Three strikes to knock cancer out: BRCA1 and BRCA2 are tumor-suppressor genes, and patients with mutations in these genes are predisposed to breast, ovarian, and other cancers. Because BRCA1 and BRCA2 mutations affect pathways involved in DNA break repair, these patients' tumors are usually vulnerable to treatments that further damage DNA repair, such as poly(ADP-ribose) polymerase (PARP) inhibitors, but they can acquire resistance to therapy. Using a genome-wide screening approach, Fugger et al. identified a protein called DNPH1 as a “nucleotide sanitizer” that prevents the incorporation of abnormal nucleotides into DNA (see the Perspective by Kriaucionis). The authors examined its mechanism of action and demonstrated how it can be targeted to expedite the killing of BRCA1 -mutant cancer cells in combination with PARP inhibitor treatment. Science , this issue p. 156 ; see also p. 127

Exosomal miRNA transfer is a mechanism for cell–cell communication that is important in the immune response, in the functioning of the nervous system and in cancer. Syncrip/hnRNPQ is a highly conserved RNA-binding protein that mediates the exosomal partition of a set of miRNAs. Here, we report that Syncrip’s amino-terminal domain, which was previously thought to mediate protein–protein interactions, is a cryptic, conserved and sequence-specific RNA-binding domain, designated NURR (N-terminal unit for RNA recognition). The NURR domain mediates the specific recognition of a short hEXO sequence defining Syncrip exosomal miRNA targets, and is coupled by a non-canonical structural element to Syncrip’s RRM domains to achieve high-affinity miRNA binding. As a consequence, Syncrip-mediated selection of the target miRNAs implies both recognition of the hEXO sequence by the NURR domain and binding of the RRM domains 5′ to this sequence. This structural arrangement enables Syncrip-mediated selection of miRNAs with different seed sequences.

Microtubules are cytoskeletal polymers whose function depends on their property to switch between states of growth and shrinkage. Growing microtubules are thought to be stabilized by a GTP cap at their ends. The nature of this cap, however, is still poorly understood. End Binding proteins (EBs) recruit a diverse range of regulators of microtubule function to growing microtubule ends. Whether the EB binding region is identical to the GTP cap is unclear. Using mutated human tubulin with blocked GTP hydrolysis, we demonstrate that EBs bind with high affinity to the GTP conformation of microtubules. Slowing-down GTP hydrolysis leads to extended GTP caps. We find that cap length determines microtubule stability and that the microtubule conformation changes gradually in the cap as GTP is hydrolyzed. These results demonstrate the critical importance of the kinetics of GTP hydrolysis for microtubule stability and establish that the GTP cap coincides with the EB-binding region.

SAMHD1 restricts HIV-1 infection of myeloid-lineage and resting CD4⁺ T-cells. Most likely this occurs through deoxynucleoside triphosphate triphosphohydrolase activity that reduces cellular dNTP to a level where reverse transcriptase cannot function, although alternative mechanisms have been proposed recently. Here, we present combined structural and virological data demonstrating that in addition to allosteric activation and triphosphohydrolase activity, restriction correlates with the capacity of SAMHD1 to form “long-lived” enzymatically competent tetramers. Tetramer disruption invariably abolishes restriction but has varied effects on in vitro triphosphohydrolase activity. SAMHD1 phosphorylation also ablates restriction and tetramer formation but without affecting triphosphohydrolase steady-state kinetics. However phospho-SAMHD1 is unable to catalyse dNTP turnover under conditions of nucleotide depletion. Based on our findings we propose a model for phosphorylation-dependent regulation of SAMHD1 activity where dephosphorylation switches housekeeping SAMHD1 found in cycling cells to a high-activity stable tetrameric form that depletes and maintains low levels of dNTPs in differentiated cells. SAMHD1 is a restriction factor that blocks infection of certain immune cells by HIV-1. It was discovered to be an enzyme that catalyses the breakdown of dNTPs, suggesting that it inhibits HIV-1 replication by reducing cellular dNTP pools to such low levels that reverse transcriptase cannot function. However, recently, alternative mechanisms have been proposed. SAMHD1 is also regulated by phosphorylation, although the effects of phosphorylation on protein function are unclear. In order to address these issues, we carried out combined structural and virological studies and have demonstrated that in addition to allosteric activation and triphosphohydrolase activity, restriction correlates with the capacity of SAMHD1 to form “long-lived” enzymatically competent tetramers. Disrupting the tetramer in various ways always abolished restriction but had differing effects on enzyme activity in vitro. SAMHD1 phosphorylation also prevented restriction and tetramer formation but without affecting enzyme catalysis under steady-state dNTP conditions. However phosphorylated SAMHD1 was unable to catalyse dNTP turnover at very low nucleotide levels that more accurately represent conditions in the cells in which restriction takes place. Based on our findings we propose a model for phosphorylation-dependent regulation of SAMHD1 activity and substantiate that degradation of dNTPs by SAMHD1 is sufficient to restrict HIV-1.

HIV-1 capsid influences viral uncoating and nuclear import. Some capsid is detected in the nucleus but it is unclear if it has any function. We reported that the antibiotic Coumermycin-A1 (C-A1) inhibits HIV-1 integration and that a capsid mutation confers resistance to C-A1, suggesting that capsid might affect post-nuclear entry steps. Here we report that C-A1 inhibits HIV-1 integration in a capsid-dependent way. Using molecular docking, we identify an extended binding pocket delimited by two adjacent capsid monomers where C-A1 is predicted to bind. Isothermal titration calorimetry confirmed that C-A1 binds to hexameric capsid. Cyclosporine washout assays in Jurkat CD4+ T cells expressing engineered human TRIMCyp showed that C-A1 causes faster and greater escape from TRIMCyp restriction. Sub-cellular fractionation showed that small amounts of capsid accumulated in the nuclei of infected cells and C-A1 reduced the nuclear capsid. A105S and N74D capsid mutant viruses did not accumulate capsid in the nucleus, irrespective of C-A1 treatment. Depletion of Nup153, a nucleoporin located at the nuclear side of the nuclear pore that binds to HIV-1 capsid, made the virus less susceptible to TRIMCyp restriction, suggesting that Nup153 may help maintain some integrity of the viral core in the nucleus. Furthermore C-A1 increased binding of CPSF6, a nuclear protein, to capsid. Our results indicate that capsid is involved in post-nuclear entry steps preceding integration.

RAD52 is important for the repair of DNA double-stranded breaks^1,2, mitotic DNA synthesis^3–5 and alternative telomere length maintenance^6,7. Central to these functions, RAD52 promotes the annealing of complementary single-stranded DNA (ssDNA)^8,9 and provides an alternative to BRCA2/RAD51-dependent homologous recombination repair¹⁰. Inactivation of RAD52 in homologous-recombination-deficient BRCA1- or BRCA2-defective cells is synthetically lethal^11,12, and aberrant expression of RAD52 is associated with poor cancer prognosis^13,14. As a consequence, RAD52 is an attractive therapeutic target against homologous-recombination-deficient breast, ovarian and prostate cancers^15–17. Here we describe the structure of RAD52 and define the mechanism of annealing. As reported previously^18–20, RAD52 forms undecameric (11-subunit) ring structures, but these rings do not represent the active form of the enzyme. Instead, cryo-electron microscopy and biochemical analyses revealed that ssDNA annealing is driven by RAD52 open rings in association with replication protein-A (RPA). Atomic models of the RAD52–ssDNA complex show that ssDNA sits in a positively charged channel around the ring. Annealing is driven by the RAD52 N-terminal domains, whereas the C-terminal regions modulate the open-ring conformation and RPA interaction. RPA associates with RAD52 at the site of ring opening with critical interactions occurring between the RPA-interacting domain of RAD52 and the winged helix domain of RPA2. Our studies provide structural snapshots throughout the annealing process and define the molecular mechanism of ssDNA annealing by the RAD52–RPA complex.