Limited DNA end resection is the key to impaired homologous recombination (HR) in BRCA1mutant cells. Here through a loss-of-function CRISPR screen we identify DYNLL1 as an inhibitor of DNA end resection. Loss of DYNLL1 allows DNA end resection and restores HR in BRCA1mutant cells, thereby inducing resistance to platinum drugs and poly(ADP-ribose) polymerase

Highlights d Lung squamous cell carcinoma has a moderate level of intratumor genetic heterogeneity d Transcriptomic heterogeneity impacts cancer pathways, driving phenotypic heterogeneity d Neo-epitope burden negatively correlates with immune infiltration d Non-genetic heterogeneity influences tumor evolutionary dynamics

Inhibitors of poly-ADP-ribose polymerase 1 (PARPi) are highly effective in killing cells deficient in homologous recombination (HR); thus, PARPi have been clinically utilized to successfully treat BRCA2-mutant tumors. However, positive response to PARPi is not universal, even among patients with HR-deficiency. Here, we present the results of genome-wide CRISPR knockout and activation screens which reveal genetic determinants of PARPi response in wildtype or BRCA2-knockout cells. Strikingly, we report that depletion of the ubiquitin ligase HUWE1, or the histone acetyltransferase KAT5, top hits from our screens, robustly reverses the PARPi sensitivity caused by BRCA2-deficiency. We identify distinct mechanisms of resistance, in which HUWE1 loss increases RAD51 levels to partially restore HR, whereas KAT5 depletion rewires double strand break repair by promoting 53BP1 binding to double-strand breaks. Our work provides a comprehensive set of putative biomarkers that advance understanding of PARPi response, and identifies novel pathways of PARPi resistance in BRCA2-deficient cells.

Editing in microRNAs, particularly in seed can significantly alter the choice of their target genes. We show that out of 13 different human tissues, different regions of brain showed higher adenosine to inosine (A-to-I) editing in mature miRNAs. These events were enriched in seed sequence (73.33%), which was not observed for cytosine to uracil (17.86%) editing. More than half of the edited miRNAs showed increased stability, 72.7% of which had ΔΔG values less than −6.0 Kcal/mole and for all of them the edited adenosines mis-paired with cytosines on the pre-miRNA structure. A seed-editing event in hsa-miR-411 (with A – C mismatch) lead to increased expression of the mature form compared to the unedited version in cell culture experiments. Further, small RNA sequencing of GBM patients identified significant miRNA hypoediting which correlated with downregulation of ADAR2 both in metadata and qRT-PCR based validation. Twenty-two significant (11 novel) A-to-I hypoediting events were identified in GBM samples. This study highlights the importance of specific sequence and structural requirements of pre-miRNA for editing along with a suggestive crucial role for ADAR2. Enrichment of A-to-I editing in seed sequence highlights this as an important layer for genomic regulation in health and disease, especially in human brain.