| dc.description.abstract | Ovarian cancer is the most leathal cancer among gynecologic cancers in the United States and is the fifth leading cause of cancer death among American women. Approximately sixty percent of women with ovarian cancer are diagnosed with advanced stage diseases. High-grade serous ovarian carcinoma (HGS-OvCa) is the most common subtype of epithelial ovarian cancer and accounts for around 70% of all ovarian cancers in the U.S. Although the platinum-taxane combination has been the standard of care for treatment of ovarian cancer for over 15 years, an emerging science indicates poly (ADP-ribose) polymerase inhibitors (PARPi) are also active in a substantial portion of HGS-OvCa. Cancer cells that show “BRCAness” are highly sensitive to PARPi, including those with deleterious mutations in BRCA1/2 or defects in other components that are crucial for homologous recombination (HR) repair pathway. In HGS-OvCa, genes involved in HR are altered in about 50% of cases, making these cancers sensitive to PARPi. However, FOXM1 transcription factor network is activated in more than 84% of cases in HGS-OvCa, and activation of the FOXM1 pathway has been shown to upregulate genes involved in HR. Yet, the role of FOXM1 in PARPi response is not well studied. The MYC proto-oncogene is reported to be amplified in a significant number of epithelial ovarian cancer, and its role in the regulation of DNA repair and the PARPi response is not yet characterized. Therefore, we evaluated the roles of FOXM1 and MYC in PARPi response and explored ways to target them and develop new combinational therapies to overcome PARPi resistance. Besides, our understanding of PARPi resistance mechanism is not complete, and there is an urgent need to identify other potential regulators of PARPi response. So, we performed a genome-wide CRISPR knockout screen to identify novel regulators of PARPi response in ovarian cancer cells. To identify potential regulators of PARPi response, we first evaluated the role of FOXM1 and found that PARPi olaparib induced the expression and nuclear localization of FOXM1, suggesting FOXM1 might play an important role in the adaptive response to PARPi. Using ChIP-qPCR, we observed that olaparib treatment enhanced the binding of FOXM1 to genes involved in HR. Using qRT-PCR, we found FOXM1 knockdown by RNAi or treatment with thiostrepton, a natural peptide thiazole antibiotic, led to significantly decreased expression of FOXM1 and HR repair genes, such as BRCA1, FANCF, BRCC3, BRIP1, NBS1 and Csk1. Consequently, with short term SRB assay and long-term colony formation assay, both FOXM1 knockdown and inhibition by thiostrepton showed enhanced sensitivity to olaparib. Using comet and PARP trapping assays, we also observed increased DNA damage and PARP1 trapping in FOXM1-inhibited cells treated with olaparib. Finally, thiostrepton treatment lead to the decreased expression of BRCA1 in rucaparib-resistant breast cancer cells and enhanced sensitivity to rucaparib. Collectively, these results suggest that FOXM1 plays an important role in the adaptive response induced by olaparib and that FOXM1 inhibition by thiostrepton contributes to “BRCAness” and enhances sensitivity to PARP inhibitors. Second, we found that FOXM1 silencing results in c-MYC upregulation, and that FOXM1 knockdown in itself is not sufficient to downregulate several HR genes, including BRCA1, BRCA2, FANCF and BRIP1. Analysis of the ENCODE (Encyclopedia of DNA Elements) database found that both FOXM1 and c-MYC bind to overlapping regions in the promoters of HR genes, such as BRCA1 and RAD51, suggesting that these HR genes may be co-regulated by FOXM1 and c-MYC. In addition, using the TCGA (The Cancer Genome Atlas) database we discovered that FOXM1, c-MYC, and BRD4 are highly expressed in ovarian cancer and are related to each other, making them good therapeutic targets to explore. Using Western blot, we showed that BET bromodomain inhibitor, (+)-JQ1, suppresses c-MYC and FOXM1 expression as well as BRCA1 and RAD51 genes, and by ChIP-qPCR we saw decreased binding of both FOXM1 and c-MYC to their promoters, suggesting a direct regulation of these genes by FOXM1 and c-MYC. Treatment of 3 ovarian cancer cell lines (ES-2, OV90 and OVCA420*) with (+)-JQ1, lead to decreased cell viability in a dose-dependent manner for all cell lines. With SRB assay, we also observed mild to moderate synergistic effects of PARPi in both primary resistant ovarian cancer cells and acquired resistant cells derived from breast cancer MDA-MB-436 cell. Strikingly, adding sub-lethal doses of (+)-JQ1 and olaparib completely inhibited colony formation. Taken together, these data suggest that inhibition of FOXM1 and c-MYC by BET inhibitor, (+)-JQ1, induces “BRCAness” by downregulating BRCA1 and RAD51, and sensitizes resistant cancer cells to PARP inhibitors. Third, using the sensitivity of PARPi olaparib as a surrogate, we performed a CRISPR/Cas9 based genome-scale loss-of-function screen and identified C12orf5, a gene that encodes a metabolic regulator, TP53 induced glycolysis and apoptosis regulator (TIGAR) as a potential genetic determinant of PARPi response. To unveil the mechanisms involved in this process, further studies showed that TIGAR knockdown increases cell apoptosis, elevates the production of reactive oxygen species (ROS), arrests cells at S-phase, all of these contributes to accumulation of DNA damage and sensitizes cancer cells to olaparib. More interestingly, RNA-sequencing analysis showed that TIGAR knockdown leads to decreased expression of BRCA1 and GSEA analysis indicates that TIGAR knockdown unexpectedly produces a gene signature that is similar as BRCA1-downregulated cells. Furthermore, TIGAR knockdown results in downregulation of the Fanconi Anemia (FA) pathway. All of these findings provide us more information about the mechanisms involved in sensitization of olaparib after TIGAR knockdown. Interestingly, TIGAR knockdown inhibits cell growth by induction of cellular senescence as shown by SA-β-Gal staining analysis. Concomitantly, TIGAR knockdown resulted in less efficacy in spheroid formation and enhanced the therapeutic effects of olaparib. Finally, analysis of TCGA database demonstrated that TIGAR amplification was found in different types of cancer including ovarian cancer (amplification was seen in about 11% of cases), and that elevated expression levels of TIGAR mRNA is associated with poor overall survival in high-grade serous ovarian cancer. Taken together, our data suggest that TIGAR negatively regulates PARPi sensitivity and is a previously unappreciated therapeutic target in ovarian cancer. In summary, our studies showed that FOXM1 plays an important role in the adaptive response to PARPi and in its resistance by positively regulating HR repair pathway through upregulation of BRCA1, RAD51, and other HR genes, such as FANCD2, FANCF, RAD51D. Targeting FOXM1 with siRNA or treatment with thiostrepton induces a “BRCAness” like phenotype and sensitizes both primary and acquired resistant cells to PARP inhibitors. The combination of FOXM1 inhibitor, such as thiostrepton, and PARPi represents a new strategy to overcome PARPi resistance and in turn enhance their cytotoxic effects. We also showed that c-MYC works together with FOXM1 to regulate HR genes, including BRCA1 and RAD51. Inhibition of FOXM1 and c-MYC by BET inhibitor downregulates HR genes, such as BRCA1 and RAD51, and enhances cytotoxic effects of PARPi in resistant cells, suggesting that combination of BET inhibitor with PARPi can be another effective way to overcome PARPi resistance in HR proficient cancer cells. Lastly but not least, we identified TIGAR as a novel regulator of PARPi response. Our data suggest that TIGAR negatively regulates PARPi sensitivity, and TIGAR knockdown leads to increased cytotoxicity to olaparib by enhancing DNA damage, induction of apoptosis, and most importantly induction of “BRCAness” by downregulating BRCA1 and the FA pathway. We also found that efficient TIGAR knockdown leads to cellular senescence. Collectively, targeting TIGAR could serve as a potential combination therapy to enhance PARPi sensitivity or as a monotherapy to induce dormancy in tumors to prevent cancer progression. | |
