Distinct states of nucleolar stress induced by anti-cancer drugs

Ribosome biogenesis is one of the most essential and energy-consuming cellular functions. It takes place mainly in the nucleolus. For cancer cells, the nucleolar function is especially important due to the high demand for ribosomes to support continuous proliferation. The goal of this study was to assess the effects of existing chemotherapy drugs on the nucleolar state. For this, we conducted an imaging-based screen for anticancer drugs that induce morphological re-organization consistent with nucleolar stress. For a readout, we developed a novel parameter termed "nucleolar normality score", which measures ratios of dense fibrillar center and granular component in the nucleolus and nucleoplasm. We show that multiple classes of drugs cause nucleolar stress, including DNA intercalators, inhibitors of mTOR/PI3K, heat shock proteins, proteasome, and cyclin-dependent kinases (CDKs). Different classes of drugs induced morphologically and molecularly distinct states of nucleolar stress. By applying phospho-proteomics and live imaging strategies, we characterized in detail the nucleolar stress induced by inhibition of transcriptional CDKs, particularly CDK9, the main CDK that targets RNA Pol II. Inhibition of CDK9 dramatically reduced rRNA production, caused dissociation of RNA Polymerase I catalytic subunit POLR1A from ribosomal DNA and dispersal of the nucleolar granular component, a stress we refer to as the "bare scaffold" state. We identified multiple nucleolar CDK phosphorylation substrates, including RNA Pol I - associated protein Treacle, and demonstrated that CDK9 can phosphorylate Treacle in vitro. This implies that transcriptional CDKs coordinate the action of RNA pol I and RNA pol II. Furthermore, molecular dynamics analysis of the endogenous nucleolar protein NPM1 demonstrated that CDK inhibition vastly increased its mobility, consistent with the loss of nucleolar integrity. We conclude that many classes of chemotherapy compounds directly or indirectly target nucleolar structure and function, and recommend considering this in anticancer drug development. ### Competing Interest Statement The authors have declared no competing interest.