Identification of DLX1 as a FOXM1 downstream target in mediating ovarian cancer oncogenesis

Abstract

Emerging evidences have documented that aberrant expression of FOXM1 is closely associated with human cancers. A recent comprehensive genome analysis has revealed that FOXM1 signaling is one of the major pathways involved in ovarian cancer oncogenesis. However, the regulatory network of FOXM1 in exerting the metastatic phenotypes remains unknown. Therefore, the identification of FOXM1 downstream targets will assist in understanding of its molecular mechanism in ovarian cancer oncogenesis.

In this study, by bioinformatics and a series of functional analyses, we identified DLX1 as a novel target of FOXM1. Our results clearly demonstrated that enforced expression of FOXM1 (FOXM1B and FOXM1C) could increase DLX1 in mRNA and protein levels. Conversely, depletion of FOXM1 by Thiostrepton (FOXM1 specific inhibitor) or RNAi knockdown could reduce DLX1 expression. Importantly, we demonstrated that the changes of DLX1 expression were in concomitant with the expression of a positive control gene, Cyclin-D1. Additionally, the luciferase promoter assay further showed that there are two conserved FOXM1 binding sites TFBS1 and TFBS2 which located at -61~-52bp upstream and -737~727bp upstream of the transcription factor binding sites (TSS) of DLX1 promoter respectively. In comparison of two binding sites, the more conserved binding site, TFBS1, seems have higher importance of FOXM1 binding in DLX1 transcriptional activation.

Furthermore, our study using immunohistochemical and Q-PCR analyses showed that DLX1 was frequently up-regulated in ovarian cancer samples. Noticeably, clinicopathological analysis revealed that the upregulated DLX1 was significantly associated with not only the overexpressed FOXM1 (P=0.001) but also high grade ovarian cancer (P<0.001). Previous studies have reported that DLX1 is a homeobox transcription factor controlling neuron migration and proliferation in embryogenesis. However, the oncogenic functions of DLX1 are rarely reported. In this study, we revealed that DLX1 could promote ovarian cancer cell proliferation and cell migration which are the main phenomena found in high grade tumors. To the best of our knowledge, this is the first report showing the regulation of FOXM1 on DLX1 and the metastatic functions exerted by DLX1 in ovarian cancer cells. Although ovarian cancer cells are epithelial cell type which is different from neurons, the similar cell functions derived from DLX1 reflecting that both cell types share the similar signaling pathway of DLX1. However, further investigation on the downstream network of DLX1 and the in vivo tumorigenic capacities in ovarian cancer cells are warranted.

To conclude, we have identified DLX1 as a novel target of FOXM1 and frequently up-regulated in high grade ovarian cancer. The in vitro tumorigenic assay demonstrated DLX1 could promote cell proliferation and cell migration which are the metastatic properties usually found in high grade ovarian cancer. Therefore, these data highlight the possibilities of using DLX1 as a biomarker and therapeutic target in combating ovarian cancer in the future.