Introduction
Chromophobe renal cell carcinoma (KICH) is a subtype of renal cell carcinoma (RCC) [
1], which is first described in 1985 and accounts for 5% of RCC [
2,
3]. The lack of rich vascular networks, which usually could be observed in clear-cell carcinomas, is a characteristic of KICH [
4]. Numerous studies revealed that the prognosis of KICH is better than other subtypes of RCC [
5]. In addition, the prognosis of most KICH was not related to the pathological stage, while the other subtypes were associated with the pathological stage [
6]. However, KICH is the main kind of primary subtype of the sarcomatoid differentiated RCC, which is generally related to a bad prognosis [
7]. Therefore, it is of critical importance to identify prognosis-related biomarkers, which are helpful for the diagnosis and treatment strategy selection of KICH.
Centrosome-associated protein E (CENPE) is a plus end-directed kinetochore motor protein, which accumulates at the G2 phase of the cell cycle and plays a vital role in mitosis [
8,
9]. Previous studies confirmed that CENPE promoted the proliferation of multiple cancers, such as ovarian cancer, lung adenocarcinoma and lung adenocarcinoma [
8,
10,
11]. Besides, Zhu et al. claimed that CENPE could be a biomarker of esophageal adenocarcinoma [
9]. Moreover, the overexpressed CENPE was closely related to the poor prognosis in breast cancer (BC), which indicated that CENPE could be a potential prognosis biomarker of BC [
12]. Although Wang et al. determined that CENPE could promote development and metastasis through the Wnt/β-catenin signal pathway in clear cell renal cell carcinoma [
13], the function of CENPE in KICH remains unclear.
Additionally, lactate dehydrogenase A (LDHA) is one of the key enzymes in glycolysis progress which have crucial effects on cancer cell growth [
14]. Meanwhile, the overproduction of lactate, which depends on the function of LDHA, in glycolysis causes triggering immune escape and promotes the development of tumors [
15]. It has been identified that LDHA was overexpressed in various cancers and involved in tumorigenesis and tumor growth [
16]. Furthermore, Huo et al. unraveled that the LDHA-mediated glycolysis was inhibited by LINC00671 in papillary thyroid cancer cells, which contributed to the suppression of tumor cell growth and metastasis [
14]. Similarly, the effect of LDHA on KICH is also not reported, while it has been found that LDHA was involved in clear cell renal cell carcinoma [
17].
Thus, in this study, we tried to explore the expression, function and regulatory axis of CENPE and LDHA in the development of KICH.
Methods
The collection of RNA expression and clinical information
The expression of mRNA and miRNA and corresponding clinical information (age, gender, race, M stage, N stage, T stage) were obtained from the University of California Santa Cruz (UCSC) Xena database (
https://xenabrowser.net/). After removing missing data, 64 KICH samples and 24 normal samples were obtained and used for subsequent analysis. Besides, mRNA and miRNA were considered not expressed if the expressions were not detected in at least 10% samples.
The differently expressed genes (DEGs) analysis
After normalization of the raw count data with transcripts per million (TPM) method and undergoing a log2 transformation, 58,939 genes were annotated. Then, we used the R package t.test function to evaluate the significance of each gene in the tumor group and the normal group, and used the p.adjust function to calculate the false discovery rate (FDR) of each gene, and 14,565 DEGs with an absolute log2 fold change (FC) > 1 and p-value < 0.05 were obtained.
Gene Ontology (GO) term, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis
Then we identified the function of DEGs in KICH progression using GO and KEGG enrichment analyses. For gene set functional enrichment analysis, we used Kegg REST API (
https://www.Kegg.jp/Kegg/rest/KeggAPI.html) and org.Hs.eg.db (version 3.1.0) to obtain the latest gene annotation of KEGG and GO Pathway, which was used as the background, and then mapped the genes into the background set, and used R software package clusterProfiler (version 3.14.3) for enrichment analysis. The
p < 0.05 and FDR < 0.25 were considered statistically significant.
The identification of hub gene
Least absolute shrinkage and selection operator (LASSO) regression was a kind of method used for selecting candidate genes which were closely related to prognosis. After we integrate the data of overall survival (OS) time, survival status and gene expression, the lasso-cox regression was performed using R software package glmnet with fivefold cross-validation (CV). After optimal lambda value was determined, hub genes with nonzero coefficients were screened out. Meanwhile, the risk scores and risk model were determined according to the gene expression value and LASSO coefficients. After the lasso-cox method, we further evaluated the prognosis value of hub genes and clinical characteristics in KICH through multivariate cox regression in SPSS 25.
Survival analysis
The Kaplan–Meier (K–M) survival curve was used for assessing the effect of mRNA, miRNA and risk score on prognosis in KICH. Before survival analysis, the KICH samples were divided into two groups according to the optimal truncation value determined using the maxstat of the R package. Then, we evaluated the difference between the two groups in the OS time through the survfit of the R package.
Gene set enrichment analysis (GSEA)
We explored the function of hub genes in KICH using GSEA. At first, we obtained GSEA software (version 3.0) from the website of GSEA. Then, GESA was performed by the GSEA software using a gene set database (
http://www.gsea-msigdb.org/gsea/downloads.jsp) downloaded from the molecular signatures database. The
p < 0.05 and FDR < 0.25 were considered statistically significant.
Immune infiltrate analysis
The tumor Immune Estimation Resource (TIMER) database (
https://cistrome.shinyapps.io/timer/) server is a comprehensive resource for the systematical analysis of immune infiltrates across diverse cancer types. We analyzed the association between hub genes and 6 types of immune genes, such as B cell, CD8 + T cell, CD4 + T cell, macrophage, neutrophil and dendritic cell (DC) using TIMER.
P-values of less than 0.05 were considered statistically significant.
The prediction of miRNA and construction of a miRNA-mRNA regulatory network
Statistical analysis
In this study, all experiments were repeated three times. The data were analyzed using SPSS 25 and presented as mean ± standard deviation. The comparison between the two groups was evaluated by independent sample t-test, while the comparison among multiple groups was conducted by one-way analysis of variance (ANOVA) followed by post-hoc comparisons. The survival difference between the 2 groups was determined by log-rank test, and the correlation analysis was identified by the Pearson test. p < 0.05 was considered a significant difference.
Discussion
In this study, according to the TCGA dataset, we identified 20,065 differentially expressed genes, including 6162 upregulated genes and 13,903 downregulated genes, in KICH samples compared with normal samples. Besides, the results of GO and KEGG enrichment analysis suggested that the DEGs were involved in the development of tumors and immune [
18‐
21]. Then, LASSO regression analysis was applied to screen out 2 hub genes, namely CENPE and LDHA. Besides, we established a risk score model based on these 2 genes, and the results of ROC curves indicated that the model could accurately predict the prognosis for patients with KICH, with the AUC of 0.93 and 0.97 for the 3-year and 5-year survival time, respectively. Furthermore, the results of the K–M plot showed that the high risk score was associated with a bad prognosis of patients diagnosed with KICH, which suggested that high expressions of CENPE and LDHA may be involved in the development of KICH.
CENPE had been regarded as a potential biomarker of diverse cancers, such as invasive ductal carcinoma and non-small cell lung cancer [
11,
22]. As a mitotic cell cycle-associated gene, CENPE has an essential and positive function in the process of mitotic cytoplasmic separation, which was related to the cell cycle [
9,
23], and the rapid proliferation is a common characteristic of tumor cells [
24,
25]. Shi et al. certificated that the silencing CENPE led to the inhibition of cell proliferation and promotion of apoptosis in acute myeloid leukemia cells [
26]. In addition, Wang et al. found that overexpressing CENPE promoted the cell viability, migration, and invasion of neuroblastoma [
27]. These results suggested that the high expression of CENPE could promote the development of tumors. In our study, we found that CENPE was upregulated in KICH samples compared to the normal samples. Besides, the high expression of CENPE could independently predict a bad prognosis in KICH samples according to the K–M plot and multivariate cox regression analysis. These results indicated that CENPE was a potential biomarker of KICH with cancer promoting activity.
RCC is essentially a metabolic disease characterized by a reprogramming of energetic metabolism [
28‐
31]. In particular the metabolic flux through glycolysis is partitioned [
32‐
34], and mitochondrial bioenergetics and OxPhox are impaired, as well as lipid metabolism [
32,
35‐
37]. As a kind of lactate dehydrogenase enzyme, LDHA played a vital role in glucose metabolism and many signal pathways related to cancer development [
38,
39], and it has been regarded as a biomarker for prognosis of many cancers [
40‐
42]. Additionally, the development of tumors could be modulated through regulating glucose metabolism [
43]. Although oxidative phosphorylation was the preferred energy production process, cancer cells including clear cell renal carcinoma (ccRCC) usually acquired energy from glycolysis [
44], and further studies proved that LDHA could also promote tumor progression through glycolysis [
45]. However, in direct contrast to ccRCC which is the most common subtype of RCC, KICH have an irregular metabolic program [
46]. Previous studies identified that the increased oxidative phosphorylation was observed in KICH [
47]. In this study, we found that LDHA expression downregulated in KICH samples compared to the normal samples, but it was high-expressed in most cancers. In addition, the K–M plot and multivariate cox regression analysis verified that high expression of LDHA also could independently predict bad prognosis in KICH samples, which suggested that LDHA was a potential biomarker of KICH. Moreover, the GSEA results showed that LDHA negatively regulated oxidative phosphorylation. It meant that the LDHA provided the energy to promote the KICH growth through glycolysis and aggravate the KICH prognosis through oxidative phosphorylation.
In addition, RCC is one of the most immune-infiltrated tumors [
48,
49]. Emerging evidence suggests that the activation of specific metabolic pathway have a role in regulating angiogenesis and inflammatory signatures [
50,
51]. Features of the tumor microenvironment heavily affect disease biology and may affect responses to systemic therapy [
52‐
55]. Previous studies demonstrated that LDHA was involved in the development of cancer through multiple signal pathways, including AKT/ mammalian target of rapamycin (mTOR), the c-Jun NH (2)-terminal kinase (JNK) and RIG-I like receptor signal pathway [
56‐
58]. Besides, we found LDHA also promoted cancer progression through being involved in the cell cycle and DNA replication. Furthermore, RIG-I like the receptor signaling pathway was related to the immune response [
59], which suggested that LDHA may be also involved in the immune response. Our results of immune infiltration analysis demonstrated that LDHA was significantly and positively related to the B cell, CD8 + T cell, macrophage and DC. Thus, combined with the expression and prognosis value of LDHA in KICH, we inferred that the downregulation of LDHA might inhibit the anti-tumor immune response and promote the initiation of KICH. Besides, the GSEA enrichment analysis revealed that the CENPE played a positive role in the cell cycle, p53 signaling pathway and DNA replication, which cloud promote tumor cell proliferation [
60‐
62]. It was worth noting that the function of highly expressed CENPE was also enriched in the T cell receptor signaling pathway. It had been reported that high expression of CENPE promoted immune cell infiltration, especially B cells [
63]. Similarly, we also found that CENPE was significantly and positively related to CD8 + T cell and macrophage. However, it seems to be against the function of CENPE as a carcinogenic factor, because the CD8 + T cell and macrophage could inhibit the development of cancer [
64,
65].
It had been reported that CNEPE and LDHA could be targeted by miRNA in some cancers [
66,
67]. What’s more, there was a wide regulatory spectrum between mRNA and miRNA, because the mRNA could bind to several miRNAs and the miRNA could bind to hundreds of mRNAs [
68]. According to the four prediction datasets and expression datasets, we found hsa-miR-577 was the common target of CENPE and LDHA. We speculated that hsa-miR-577 may target CENPE and LDHA and influence the progression of KICH. It had been reported that hsa-miR-577 was involved in the progression of renal cancer except for KICH [
69,
70]. Our study demonstrated that hsa-miR-577 was upregulated in KICH and might promote KICH progression through targeting CENPE and LDHA. The detailed regulation of hsa-miR-577 in KICH needs further investigation.
Though this study provided 2 important biomarkers for KICH, several limitations should be acknowledged. For example, the factors contributed to the LDHA downregulated in tumor samples remains unclear, and the mechanism of CENPE as a carcinogenic factor positively regulating the immune cells was not clarified. These problems might be the direction of the future research.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.