Non-Coding Transcriptomic and Epigenetic Changes Implicated in Ovarian Cancer Platinum Chemoresistance

Arun Krishnaraj

Mentors:  Dr. Sreejith Nair, Department of Oncology – Georgetown University Medical Center; Dr. Matthew McCoy, Innovation Center for Biomedical Informatics (ICBI) – Georgetown University Medical Center

Date/Time: August 23rd, 2022 at 1:40pm.

Abstract: Platinum (II) chemotherapeutics like cisplatin are widely utilized in the treatment of malignant cancers, acting through DNA-adduct formation and induction of apoptosis. Ovarian cancer is the second most common gynecologic cancer and most lethal gynecologic disease, primarily treated with the cisplatin analog carboplatin. Despite initial efficacy, up to 70% of women experience primary tumor recurrence within 6 months and are clinically characterized as platinum-resistant. Lack of targeted therapies and poor patient response to alternative monotherapies make platinum resistance a crucial area of research in ovarian cancer. Multiple mechanisms have been proposed to contribute to platinum resistance, including changes in drug influx/efflux, nuclear localization, or anti-apoptotic mechanisms. Studies have also provided evidence of cisplatin association with ribosomal and telomeric RNAs, providing for a cytoplasmic sequestration model; these mechanisms may modify cisplatin localization or DNA access patterns crucial to platinum chemosensitivity. Changes in nuclear condensate formation or improper cisplatin localization may be related to long non-coding RNAs (lncRNAs), which have been shown to play increasingly diverse roles in transcriptomic regulation, cell proliferation, and therapy response. The goal of this internship project is to evaluate the role of non-coding transcriptomic and epigenetic factors in modified cisplatin localization and cellular response in ovarian cancer. Integrated bioinformatics analysis of the genomic and transcriptomic changes which underlie platinum chemoresistance in ovarian cancer may provide insight to the mechanisms responsible and improve existing standards of care.

TCGA bulk RNA-Seq data for ovarian tumor samples was first used to identify differentially expressed (DE) coding transcripts and lncRNAs between primary and recurrent tumors. This analysis aimed to identify RNAs associated with tumor recurrence despite platinum chemotherapy but was limited by a lack of statistical power. A publicly available RNA-Seq dataset was then used to identify DE lncRNAs between platinum resistant and sensitive ovarian cancer cell lines. Paired histone modification ChIP-Seq data was used to identify changes in genomic enhancer and super-enhancer distribution correlated with resistance status; these data were combined to obtain a list of enhancer-mediated DE lncRNAs by platinum resistance status. Total RNA-Seq was performed for cisplatin resistant and sensitive cell lines in the presence/absence of the drug. Analysis identified a generally muted transcriptional effect of drug treatment in the resistant line, as well as a large number of DE lncRNAs between resistance groups. Based on the hypothesis of lncRNA-mediated dysregulation of cisplatin localization, DE lncRNAs were stratified by features previously associated with RNA phase-separation potential: processed transcript length and consensus m6a RNA modification sites. These putative phase-separating lncRNAs were combined with the aforementioned enhancer-mediated DE lncRNAs to obtain a set of platinum resistance-associated lncRNAs for further analysis. Further mechanistic studies employing super-resolution microscopy and genomic assays will evaluate these lncRNA candidates for their involvement in modified cisplatin localization and cellular response in platinum-resistant ovarian cancers. Ultimately, these lncRNA candidates may be profiled for utility as therapeutic targets to resensitize ovarian cancers to platinum chemotherapeutics or prevent resistance from developing.