HDACs control RUNX2 expression in cancer cells through redundant and cell context-dependent mechanisms

RUNX2 is a Runt-related transcription factor required during embryogenesis for skeletal development and morphogenesis of other organs including thyroid and breast gland. Consistent evidence indicates that RUNX2 expression is aberrantly reactivated in cancer and supports tumor progression. The mechanisms leading to RUNX2 expression in cancer has only recently began to emerge. Previously, we showed that suppressing the activity of the epigenetic regulators HDACs significantly represses RUNX2 expression highlighting a role for these enzymes in RUNX2 reactivation in cancer. However, the molecular mechanisms by which HDACs control RUNX2 are still largely unexplored. Here, to fill this gap, we investigated the role of different HDACs in RUNX2 expression regulation in breast and thyroid cancer, tumors that majorly rely on RUNX2 for their development and progression.

Overall, our data provide new insights into the molecular mechanisms controlling RUNX2 in cancer and consolidate the rationale for the use of HDACi as potential pharmacological strategy to counteract the pro-oncogenic program controlled by RUNX2 in cancer cells.

Proliferation assays and evaluation of RUNX2 mRNA levels by qRT-PCR were used to evaluate the effect of several HDACi and specific siRNAs on a panel of cancer cell lines. Moreover, ChIP and co-IP assays were performed to elucidate the molecular mechanism underneath the RUNX2 transcriptional regulation. Finally, RNA-sequencing unveiled a new subset of genes whose transcription is regulated by the complex RUNX2-HDAC6.

In this study, we showed that Class I HDACs and in particular HDAC1 are required for RUNX2 efficient transcription in cancer. Furthermore, we found an additional and cell-specific function of HDAC6 in driving RUNX2 expression in thyroid cancer cells. In this model, HDAC6 likely stabilizes the assembly of the transcriptional complex, which includes HDAC1, on the RUNX2 P2 promoter potentiating its transcription. Since a functional interplay between RUNX2 and HDAC6 has been suggested, we used RNA-Seq profiling to consolidate this evidence in thyroid cancer and to extend the knowledge on this cooperation in a setting in which HDAC6 also controls RUNX2 expression. Conclusions: Overall, our data provide new insights into the molecular mechanisms controlling RUNX2 in cancer and consolidate the rationale for the use of HDACi as potential pharmacological strategy to counteract the pro-oncogenic program controlled by RUNX2 in cancer cells.