Cancer-associated mutations in the iron-sulfur domain of FANCJ affect G-quadruplex metabolism

Author summary Breast and ovarian cancers are often linked to a genetic predisposition, most commonly through mutations in the breast cancer susceptibility genesBRCA1andBRCA2, but also other genes, such asFANCJ/BRIP1, are associated with an increased disease risk. The small molecule CX-5461 is currently in phase I/II clinical trials for patients with BRCA-deficient tumours. It was originally identified as an rDNA transcription inhibitor, but-more recently-also found to bind and stabilise G4 DNA secondary structures. In this study we show that FANCJ-deficient cells have an increased sensitivity towards CX-5461. Our data further suggest that single amino acid changes in FANCJ that abolish its helicase activity or its ability to bind an iron-sulfur co-factor are sufficient to render cells more sensitive to CX-5461 treatment. Combined, our findings support a model whereby FANCJ resolves G4 structures in the context of the replisome to allow replication through guanine-rich regions of the genome. Mechanistically, FANCJ's ability to resolve G4 structures largely depends on an intact helicase domain and partially on the iron-sulfur cluster-binding domain. The latter finding is important since a number of cancer-associated mutations are located within the iron-sulfur domain ofFANCJ. FANCJ/BRIP1 is an iron-sulfur (FeS) cluster-binding DNA helicase involved in DNA inter-strand cross-link (ICL) repair and G-quadruplex (G4) metabolism. Mutations inFANCJare associated with Fanconi anemia and an increased risk for developing breast and ovarian cancer. Several cancer-associated mutations are located in the FeS domain ofFANCJ, but how they affect FeS cluster binding and/or FANCJ activity has remained mostly unclear. Here we show that the FeS cluster is indispensable for FANCJ's ability to unwind DNA substratesin vitroand to provide cellular resistance to agents that induce ICLs. Moreover, we find that FANCJ requires an intact FeS cluster for its ability to unfold G4 structures on the DNA template in a primer extension assay with the lagging-strand DNA polymerase delta. Surprisingly, however, FANCJ variants that are unable to bind an FeS cluster and to unwind DNAin vitrocan partially suppress the formation of replisome-associated G4 structures that we observe in aFANCJknock-out cell line. This may suggest a partially retained cellular activity of FANCJ variants with alterations in the FeS domain. On the other hand,FANCJknock-out cells expressing FeS cluster-deficient variants display a similar-enhanced-sensitivity towards pyridostatin (PDS) and CX-5461, two agents that stabilise G4 structures, asFANCJknock-out cells. Mutations inFANCJthat abolish FeS cluster binding may hence be predictive of an increased cellular sensitivity towards G4-stabilising agents.