Reduced expression of SOX7 in ovarian cancer: a novel tumor suppressor through the Wnt/β-catenin signaling pathway

Ovarian cancer is one of the three most common cancers in female with high morbidity and mortality rates [1]. The disease is also among the most lethal cancers in females, primarily because of its aggressive metastasis within the peritoneal cavity. It is hardly detectable at the early stage, largely due to the lack of specific symptoms and reliable screening [2],[3]. Among all types, epithelial ovarian cancer (EOC) is the most common, accounting for 85% to 90% of ovarian cancers [4], and this type is highly lethal. This situation calls for the investigation of pathogenesis of EOC and the identification of molecular markers for early diagnosis or for use as targets in treatment.

SOX7, a member of subfamily SOXF along with SOX17 and SOX18, has been identified as a developmental regulator in hematopoiesis and cardiogenesis [5]–[9]. The human SOX7 gene is located at the chromosomal region of 8p23.1 and is approximately 7.7 kbps in length [10]; a frequently methylated CpG island on the promoter region of the SOX7 gene plays a role in regulating SOX7 gene expression. SOX7 is frequently down-regulated in many human cancers, such as prostate, colon, lung, and breast cancers, and its reduced expression often correlates with poor prognoses [11]–[13]. It is reported that SOX7 can directly bind β-catenin and negatively regulate its activity [14],[15].

The Wnt/β-catenin signal transduction pathway is widely acknowledged as playing a dominating role in human diseases, especially in the occurrence and development of cancer [16]. Although the molecular mechanism by which SOX7 produces its tumor suppressive effects has yet to be fully determined, SOX7 has been shown to interact with β-catenin and inhibit cell proliferation mediated by the Wnt signaling pathways. Evidence is accumulating to show that SOX7 may disrupt the transcriptional function of the β-catenin-TCF/LEF complex and inhibit the activity of Wnt target genes including cyclin D1, c-Myc and COX-2 [15],[17]. Additionally, recent studies indicate that SOX7 may act as a tumor suppressor [11],[13],[14]. Ectopic SOX7 expression is reported to have a regulating function against cell growth and promote apoptosis; meanwhile, SOX7 gene knockdown may lead to neoplastic transformation [14],[18]. As SOX7 negatively regulates the Wnt/β-catenin signaling pathway by impeding the transcriptional machinery of β-catenin/TCF/LEF-1, inactivation of SOX7 may be associated with the pathogenesis of cancers, e.g., endometrial cancer and prostate cancer [14],[15]. Additionally, SOX7 may be involved in aspirin-mediated growth inhibition of COX2 colorectal cancer cells, in which aspirin may up-regulate the expression of SOX7 by activating the MAPK [17]. There are also reports indicating that the decreased expression of SOX7 is an important feature of lung adenocarcinoma [12]. Based on whole genome analysis to measure the expression of SOX7 on a series of non-small cell lung cancers (NSCLC), investigators have shown that SOX7 is a novel tumor suppressor gene silenced in the majority of NSCLC [19]. Numerous studies also suggest that promoter methylation might be an important mechanism of SOX7 down-regulation in breast cancer and MDS [11],[13]. On the other hand, reports are also available to show that SOX7 mRNA is significantly up-regulated in cancer, such as in pancreatic, gastric, and esophageal cancer cell lines, as well as in primary gastric cancer [20]. To date, nevertheless, the contributions and molecular mechanisms of SOX7 in ovarian cancer are largely unknown.

In this study, we investigated the changes and possible roles of SOX7 in ovarian cancer using the microarray gene expression techniques and validated the results with clinical tissues. Both kinds of data showed that the expression of SOX7 was significantly reduced in ovarian cancer and the levels of expression were correlated with tumour progression. This effect might be mediated by the down-regulation of Wnt target genes cyclinD1 and COX2.

Ovarian cancer remains to be a leading cause of death from gynecological malignancies. It is a huge challenge of current basic and clinical research to seek novel molecular markers for more accurate and efficient use in early diagnosis, treatment or prognosis of ovarian cancer. In this study, we chose ovarian cancer to work on primarily due to the fact that this disease is so devastating in females and that to date relatively little has been done on SOX7 in ovarian cancer. Our results obtained from different platforms indicate that the expression levels of SOX7 were significantly reduced in all types of ovarian cancers studied here, though at different extents. LG is the most malignant among ovarian cancer and may progress from SBOT, while HG, much less malignant than LG, likely develops from other kinds of precursors, such as normal epithelium of ovary or distal fallopian tube [24]. HG is a well-differentiated neoplasm and closely resembles normal tissues in many ways, so its prognosis is generally much better than LG [25]. As a result, the gene expression levels of SOX7 showed opposite tendencies to malignancy degrees. Based on the findings in this study, we propose that SOX7 is a key factor during ovarian cancer progression and is a useful prognostic marker.

Of great significance, we found that SOX7 was negatively correlated with Cyclin D1 and could be a negative regulator of the Wnt/β-catenin pathway. Cyclin D1, which controls the cell cycle, is the target gene of β-catenin and plays an important role in ovarian cancer. It is known that the Wnt/β-catenin signaling pathway is activated in epithelial ovarian cancer [26] and strongly involved in ovarian cancer development [27]. As SOX7 has been reported to block the transcriptional function of the Wnt/β-catenin signaling pathway and inhibit the activity of Wnt target genes including cyclin D1, c-Myc and COX-2 [15],[17], we wondered whether there might be a correlation between SOX7 and these supposed target genes. Our results demonstrated that the expression levels of SOX7 and its targets, COX-2 and cyclin D1, have an inverse relationship, further supporting our hypothesis that SOX7 is a negative regulator in the Wnt/β-catenin signaling pathway in ovarian cancer. Although our data showed a marginal correlation of cyclin D1 with SOX7 (p = 0.0555) at the mRNA level, the correlation at the protein level was strong, so both had similar trends.

SOX7, a member of the SOX-F subfamily encoding transcription factors that have a pivotal role in cardiovascular development [8],has been implicated as a tumor suppressor in a variety of human cancers, e.g., colorectal, prostate, breast, liver and lung cancers [13],[14],[19],[28]. Additionally, SOX7 may have potential usage as an independent prognostic marker in prostate and lung cancers as well as MDS patients [11],[12],[29]. Several lines of evidence indicate SOX7 as a negative regulator of the Wnt/β-catenin signaling pathway [14],[15]. Of great interest, a recent report based on whole genomic copy number analysis demonstrates SOX7 as a novel tumor suppressor, which is silenced in the majority of non-small cell lung cancer (NSCLC) samples [19]. Molecules modulating SOX7 expression, such as miR-184 in hepatocellular carcinoma (HCC), have also been reported [28]. Another study published recently also showed that SOX7 could suppress HCC in vivo and in vitro[30]. There are also contradicting reports to demonstrate that SOX7 mRNA was significantly up-regulated in several human cancer cells [20]. These results have prompted us to hypothesize that SOX7 might be a true tumor suppressor but behaves differently in different cancers.

Our work reported here suggests, for the first time, that SOX7 may play an important role as a tumor suppressor in ovarian cancer progression. Our results also revealed SOX7 as a negative regulator in the Wnt/β-catenin signaling pathway in ovarian cancer. Although further studies are needed to elucidate the underlining mechanisms of interactions between SOX7 and the Wnt/β-catenin signaling pathway, our result demonstrates the suppressive function of SOX7 in the carcinogenic process of ovarian cancer.

HL designed the study, BL carried out bioinformatics analysis and drafted the manuscript, HYL and ZQY participated in sample collection and experimental process, QS, SYY, JJK, DY and KF contributed reagents/materials/analysis tools. SLL finalized the manuscript. All authors read and approved the final manuscript.