RESUMO
Aneuploidy is a hallmark of human cancer, yet the molecular mechanisms to cope with aneuploidy-induced cellular stresses remain largely unknown. Here, we induce chromosome mis-segregation in non-transformed RPE1-hTERT cells and derive multiple stable clones with various degrees of aneuploidy. We perform a systematic genomic, transcriptomic and proteomic profiling of 6 isogenic clones, using whole-exome DNA, mRNA and miRNA sequencing, as well as proteomics. Concomitantly, we functionally interrogate their cellular vulnerabilities, using genome-wide CRISPR/Cas9 and large-scale drug screens. Aneuploid clones activate the DNA damage response and are more resistant to further DNA damage induction. Aneuploid cells also exhibit elevated RAF/MEK/ERK pathway activity and are more sensitive to clinically-relevant drugs targeting this pathway, and in particular to CRAF inhibition. Importantly, CRAF and MEK inhibition sensitize aneuploid cells to DNA damage-inducing chemotherapies and to PARP inhibitors. We validate these results in human cancer cell lines. Moreover, resistance of cancer patients to olaparib is associated with high levels of RAF/MEK/ERK signaling, specifically in highly-aneuploid tumors. Overall, our study provides a comprehensive resource for genetically-matched karyotypically-stable cells of various aneuploidy states, and reveals a therapeutically-relevant cellular dependency of aneuploid cells.
Assuntos
Aneuploidia , Dano ao DNA , Sistema de Sinalização das MAP Quinases , Ftalazinas , Humanos , Sistema de Sinalização das MAP Quinases/efeitos dos fármacos , Ftalazinas/farmacologia , Linhagem Celular Tumoral , Piperazinas/farmacologia , Quinases raf/metabolismo , Quinases raf/genética , Neoplasias/genética , Neoplasias/metabolismo , Neoplasias/patologia , Sistemas CRISPR-Cas , Linhagem Celular , Proteínas Proto-Oncogênicas c-raf/metabolismo , Proteínas Proto-Oncogênicas c-raf/genética , Resistencia a Medicamentos Antineoplásicos/genéticaRESUMO
Aneuploidy results in a stoichiometric imbalance of protein complexes that jeopardizes cellular fitness. Aneuploid cells thus need to compensate for the imbalanced DNA levels by regulating their RNA and protein levels, but the underlying molecular mechanisms remain unknown. Here, we dissected multiple diploid vs. aneuploid cell models. We found that aneuploid cells cope with transcriptional burden by increasing several RNA degradation pathways, and are consequently more sensitive to the perturbation of RNA degradation. At the protein level, aneuploid cells mitigate proteotoxic stress by reducing protein translation and increasing protein degradation, rendering them more sensitive to proteasome inhibition. These findings were recapitulated across hundreds of human cancer cell lines and primary tumors, and aneuploidy levels were significantly associated with the response of multiple myeloma patients to proteasome inhibitors. Aneuploid cells are therefore preferentially dependent on several key nodes along the gene expression process, creating clinically-actionable vulnerabilities in aneuploid cells.
RESUMO
Precision in the diagnosis of diverse central nervous system (CNS) tumor types is crucial for optimal treatment. DNA methylation profiles, which capture the methylation status of thousands of individual CpG sites, are state-of-the-art data-driven means to enhance diagnostic accuracy but are also time consuming and not widely available. Here, to address these limitations, we developed Deep lEarning from histoPathoLOgy and methYlation (DEPLOY), a deep learning model that classifies CNS tumors to ten major categories from histopathology. DEPLOY integrates three distinct components: the first classifies CNS tumors directly from slide images ('direct model'), the second initially generates predictions for DNA methylation beta values, which are subsequently used for tumor classification ('indirect model'), and the third classifies tumor types directly from routinely available patient demographics. First, we find that DEPLOY accurately predicts beta values from histopathology images. Second, using a ten-class model trained on an internal dataset of 1,796 patients, we predict the tumor categories in three independent external test datasets including 2,156 patients, achieving an overall accuracy of 95% and balanced accuracy of 91% on samples that are predicted with high confidence. These results showcase the potential future use of DEPLOY to assist pathologists in diagnosing CNS tumors within a clinically relevant short time frame.