OCT4 promotes tumorigenesis and inhibits apoptosis of cervical cancer cells by miR-125b/BAK1 pathway

Abstract

Octamer-binding transcription factor 4 (OCT4) is a key regulatory gene that maintains the pluripotency and self-renewal properties of embryonic stem cells. Although there is emerging evidence that it can function as oncogene in several cancers, the role in mediating cervical cancer remains unexplored. Here we found that OCT4 protein expression showed a pattern of gradual increase from normal cervix to cervical carcinoma in situ and then to invasive cervical cancer. Overexpression of OCT4 in two types of cervical cancer cells promotes the carcinogenesis, and inhibits cancer cell apoptosis. OCT4 induces upregulation of miR-125b through directly binding to the promoter of miR-125b-1 confirmed by chromatin immunoprecipitation analysis. MiRNA-125b overexpression suppressed apoptosis and expression of BAK1 protein. In contrast, miR-125b sponge impaired the anti-apoptotic effect of OCT4, along with the upregulated expression of BAK1. Significantly, Luciferase assay showed that the activity of the wild-type BAK1 3'-untranslated region reporter was suppressed and this suppression was diminished when the miR-125b response element was mutated or deleted. In addition, we observed negative correlation between levels of BAK1 and OCT4, and positive between OCT4 and miR-125b in primary cervical cancers. These findings suggest an undescribed regulatory pathway in cervical cancer, by which OCT4 directly induces expression of miR-125b, which inhibits its direct target BAK1, leading to suppression of cervical cancer cell apoptosis.

Figure 1

OCT4 protein expression in normal cervical (NC) tissue and cervical cancer lesions. (a) Representative immunostained specimens showing OCT4 expression in NC (n=42), cervical carcinoma in situ (CIS; n=20) and invasive cervical carcinoma (ICC; n=44). Scale bars=20 μm. (b) Bar chart showing the frequency of OCT4-positive samples among different cervical lesions. (c) The scatter plot shows the immunoreactivity scores (IRS) for OCT4 staining in tissue samples (NC versus CIS, P=0.027; CIS versus ICC, P=0.0030; NC versus ICC, P=1.61E-09, one-way ANOVA test). (d) Western blot analysis of OCT4 protein levels in NC (n=8) and ICC (n=8) specimens. (e) Ratio of OCT4 to β-actin staining density in NC and ICC specimens determined with the AlphaView SA software (P=0.0055; Student's t-test)

Figure 2

OCT4 promotes tumor formation by cervical cancer cells in vivo. (a) OCT4 expression in cervical cancer cell lines was detected by immunocytochemistry and the positive rates were summarized. The arrow indicates the OCT4 positive cell. (b) OCT4 expression in cervical cancer cell lines was detected by western blotting. (c) Western blot analysis of OCT4 protein and in HeLa and SiHa parental, control- and OCT4-stably transfected cells. For western blot analyses, Tera-1 cells were used as a positive control, and β-actin served as the loading control. A tumor formation assay was performed with three mice per group. Tumor growth curves were calculated based on monitoring performed every 4 days post-transplant. At 36 days post-transplant for HeLa cells (d) and 32 days post-transplant for SiHa cells (e), the xenograft tumors were dissociated and weighed (f and g). The data were analyzed and shown as mean±S.E.M. (d and e were determined by two-way ANOVA test, *P<0.05 whereas f and g were determined by Student's t-test). (h) Serial sections of immunochemical staining for OCT4 and Ki-67 in indicated tumor xenografts. Scale bar=20 μm

Figure 3

Exogenous OCT4 expression inhibits apoptosis of human cervical cancer cells in vitro and in vivo. (a) Apoptosis analysis based on flow cytometry was performed triplicate. All of the data were summarized and presented as the mean±S.E.M. (Student's t-test), *P<0.05. (b) TUNEL assays were performed triplicate using the tumors from Figures 2d and e. Scale bar=50 μm for main image, 20 μm for inset; arrows indicate representative TUNEL+ cells. And the data were analyzed and presented as the mean±S.E.M. (Student's t-test), *P<0.05

Figure 4

Exogenous OCT4 expression inhibits apoptosis by transactivating miR-125b-1. (a) Apoptosis-related miRNAs were screened by real-time PCR in control and OCT4-expressing HeLa and SiHa cells. Real-time PCR analysis of pri-miR-125b-1 (b) and pri-miR-125b-2 (c) in OCT4-expressing and control cells. Data are presented as the mean±S.E.M. of triplicate analyses (Student's t-test). (d) Schematic structure of miR-125b-1 promoter. The pink and red ovals indicate the location of two potential OCT4-binding sites (RR1 and RR2). The bidirectional arrows (P1, P2 and P3) represent the primers used in ChIP analysis. Dual luciferase reporter assays in HeLa-GFP/OCT4 cells with a series of miR-125b-1 promoter 5′ deletion and rearrangement mutants. Data are presented as the mean±S.E.M. of triplicate analyses (Student's t-test). (e) Quantitative ChIP analysis of OCT4 binding to miR-125b-1 promoter sequences is shown for HeLa-GFP/OCT4 and SiHa-GFP/OCT4 cells. Data are presented as the mean±S.E.M. of triplicate analyses (Student's t-test). (f) Apoptosis assays were performed on OCT4-expressing or control cells following transfection with the miR-125b-expressing or -sponge vector. Data are presented as the mean±S.E.M. of triplicate analyses (Student's t-test)

Figure 5

Exogenous OCT4 expression downregulates BAK1 through the activation of miR-125b. (a) BAK1 expression was detected in the indicated cell lines by western blot analysis. (b) Highly conserved, predicted binding sites for the seed sequences of miR-125b among human (629-636), mouse (607-614) and rat (87-94) in the 3′-UTR (untranslated region) of BAK1. (c) BAK1 protein levels were determined by western blot analysis in HeLa-GFP or SiHa-GFP cells following transfection with the miR-125b-1-expressing or control vector and in HeLa-OCT4 or SiHa-OCT4 cells transfected with the miR-125b-sponge or control vector. (d) Schematic representation of the constructs used in the luciferase assay. HeLa-GFP/OCT4 cells were transfected with plasmid of BAK1wt, BAK1mut or BAK1del. Dual luciferase reporter activity was expressed relative to the pmiR-Report vector and data are presented as the mean±S.E.M. of triplicate analyses (Student's t-test). (e) Expression of OCT4 and BAK1 was examined by IHC staining in 20 cervical cancers specimens. The expression levels of two representative samples are shown. Scale bars=200 μm for upper panel, 50 μm for lower panel. (f) Nuclear OCT4 and cytoplasmic BAK1 staining were scored from 1 to 15. The correlation was significant, as determined by the Pearson's correlation test (r=−0.565; P=0.0094). (g) Expression of miR-125b was detected in 13 clinical samples from Figure 1 by real-time PCR and the correlation between OCT4 and miR-125b was significant, as determined by the Pearson's correlation test (r=0.849; P=0.0002)

Figure 6

A model of the OCT4/miR-125b/BAK1 pathway in human cervical carcinogenesis. OCT4 expression activates miR-125b-1. In turn, miR-125b-1 inhibits BAK1 translation, which inhibits apoptosis and promotes tumor progression