rDNA Chromatin Activity Status as a Biomarker of Sensitivity to the RNA Polymerase I Transcription Inhibitor CX-5461

Abstract

Hyperactivation of RNA polymerase I (Pol I) transcription of ribosomal RNA (rRNA) genes (rDNA) is a key determinant of growth and proliferation and a consistent feature of cancer cells. We have demonstrated that inhibition of rDNA transcription by the Pol I transcription inhibitor CX-5461 selectively kills tumor cells in vivo. Moreover, the first-in human trial of CX-5461 has demonstrated CX-5461 is well-tolerated in patients and has single-agent anti-tumor activity in hematologic malignancies. However, the mechanisms underlying tumor cell sensitivity to CX-5461 remain unclear. Understanding these mechanisms is crucial for the development of predictive biomarkers of response that can be utilized for stratifying patients who may benefit from CX-5461. The rDNA repeats exist in four different and dynamic chromatin states: inactive rDNA can be either methylated silent or unmethylated pseudo-silent; while active rDNA repeats are described as either transcriptionally competent but non-transcribed or actively transcribed, depending on the level of rDNA promoter methylation, loading of the essential rDNA chromatin remodeler UBF and histone marks status. In addition, the number of rDNA repeats per human cell can reach hundreds of copies. Here, we tested the hypothesis that the number and/or chromatin status of the rDNA repeats, is a critical determinant of tumor cell sensitivity to Pol I therapy. We systematically examined a panel of ovarian cancer (OVCA) cell lines to identify rDNA chromatin associated biomarkers that might predict sensitivity to CX-5461. We demonstrated that an increased proportion of active to inactive rDNA repeats, independent of rDNA copy number, determines OVCA cell line sensitivity to CX-5461. Further, using zinc finger nuclease genome editing we identified that reducing rDNA copy number leads to an increase in the proportion of active rDNA repeats and confers sensitivity to CX-5461 but also induces genome-wide instability and sensitivity to DNA damage. We propose that the proportion of active to inactive rDNA repeats may serve as a biomarker to identify cancer patients who will benefit from CX-5461 therapy in future clinical trials. The data also reinforces the notion that rDNA instability is a threat to genomic integrity and cellular homeostasis.