Comparative transcriptomic analysis reveals differences in gene expression and regulatory pathways between nonacral and acral melanoma in Asian individuals

Abstract

Melanoma predominantly occurs in White individuals, which is associated with factors such as exposure to UV radiation and skin pigmentation. Despite its low incidence, melanoma is the primary cause of skin cancer-related death in Asia, typically in areas with low sun exposure. In our previous whole-exome sequencing study, we identified mutational signature 12 as the most prevalent variant in Asian patients, differing from the common UV-associated mutational signature 7 observed in White individuals. We also observed major differences between acral melanoma (AM) and nonacral melanoma (NAM) in terms of signatures 7, 21, and 22. Notably, few studies have investigated the genomic differences between AM and NAM in Asian individuals. Therefore, in this study, we conducted transcriptomic sequencing to examine the disparities in RNA expression between AM and NAM. Ribosomal RNA depletion was performed to enhance the detection of functionally relevant coding and noncoding transcripts. Ingenuity pathway analysis revealed significant differences in gene expression and regulatory pathways between AM and NAM. The results also indicate that the genes involved in cell cycle signaling or immune modulation and programmed cell death protein 1/programmed cell death 1 ligand 1 signaling were differentially expressed in NAM and AM. In addition, high CDK4 expression and cell cycle variability were observed in AM, with high immunogenicity in NAM. Overall, these findings provide further insights into the pathogenesis of melanoma and serve as a reference for future research on this major malignant disease.

Keywords: Asians; RNA sequencing; melanoma; transcriptome.

FIGURE 1

Bubble plots revealing the differences in gene count and regulatory pathways between nonacral melanoma (NAM) and acral melanoma (AM). The size of each bubble represents the number of genes overlapping with the pathways. Blue bubbles represent negative values, and orange bubbles represent positive values. (a) In the case of NAM, most of the upregulated genes were distributed in the cellular stress and injury pathway, cancer pathway, and cellular immune response pathway. (b) In the case of AM, most of the downregulated genes were distributed in the cellular stress and injury pathway, cancer pathway, and cellular immune response pathway.

FIGURE 2

Differences in signaling pathways between nonacral melanoma (NAM) and acral melanoma (AM). Each horizontal bar denotes a distinct pathway with a unique z score. Red bars indicate activation, and blue bars indicate suppression. The following pathways were identified as the most significantly different regulatory pathways between NAM and AM: S100 protein family signaling pathway, wound healing signaling pathway, phagosome formation pathway, stearate biosynthesis I pathway, focal adhesion kinase (FAK) signaling pathway, epithelial mesenchymal transition (EMT) regulation signaling pathway, cAMP‐response element‐binding protein signaling pathway (CREB), tumor microenvironment signaling pathway, and natural killer cell signaling pathway. HIF1α, hypoxia‐inducible factor 1 α; IL‐8, interleukin 8; MAPK, mitogen‐activated protein kinase; NF‐κB, nuclear factor–κB.

FIGURE 3

Melanoma and cell cycle mechanistic networks with multiple affected cell cycle signaling networks, predicting the activated state of the network on the basis of transcriptome data, with subsequent predicted effects on downstream effector molecules. The figure shows the differences in cell cycle family networks between acral melanoma (AM) and nonacral melanoma (NAM). In NAM, the expression of genes CDK2, CDK5, CDK5RAP2, CDK11A, CDK11B, CDK20, CDKN2A, and E2F3 was significantly upregulated. In AM, the expression of genes CDK7, CDK8P2, CDK16, CDKAL1, CDH1, E2F6, E2F8, and MITF was significantly upregulated.

FIGURE 4

S100 mechanistic networks with multiple affected S100 signaling networks, predicting the activated state of the network on the basis of transcriptome data, with subsequent predicted effects on downstream effector molecules. In NAM, the expression of genes S100A1, S100A4, S100A7, S100A13, and S100PBP was significantly upregulated. In AM, the expression of genes S100A2, S100A7A, S100A8, and S100P was significantly upregulated. However, the expression of gene S100B did not differ between nonacral melanoma (NAM) and acral melanoma (AM).

FIGURE 5

Immune modulator mechanistic networks with multiple affected melanoma signaling networks, predicting the activated state of the network on the basis of transcriptome data, with subsequent predicted effects on downstream effector molecules. The figure shows the differences in melanoma networks between acral melanoma (AM) and nonacral melanoma (NAM). In NAM, the expression of genes IL1B, IL2, IL10, IL12A, IL13, IL26, IL36, IFNG, and TGFB1 was significantly upregulated. In AM, the expression of genes IL1A, IL6, and IL32 was significantly upregulated. According to the results of human leukocyte antigen (HLA) gene analysis, the expression of genes HLA‐A, HLA‐B, HLA‐DMA, HLA‐DMB, and HLA‐K was significantly upregulated in NAM.

FIGURE 6

Programmed cell death protein 1/programmed cell death 1 ligand 1 (PD‐1/PD‐L1) mechanistic networks with multiple affected immune modulator signaling networks, predicting the activated state of the network on the basis of transcriptome data, with subsequent predicted effects on downstream effector molecules. The figure shows the differences in immune modulator networks between acral melanoma (AM) and nonacral melanoma (NAM). According to the results of PD‐1/PD‐L1 signaling analysis, the expression of genes IL2RA, IL2RB, IL2RG, HLA‐A, HLA‐B, CD274, and CDK2 was significantly upregulated in NAM.

FIGURE 7

Immunohistochemistry analyses of CD8, programmed cell death protein 1 (PD‐1), programmed cell death 1 ligand 1 (PD‐L1) on both nonacral melanoma (NAM) and acral melanoma (AM) tumor samples. The results from representative cases are shown to reveal a higher expression of CD8, PD‐1, and PD‐L1 (brown color) in the NAM samples compared with the AM samples. Bars = 100 μm.

FIGURE 8

Quantitative polymerase chain reaction analyses targeting CD8, programmed cell death protein 1 (PD‐1), and programmed cell death 1 ligand 1 (PD‐L1) expression of both nonacral melanoma (NAM) and acral melanoma (AM) tumor samples. The findings revealed higher expression of these CD8, PD‐1, and PD‐L1 immune markers in the NAM samples compared with the AM samples. *p‐value <0.05 was considered to be statistically significant.