Differential Expression of E2F Transcription Factors and Their Functional and Prognostic Roles in Human Prostate Cancer

Abstract

Given the tumor heterogeneity, most of the current prognostic indicators cannot accurately evaluate the prognosis of patients with prostate cancer, and thus, the best opportunity to intervene in the progression of this disease is missed. E2F transcription factors (E2Fs) have been reported to be involved in the growth of various cancers. Accumulating studies indicate that prostate cancer (PCa) carcinogenesis is attributed to aberrant E2F expression or E2F alteration. However, the expression patterns and prognostic value of the eight E2Fs in prostate cancer have yet to be explored. In this study, The Cancer Genome Atlas (TCGA), Kaplan-Meier Plotter, Metascape, the Kyoto Encyclopedia of Genes and Genomes (KEGG), CIBERSORT, and cBioPortal and bioinformatic analysis were used to investigate E2Fs in patients with PCa. Our results showed that the expression of E2F1-3, E2F5, and E2F6 was higher in prostate cancer tissues than in benign tissues. Furthermore, elevated E2F1-3 and E2F5 expression levels were associated with a higher Gleason score (GS), advanced tumor stage, and metastasis. Survival analysis suggested that high transcription levels of E2F1-3, E2F5, E2F6, and E2F8 were associated with a higher risk of biochemical recurrence. In addition, we developed a prognostic model combining E2F1, E2F6, Gleason score, and the clinical stage that may accurately predict a biochemical recurrence-free survival. Functional enrichment analysis revealed that the E2F family members and their neighboring genes were mainly enriched in cell cycle-related pathways. Somatic mutations in different subgroups were also investigated, and immune components were predicted. Further experiments are warranted to clarify the biological associations between Pca-related E2F family genes, which may influence prognosis via the cell cycle pathway.

Keywords: E2F transcription factors; biochemical recurrence; bioinformatic analysis; prognosis; prostate cancer.

FIGURE 1

The transcription levels of E2F family members in prostate cancer patients. (A) Analysis of E2F gene family alterations in prostate cancer (cBioPortal). (B) Comparison of mRNA expression between benign tissues and prostate cancer tissues. (C) Analysis of the relationship between E2F expression and Gleason score in PCa patients. **p < 0.01, *p < 0.05, ns: p > 0.05.

FIGURE 2

Correlation between E2F mRNA levels and clinicopathological features. (A) Correlation between mRNA levels and tumor stage in PCa patients. (B) Transcription levels in PCa patients with metastasis and without metastasis. **p < 0.01, *p < 0.05, ns: p > 0.05.

FIGURE 3

Prognostic value of E2Fs in patients with prostate cancer. (A) BCR-free survival curves in prostate cancer patients with high and low expressions of E2Fs. (B) Overall survival curves in prostate cancer patients with high and low expressions of E2Fs.

FIGURE 4

The prognostic value of E2F expression combined with clinical features in prostate cancer. (A) Stepwise Cox regression analysis of E2Fs, Gleason score, and pathological tumor stage in the TCGA-PRAD cohort. Nomograms (B), including the calibration plots (C) for the prediction of relapse-free survival (RFS) for PCa patients at 3 and 5 years. (D) BCR-free survival curves in prostate cancer patients with high and low risk scores. (E) AUC curves for the ability of the risk score to predict 1-, 3- and 5-year RFS in prostate cancer patients.

FIGURE 5

Landscape of mutation profiles in PCa patients with an aberrant expression of E2Fs. (A–J) Waterfall plots represent mutation information in each PCa patient sample in the high and low E2F1, E2F2, E2F3, E2F5, and E2F6 expression groups.

FIGURE 6

Intratumoral immune cell composition analysis. (A) The proportions of 22 immune cell types in prostate cancer from TCGA datasets. (B–F) The boxplot shows the different levels of 22 infiltrating immune cell types in groups with high and low E2F1, E2F2, E2F3, E2F5, and E2F6 expressions. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

FIGURE 7

Functional enrichment analysis of E2Fs and neighboring genes in prostate cancer. (A) The top 20 related genes of E2F1, E2F2, E2F3, E2F5, and E2F6 in TCGA-PRAD. (B) Protein–protein interaction (PPI) network among top E2F-related genes made by STRING. (C) Bar graph of the enriched terms across E2F genes colored by p values. (D) Network of enriched GO terms colored by cluster ID (left) and by p value (right). (E) The three most significant MCODE components from the PPI network (Metascape).

FIGURE 8

Cell cycle and prostate cancer pathways regulated by the altered E2Fs in prostate cancer. (A) Cell cycle pathways regulated by the altered E2Fs in prostate cancer. (B) Prostate cancer progression pathways regulated by the altered E2Fs in prostate cancer.