Actin-like Protein 6A Expression Correlates with Cancer Stem Cell-like Features and Poor Prognosis in Ovarian Cancer

Abstract

Ovarian cancer has the highest mortality rate among gynecological cancers, often diagnosed at the late stage and lacking an effective targeted therapy. Although the study of malignant features of cancer, considered to be cancer stem cells (CSCs), is emerging, the aim of this study was to predict and explore the possible mechanism and clinical value of genetic markers in the development of ovarian cancer from a combined database with CSCs features. The common differentially expressed genes (DEGs) were selected in GSE185833 and GSE176246 datasets from the Gene Expression Omnibus (GEO). The GSE185833 dataset was created to reveal gene expression profiles of peritoneal metastasis tissues using single-cell sequencing, and the GSE176246 dataset was determined from gene expression profiles of chemotherapy-refractory ovarian cancer cell lines compared with ovarian cancer cell lines by RNA-seq analysis. By analyzing the correlation between common DEGs and prognosis of ovarian cancer and its possible pathways and functions were predicted by The Cancer Genome Atlas (TCGA) database. The expression levels of 11 genetic markers were significantly elevated in highly invasive and chemoresistant ovarian cancer. The expression of Actin-like protein 6A (ACTL6A) was found to be correlated with survival prognosis, and the total survival time of the patients with high expression of ACTL6A was shorter than those with low expression. Gene set enrichment analysis (GSEA) showed that ACTL6A positively enriched the gene set of 'Cell cycle' and ACTL6A negatively enriched the gene set of focal adhesion. CP724714, a human epidermal growth factor receptor 2 (HER2) inhibitor, could serve as a therapeutic option when ACTL6A levels are high in ovarian cancer cells. The high expression of ACTL6A is a poor prognostic factor in ovarian cancer and may serve as an effective biomarker for predicting treatment-refractory, metastasis, and prognosis of patients with ovarian cancer. The use of HER2 inhibitors is a promising therapeutic strategy against chemoresistant ovarian cancer.

Keywords: ACTL6A; cell cycle; metastasis; ovarian cancer; prognosis; resistance.

Figure 1

The common genes associated chemoresistance and metastasis in ovarian cancer. (A) The Venn diagram shows the mapping results among ovarian 2162 common genes associated metastasis in the GSE185833 dataset. (B) The Venn diagram shows the mapping results among ovarian 623 common genes associated chemoresistance in GSE176246 dataset. (C) The Venn diagram shows the mapping results among ovarian 11 common genes associated chemoresistance and metastasis in GSE185833 and GSE176246 datasets. (D) The mRNA expression levels of 11 DEGs, including CREBBP, ACTL6A, ANXA5, EPB41L4A-AS1, FAM217A, IFI27L2, LNX2, MED28, TACR1, TLE4, and ZNF763 between the peritoneal metastasis tissues and their corresponding primary ovarian cancer. (E) The mRNA expression levels of 11 DEGs, including CREBBP, ACTL6A, ANXA5, EPB41L4A-AS1, FAM217A, IFI27L2, LNX2, MED28, TACR1, TLE4, and ZNF763 between the chemoresistant ovarian cancer cells and their corresponding parental ovarian cancer cells.

Figure 2

GACTL6A has poor prognosis for overall survival (OS) and progression-free survival (PFS) (A) Heat map of hazard ratios (HR) illustrating cancer-11 DEGs with altered prognosis. We selected the median as the suitable expression threshold for splitting the high-expression and low-expression cohorts. The hazards ratio (HR) was calculated based on the Kaplan–Meier survival method, and the HR value was scaled in decibel. (B) Formation of spheres under the stem cell selective condition on day 8 after culturing from parental OVS1. (C) Changes in gene expression of CREBBP, ACTL6A, ANXA5, EPB41L4A-AS1, FAM217A, IFI27L2, LNX2, MED28, TACR1, TLE4, and ZNF763 in a heat map. (D) All ovarian cancer patients revealed that patients with high expression of ACTL6A had worse overall survival (OS) compared to the ACTL6A low expression patients. (E) All ovarian cancer patients revealed that patients with high expression of ACTL6A had worse progression-free survival (PFS) compared to the ACTL6A low expression patients. (F) Chemotherapy ovarian cancer patients revealed that patients with high expression of ACTL6A had worse overall survival (OS) compared to the ACTL6A low expression patients. (G) Chemotherapy ovarian cancer patients revealed that patients with high expression of ACTL6A had worse progression-free survival (PFS) compared to the ACTL6A low expression patients.

Figure 3

Most cell cycle genes in positive correlation and focal adhesion genes in negative correlation with ACTL6A in ovarian cancer. (A) Association results of ACTL6A in ovarian cancer (Number of patients = 303); Pearson test was performed to analyze LinkFinder. Volcano plots show the statistical association results of ACTL6A for all ovarian cancer DEGs. (B) Data for the top 50 positive association genes are visualized in a heat map. (C) Data for the top 50 negative association genes are visualized in a heat map. (D) In the bar plots, the color scale represents the FDR value, and the length of the bar represents the normalized enrichment score (NES) value: positive correlated, blue Number of patients = 303 > 0; and negative correlated, orange < 0. (E) GSEA results showing cell cycle is differentially enriched pathway in ACTL6A-related genes. NES, normalized enrichment score. (F) GSEA results showing focal adhesion is differentially enriched pathway in ACTL6A-related genes. NES, normalized enrichment score. (G) KEGG pathway annotations of the cell cycle pathway.

Figure 4

CP724714 as a therapeutic option in the context of high ACTL6A in ovarian cancer. Q-omics analysis was used to analyze cross-association scores for the identification of candidate drugs acting on ovarian cancer cells based on ACTL6A expression. (A) X-score: log(fold-change) of ACTL6A expression between samples of high and low response of target drug. Y-score: log(fold-change) of target drug response between samples of high and low ACTL6A expression. The blue dot hits a negative association with p-value(X) < 0.05 and p-value(Y) < 0.05. (B) The boxplots of −log (half maximal inhibitory concentration (IC50)) M of CP724714 in low (<50%) and high (>50%) ACTL6A expression. Fold-change: the difference of IC50 variable between low (<50%) and high (>50%) ACTL6A expression (C) The scatter plot showed that ACTL6A expression levels and IC50 of CP724714 in 30 ovary adenocarcinoma cell lines. The r value was calculated by Pearson’s correlation. (D) The yellow bar showed that 11 of 11 ovarian cancer patients show high/medium expression. The representative immunohistochemistry (IHC) staining of ACTL6A molecule with specific ACTL6A (CAB012188) antibody in a patient with ovarian cancer. Scale bar = 50 μm.