Abstract 1207 Interaction of metabolic and epigenetic control in colon cancer

Colorectal cancer (CRC) is third most common cause of cancer mortality worldwide. More then 150 000 new cases and 50 000 deaths from CRC are expected in the US in 2023. Although CRC can be treated by surgery, chemotherapy or targeted therapy, patients with advanced disease are resistant to current treatments. Acetyl-CoA synthetase (ACSS2) is a nuclear and cytosolic enzyme that converts acetate to Acetyl coenzyme A (Ac-CoA). In CRC, ACSS2 expression is decreased in both primary tumors and metastases, and lower expression correlates with reduced patient survival. Since Ac-CoA is essential for lipid synthesis and histone acetylation, we hypothesized that CRC with low ACSS2 might be sensitized to inhibition of ACSS2 or processes dependent on Ac-CoA. DOT1L (disruptor of telomeric silencing-1-like) is the only enzyme that methylates H3K79, a histone mark that promotes transcription elongation and DNA damage repair. High DOT1L expression in CRC has been linked to shorter survival and DOT1L knockdown has been shown to reduce the growth of subcutaneous CRC xenografts. We have tested the effect of inhibiting both ACSS2 and DOT1L in human CRC cell lines. CRISPR inactivation of ACSS2 increased sensitivity to DOT1L inhibitors in CRC cells. Combining small molecule inhibitors of ACSS2 and DOT1L resulted in a synergistic reduction in proliferation in multiple CRC cell lines, but had less effect in a normal colon epithelial line. RNA-seq data from CRC cells treated with ACS22 or DOT1L inhibitors alone identified relatively few gene expression changes, whereas in cells treated with both inhibitors gene expression changes were much more numerous. Interestingly, the majority of genes involved in the cholesterol synthesis pathway is deregulated by ACSS2 and/or DOT1L inhibition. Using a cell cell-based luciferase assay with an SREBP-responsive promoter, we found effects of DOT1L and ACSS2 inhibitors consistent with those seen by RNA-seq. Based on the effect on the cholesterol pathway, we tested the effects of statins, to inhibit cholesterol synthesis, together with either ACSS2 or DOT1L inhibition. These results show a synergistic effect of ACSS2 inhibition and statins in all CRC cell lines tested, as well as synergy between statins and DOT1L inhibitors in a subset. We are currently testing whether the effect of statins with ACSS2 inhibitors is due to a requirement for cholesterol, or for other biosynthetic intermediates of the cholesterol synthesis pathway. In summary we have found a synergistic interaction between ACSS2 and DOT1L inhibitors, which has profound effects on gene expression and proliferation in CRC cells in vitro, and identified cholesterol synthesis as a downstream target. Future work will test these combinations of inhibitors in pre-clinical CRC models. Research is funded from the NIH grant (CA259571).