cell carcinoma; NST, normal solid tissue.

Fig. 2.

(A) Heatmap showing significantly dysregulated mRNA expression of the sphingolipid metabolism‑related genes in ccRCC tissues compared with the NST of TCGA KIRC cohort. The data are presented in matrix format in which each row represents an individual gene and each column represents a single tissue. Each cell in the matrix represents the expression level of a gene feature in an individual tissue. The red and green colors in cells reflect relative high and low expression levels, respectively, as indicated in the scale bar. The samples were ordered from NSTs to ccRCC samples according to the standardized expression level of each gene as indicated. *p<0.05 (NST versus ccRCC). (B) Relative mRNA levels of various sphingolipid metabolism-related genes in TCGA KIRC samples. *Student t-test, p<0.05, |log2FoldChange| ≥ 1.0. TCGA KIRC, The Cancer Genome Atlas kidney clear cell carcinoma; ccRCC, clear cell renal cell carcinoma; NST, normal solid tissue.

Inter-individual correlations between sphingolipid metabolism-related genes showing altered mRNA expression levels in ccRCC.

Based on the Pearson’s correlation coefficient, we found 69 significant correlations between the mRNA expression levels of the 14 significantly dysregulated sphingolipid metabolism-related genes in ccRCC sorted from TCGA KIRC cohort (Table 1).

Pearson’s correlation analysis between inter-individual components of sphingolipid metabolism-related genes

Survival analysis of the significantly dysregulated sphingolipid metabolism-related genes in ccRCC.

Survival data were available for a total of 605 ccRCC patients for the KM survival analysis. The KM plot and log-rank test indicated that higher mRNA expression levels of B4GALNT1, SLC26A10, and SPHK1 were associated with an unfavorable overall survival in ccRCC patients (Fig. 3A-3C; p = 0.013, p < 0.001, and p < 0.001, respectively). However, higher mRNA expression levels of BNIP3, HPGD, and SMPDL3A were associated with a favorable overall survival in ccRCC patients (Fig. 3D-3F, p = 0.005, p < 0.001, and p = 0.009, respectively).

Fig. 3.

Survival analyses of the sphingolipid metabolism-related genes in ccRCC samples of TCGA KIRC cohort. Kaplan-Meier plot of overall survival in subjects with high versus low (A) B4GALNT1, (B) SLC26A10, (C) SPHK1, (D) BNIP3, (E) HPGD, and (F) SMPDL3A mRNA expression. TCGA KIRC, The Cancer Genome Atlas kidney clear cell carcinoma; ccRCC, clear cell renal cell carcinoma.

Discussion

Since the sphingolipid rheostat was proposed as an important concept in determining the cell fate, many studies have focused on the contribution of sphingolipid metabolism to the mechanism of cancer development, as well as for the identification of new targets for a cancer therapeutic strategy and prognostic factors [16]. However, the expression and potential prognostic markers of the sphingolipid metabolism-related genes remain to be investigated for many cancer types, including ccRCC. In this study, using the RNAseq data of TCGA KIRC cohort, we evaluated the dysregulation of sphingolipid metabolism-related genes in ccRCC and identified potential prognostic markers.

We identified eight sphingolipid metabolism-related genes (B4GALNT1, BNIP3, DEGS1, GAL3ST1, S1PR4, SLC26A10, SMPDL3A, and SPHK1) that were significantly upregulated in ccRCC in TCGA-KIRC cohort, and the KM survival analysis showed that higher expression levels of B4GALNT1, SLC26A10, and SPHK1 were significantly correlated with a worse prognosis. These results were found to be consistent with recent studies showing that worse overall survival was significantly associated with upregulated SPHK1 expression in non-small cell lung cancer [17] and that B4GALNT1 expression was upregulated in ccRCC [18]. Although SLC26A10 expression was found to be significantly upregulated in colorectal carcinoma compared to adenoma, the prognostic significance of SLC26A10 was not evident [19]. SPHK1 is a major enzyme of the S1P metabolism pathway, and plays an important role in tumorigenesis [20], cancer angiogenesis [21], intercellular communication of the tumor environment [22], and formation of the prometastatic environment [23]. B4GALNT1 is a key enzyme in generation of the gangliosides GM2/GD2, and has also been shown to be involved in tumorigenesis [24]. Thus, our results suggest that SPHK1 and B4GALNT1 may play crucial roles as oncogenes in ccRCC.

Similarly, we hypothesized that BNIP3 and SMPDL3A also functioned as oncogenes in ccRCC, and that their higher expression levels may contribute to poor prognoses. Noteworthy, BNIP3 and SMPDL3A were significantly upregulated in ccRCC, but higher expression levels of these genes were significantly correlated with better prognoses in TCGA-KIRC cohort. BNIP3 is a well-known tumor suppressor gene that regulates cell death and mitophagy [25], and its higher expression level has previously been correlated with a good prognosis [26], providing further support that BNIP3 plays an important role as a tumor suppressor gene in the ccRCC pathophysiology. However, further investigation is needed to clarify the reason for the increase in BNIP3 expression levels in ccRCC tissues compared with the levels in NST. SMPDL3A functions as a liver X receptor target gene; however, little is known about its potential roles in cancer. Conversely, the other six dysregulated sphingolipid metabolism-related genes (B4GALT6, HPGD, LPAR1, SFTPB, ST6GALNAC5, and UGT8) were significantly downregulated in ccRCC compared with those in NST. Notably, higher expression levels of HPGD were significantly correlated with a better prognosis in TCGA-KIRC cohort. HPGD is responsible for regulation of prostaglandin metabolism [27] and also acts as a tumor suppressor gene [28]. A decreased level of HPGD was correlated with a poor prognosis in patients with triple-negative breast cancer [29], which was found to be in line with our results, suggesting an important role in the pathophysiology of ccRCC. Moreover, in the present study we provide the first evidence for the significantly-altered regulation of SLC26A10, B4GALT6, and SMPDL3A and their prognostic significance in ccRCC. However, previous studies have demonstrated conflicting results in finding a poor prognosis associated with higher expression of BNIP3 in patients with non-small cell lung cancer [30] and uveal melanoma [31], and higher expression of HPGD in patients with breast cancer [32]. Therefore, more detailed analysis will be needed to determine the regulatory mechanisms of sphingolipid metabolism-related genes and their association with the pathophysiology of ccRCC.

Significant correlations between the sphingolipid metabolism-related genes have been reported in various CRC cohorts, including CERS2 and CERS4, CERS4 and CERS5, and CERS5 and CERS6 [33]. Inter-individual correlations between the significantly dysregulated sphingolipid metabolism-related genes of ccRCC revealed 69 significant correlations; however, no genes were correlated with these six CERS genes as none of these genes showed significantly-altered expression levels in ccRCC compared with those in NST. Therefore, cancer signaling pathways altered in ccRCC may be related to the correlation between the sphingolipid metabolism-related genes other than the CERS genes.

In conclusion, we identified the significantly-altered sphingolipid metabolism-related genes and their prognostic significance in patients with ccRCC using TCGA-KIRC cohort. Notably, our study suggests that B4GALNT1, SLC26A10, SPHK1, BNIP3, HPGD, and SMPDL3A may be potential prognostic biomarkers and promising therapeutic candidates for ccRCC.

Abbreviations

CERS, Ceramide synthase; ccRCC, clear cell renal cell carcinoma; TCGA KIRC, The Cancer Genome Atlas Kidney Clear Cell Carcinoma; NST, normal solid tissue; CRC, Colorectal cancer; B4GALNT1, beta-1,4-N-acetyl-galactosaminyltransferase 1; B4GALT6, beta-1,4-galactosyltransferase 6; BNIP3, BCL2 interacting protein 3; DEGS1, delta 4-desaturase, sphingolipid 1; GAL3ST1, galactose-3-O-sulforansferase 1; HPGD, 15-hydroxyprostaglandin dehydrogenase; LPAR1, lysophosphatidic acid receptor 1; S1PR4, sphingosine-1-phosphate receptor 4; SLC26A10, solute carrier family 26 member 10; SFTPB, surfactant protein B; SMPDL3A, sphingomyelin phosphodiesterase acid like 3A; SPHK1, sphingosine kinase 1; ST6GALNAC5, ST6 N-acetylgalactosaminide alpha-2,6-sialyltransferase 5; UGT8, UDP glycosyltransferase 8.

Acknowledgements

We would like to thank all the members of our research group for their enthusiastic participation in this study.

Notes

Funding

The present study was supported by the National Research Foundation of Korea (grant funded by the Korea Government Ministry of Science, ICT and Future Planning; grant nos. 2014R1A5A2010008, NRF-2016R1D1A1B04930619, and NRF-2019R1I1A3A01063114).

Availability of data and materials

The datasets analyzed during the present study are available from The Cancer Genome Atlas (https://www.cancer.gov/tcga) and the UCSC Xena (https://xena.ucsc.edu). The datasets generated in the present study are available from the corresponding author upon reasonable request.

Authors contributions

WJP and SK contributed to the conception and design of the study, analysis of the data, interpretation of results, and the writing of the manuscript. JYP and SK contributed to the acquisition of data and the writing of the manuscript. TKK, JWP, and SK reviewed the manuscript. SK edited the manuscript. All authors read and approved the manuscript and agree to be accountable for all aspects of the research in ensuring that the accuracy and integrity of any part of the work are appropriately investigated and resolved.

Ethics approval

Not applicable.

Patient consent

Not applicable.

Conflict of interest

All authors declare no conflicts-of-interest related to this article.

References

1. Bhatt JR, Finelli A. Landmarks in the diagnosis and treatment of renal cell carcinoma. Nat Rev Urol 2014;11:517–25.
2. Kovacs G, Akhtar M, Beckwith BJ, Bugert P, Cooper CS, Delahunt B, et al. The Heidelberg classification of renal cell tumours. J Pathol 1997;183:131–3.
3. MacLennan S, Imamura M, Lapitan MC, Omar MI, Lam TB, Hilvano-Cabungcal AM, et al. Systematic review of oncological outcomes following surgical management of localised renal cancer. Eur Urol 2012;61:972–93.
4. Li Z, Liu J, Zhang X, Fang L, Zhang C, Zhang Z, et al. Prognostic significance of cyclin D1 expression in renal cell carcinoma: a systematic review and meta-analysis. Pathol Oncol Res 2019;DOI: 10.1007/s12253-019-00776-0.
5. Kojima T, Shimazui T, Horie R, Hinotsu S, Oikawa T, Kawai K, et al. FOXO1 and TCF7L2 genes involved in metastasis and poor prognosis in clear cell renal cell carcinoma. Genes Chromosomes Cancer 2010;49:379–89.
6. Li X, Tan X, Yu Y, Chen H, Chang W, Hou J, et al. D9S168 microsatellite alteration predicts a poor prognosis in patients with clear cell renal cell carcinoma and correlates with the down-regulation of protein tyrosine phosphatase receptor delta. Cancer 2011;117:4201–11.
7. Cuvillier O, Pirianov G, Kleuser B, Vanek PG, Coso OA, Gutkind S, et al. Suppression of ceramide-mediated programmed cell death by sphingosine-1-phosphate. Nature 1996;381:800–3.
8. Ryland LK, Fox TE, Liu X, Loughran TP, Kester M. Dysregulation of sphingolipid metabolism in cancer. Cancer Biol Ther 2011;11:138–49.
9. Sedic M, Grbcic P, Pavelic SK. Bioactive sphingolipids as biomarkers predictive of disease severity and treatment response in cancer: current status and translational challenges. Anticancer Res 2019;39:41–56.
10. Geffken K, Spiegel S. Sphingosine kinase 1 in breast cancer. Adv Biol Regul 2018;67:59–65.
11. Fitzgerald S, Sheehan KM, Espina V, O'Grady A, Cummins R, Kenny D, et al. High CerS5 expression levels associate with reduced patient survival and transition from apoptotic to autophagy signalling pathways in colorectal cancer. J Pathol Clin Res 2015;1:54–65.
12. Ruckhaberle E, Rody A, Engels K, Gaetje R, von Minckwitz G, Schiffmann S, et al. Microarray analysis of altered sphingolipid metabolism reveals prognostic significance of sphingosine kinase 1 in breast cancer. Breast Cancer Res Treat 2008;112:41–52.
13. de Hoon MJ, Imoto S, Nolan J, Miyano S. Open source clustering software. Bioinformatics 2004;20:1453–4.
14. Saldanha AJ. Java Treeview--extensible visualization of microarray data. Bioinformatics 2004;20:3246–8.
15. Evans C, Hardin J, Stoebel DM. Selecting between-sample RNA-Seq normalization methods from the perspective of their assumptions. Brief Bioinform 2018;19:776–92.
16. Newton J, Lima S, Maceyka M, Spiegel S. Revisiting the sphingolipid rheostat: evolving concepts in cancer therapy. Exp Cell Res 2015;333:195–200.
17. Wang Y, Shen Y, Sun X, Hong TL, Huang LS, Zhong M. Prognostic roles of the expression of sphingosine-1-phosphate metabolism enzymes in non-small cell lung cancer. Transl Lung Cancer Res 2019;8:674–81.
18. Yang H, Li W, Lv Y, Fan Q, Mao X, Long T, et al. Exploring the mechanism of clear cell renal cell carcinoma metastasis and key genes based on multi-tool joint analysis. Gene 2019;720DOI: 10.1016/j.gene.2019.144103.
19. Carvalho B, Sillars-Hardebol AH, Postma C, Mongera S, Terhaar Sive Droste J, Obulkasim A, et al. Colorectal adenoma to carcinoma progression is accompanied by changes in gene expression associated with ageing, chromosomal instability, and fatty acid metabolism. Cell Oncol (Dordr) 2012;35:53–63.
20. Nava VE, Hobson JP, Murthy S, Milstien S, Spiegel S. Sphingosine kinase type 1 promotes estrogen-dependent tumorigenesis of breast cancer MCF-7 cells. Exp Cell Res 2002;281:115–27.
21. Dai L, Liu Y, Xie L, Wu X, Qiu L, Di W. Sphingosine kinase 1/sphingosine-1-phosphate (S1P)/S1P receptor axis is involved in ovarian cancer angiogenesis. Oncotarget 2017;8:74947–61.
22. Schneider G. S1P signaling in the tumor microenvironment. Adv Exp Med Biol 2020;1223:129–53.
23. Schneider G, Bryndza E, Abdel-Latif A, Ratajczak J, Maj M, Tarnowski M, et al. Bioactive lipids S1P and C1P are prometastatic factors in human rhabdomyosarcoma, and their tissue levels increase in response to radio/chemotherapy. Mol Cancer Res 2013;11:793–807.
24. Yoshida H, Koodie L, Jacobsen K, Hanzawa K, Miyamoto Y, Yamamoto M. B4GALNT1 induces angiogenesis, anchorage independence growth and motility, and promotes tumorigenesis in melanoma by induction of ganglioside GM2/GD2. Sci Rep 2020;10DOI: 10.1038/s41598-019-57130-2.
25. Chourasia AH, Macleod KF. Tumor suppressor functions of BNIP3 and mitophagy. Autophagy 2015;11:1937–8.
26. Chourasia AH, Tracy K, Frankenberger C, Boland ML, Sharifi MN, Drake LE, et al. Mitophagy defects arising from BNip3 loss promote mammary tumor progression to metastasis. EMBO Rep 2015;16:1145–63.
27. Zhang Y, Desai A, Yang SY, Bae KB, Antczak MI, Fink SP, et al. Tissue regeneration. Inhibition of the prostaglandin-degrading enzyme 15-PGDH potentiates tissue regeneration. Science 2015;348DOI: 10.1126/science.aaa2340.
28. Wolf I, O'Kelly J, Rubinek T, Tong M, Nguyen A, Lin BT, et al. 15-Hydroxyprostaglandin dehydrogenase is a tumor suppressor of human breast cancer. Cancer Res 2006;66:7818–23.
29. Kochel TJ, Goloubeva OG, Fulton AM. Upregulation of cyclooxygenase-2/prostaglandin E2 (COX-2/PGE2) pathway member multiple drug resistance-associated protein 4 (MRP4) and downregulation of prostaglandin transporter (PGT) and 15-prostaglandin dehydrogenase (15-PGDH) in triple-negative breast cancer. Breast Cancer (Auckl) 2016;10:61–70.
30. Giatromanolaki A, Koukourakis MI, Sowter HM, Sivridis E, Gibson S, Gatter KC, et al. BNIP3 expression is linked with hypoxia-regulated protein expression and with poor prognosis in non-small cell lung cancer. Clin Cancer Res 2004;10:5566–71.
31. Jiang Z, Yu F, Li M. Upregulation of BCL2 19 kD protein-interacting protein 3 (BNIP3) is predictive of unfavorable prognosis in uveal melanoma. Med Sci Monit 2018;24:4711–7.
32. Lehtinen L, Vainio P, Wikman H, Reemts J, Hilvo M, Issa R, et al. 15-Hydroxyprostaglandin dehydrogenase associates with poor prognosis in breast cancer, induces epithelial-mesenchymal transition, and promotes cell migration in cultured breast cancer cells. J Pathol 2012;226:674–86.
33. Jang SW, Park WJ, Min H, Kwon TK, Baek SK, Hwang I, et al. Altered mRNA expression levels of the major components of sphingolipid metabolism, ceramide synthases and their clinical implication in colorectal cancer. Oncol Rep 2018;40:3489–500.

Article information Continued

Fig. 1.

Heatmap presenting the relative mRNA expression levels of the sphingolipid metabolism-related genes in the TCGA KIRC cohort. The data are presented in matrix format in which each row represents an individual gene and each column represents a single tissue. Each cell in the matrix represents the expression level of a gene feature in an individual tissue. The red and green colors in cells reflect relative high and low expression levels, respectively, as indicated in the scale bar. The samples were ordered from NSTs to ccRCC samples according to the standardized expression level of each gene as indicated. TCGA KIRC, The Cancer Genome Atlas kidney clear cell carcinoma; ccRCC, clear cell renal cell carcinoma; NST, normal solid tissue.

Fig. 2.

(A) Heatmap showing significantly dysregulated mRNA expression of the sphingolipid metabolism‑related genes in ccRCC tissues compared with the NST of TCGA KIRC cohort. The data are presented in matrix format in which each row represents an individual gene and each column represents a single tissue. Each cell in the matrix represents the expression level of a gene feature in an individual tissue. The red and green colors in cells reflect relative high and low expression levels, respectively, as indicated in the scale bar. The samples were ordered from NSTs to ccRCC samples according to the standardized expression level of each gene as indicated. *p<0.05 (NST versus ccRCC). (B) Relative mRNA levels of various sphingolipid metabolism-related genes in TCGA KIRC samples. *Student t-test, p<0.05, |log2FoldChange| ≥ 1.0. TCGA KIRC, The Cancer Genome Atlas kidney clear cell carcinoma; ccRCC, clear cell renal cell carcinoma; NST, normal solid tissue.

Fig. 3.

Survival analyses of the sphingolipid metabolism-related genes in ccRCC samples of TCGA KIRC cohort. Kaplan-Meier plot of overall survival in subjects with high versus low (A) B4GALNT1, (B) SLC26A10, (C) SPHK1, (D) BNIP3, (E) HPGD, and (F) SMPDL3A mRNA expression. TCGA KIRC, The Cancer Genome Atlas kidney clear cell carcinoma; ccRCC, clear cell renal cell carcinoma.

Table 1.

Pearson’s correlation analysis between inter-individual components of sphingolipid metabolism-related genes

Samples Correlations between components Pearson’s correlation coefficient value p-Value*
TCGA KIRC tissues from TCGA KIRC cohort (n = 533) B4GALNT1 and B4GALT6 -0.271 < 0.001
B4GALNT1 and BNIP3 -0.129 0.003
B4GALNT1 and GAL3ST1 -0.267 < 0.001
B4GALNT1 and SLC26A10 0.768 < 0.001
B4GALNT1 and SMPDL3A -0.244 < 0.001
B4GALNT1 and SPHK1 0.181 < 0.001
B4GALNT1 and ST6GALNAC5 0.087 0.045
B4GALNT1 and UGT8 -0.110 0.011
B4GALT6 and DEGS1 -0.141 0.001
B4GALT6 and GAL3ST1 -0.127 0.003
B4GALT6 and HPGD 0.158 < 0.001
B4GALT6 and S1PR4 -0.232 < 0.001
B4GALT6 and SFTPB 0.352 < 0.001
B4GALT6 and SLC26A10 -0.346 < 0.001
B4GALT6 and SMPDL3A 0.120 0.006
B4GALT6 and SPHK1 -0.255 < 0.001
B4GALT6 and ST6GALNAC5 -0.183 < 0.001
B4GALT6 and UGT8 0.347 < 0.001
BNIP3 and DEGS1 0.123 0.004
BNIP3 and GAL3ST1 0.410 < 0.001
BNIP3 and HPGD -0.106 0.014
BNIP3 and LPAR1 -0.201 < 0.001
BNIP3 and SFTPB -0.353 < 0.001
BNIP3 and SLC26A10 -0.104 0.016
BNIP3 and SMPDL3A 0.419 < 0.001
BNIP3 and SPHK1 -0.320 < 0.001
*

Pearson’s correlation coefficient analysis.