Meta-analysis based gene expression profiling reveals functional genes in ovarian cancer

Abstract

Background: Ovarian cancer causes high mortality rate worldwide, and despite numerous attempts, the outcome for patients with ovarian cancer are still not well improved. Microarray-based gene expressional analysis provides with valuable information for discriminating functional genes in ovarian cancer development and progression. However, due to the differences in experimental design, the results varied significantly across individual datasets.

Methods: In the present study, the data of gene expression in ovarian cancer were downloaded from Gene Expression Omnibus (GEO) and 16 studies were included. A meta-analysis based gene expression analysis was performed to identify differentially expressed genes (DEGs). The most differentially expressed genes in our meta-analysis were selected for gene expression and gene function validation.

Results: A total of 972 DEGs with P-value < 0.001 were identified in ovarian cancer, including 541 up-regulated genes and 431 down-regulated genes, among which 92 additional DEGs were found as gained DEGs. Top five up- and down-regulated genes were selected for the validation of gene expression profiling. Among these genes, up-regulated CD24 molecule (CD24), SRY (sex determining region Y)-box transcription factor 17 (SOX17), WFDC2, epithelial cell adhesion molecule (EPCAM), innate immunity activator (INAVA), and down-regulated aldehyde oxidase 1 (AOX1) were revealed to be with consistent expressional patterns in clinical patient samples of ovarian cancer. Gene functional analysis demonstrated that up-regulated WFDC2 and INAVA promoted ovarian cancer cell migration, WFDC2 enhanced cell proliferation, while down-regulated AOX1 was functional in inducing cell apoptosis of ovarian cancer.

Conclusion: Our study shed light on the molecular mechanisms underlying the development of ovarian cancer, and facilitated the understanding of novel diagnostic and therapeutic targets in ovarian cancer.

Keywords: Differentially expressed genes; Gene expression omnibus; Microarray analysis; Ovarian cancer.

Figure 1. Flowcharts for microarray datasets selection and meta-analysis

(A) Selection process of microarray datasets for meta-analysis of gene expressional signature in ovarian cancer. (B) Process of meta-analysis based data exploration.

Figure 2. DEGs identified by meta-analysis

(A) PCA-3D plot for sample clustering of microarray datasets without batch effect adjustment. (B) PCA-3D plot for sample clustering of microarray datasets after batch effect adjustment. (C) Venn diagram of DEGs by meta-analysis (Meta-DE) and individual microarray dataset analysis (Individual-DE). (D) Heat-map visualization of expressional profiles for top 25 up- and down-regulated DEGs identified by meta-analysis. Genes were ranked by combined ES value, and the representative heat-map was generated by ‘Define Custom Signatures’ tool of NetworkAnalyst. Class of red: healthy donor; class of green: patient.

Figure 3. Gene expressional validation for top five up- and down-regulated genes

Relative genes levels of ovarian cancer and normal tissues measured by qPCR were shown using waterfall plot. (A–F) Represent the expression of CD24, SOX17, INAVA, WFDC2, EPCAM, AOX1 respectively. The fold change of relative genes expression (log2[T/N]) > 1 or < −1 was defined as significant. N, normal control; T, tumor of ovarian cancer.

Figure 4. WFDC2 knockdown inhibits proliferation of ovarian cancer cells

(A,B) Growth curve of OVCAR-3 and CAOV-3 cells upon transfection with NC and si-WFDC2. Transfected cells were counted and plated in 96-well plates. ***P<0.001, ****P<0.0001.

Figure 5. AOX1 knockdown inhibits apoptosis of ovarian cancer cells

(A,B) Flow cytometry analysis for apoptotic cells affected by si-AOX1 transfection in OVCAR-3 and CAOV-3 cells. Annexin V was labeled by FITC conjugated antibodies, and the DNA was stained by Propidium Iodide. *P<0.05, **P<0.01, ***P<0.001.

Figure 6. INAVA and WFDC2 knockdown inhibit migration of ovarian cancer cells

Down-regulation of INAVA and WFDC2 in OVCAR-3 and CAOV-3 exhibited inhibition of migration (magnification, ×100). Scale bar = 100 μm. The results are representative of three separate experiments.