Studies of nucleoprotein complexes involved in retroviral DNA integration

National Institute Of Diabetes And Digestive And Kidney Diseases

Investigators

Linked publications, trials & patents

Abstract

The goal of the project is to understand the detailed molecular mechanism of HIV-1 DNA integration, the structures of the nucleoprotein complexes that mediate DNA integration, the mechanism of action of integrase inhibitors, and how the virus can the virus can evolve resistance to these inhibitors. Integration of a DNA copy of the viral genome into cellular DNA is an essential step for replication of HIV-1 and other retroviruses. Integration is mediated by the virally encoded integrase protein in complex with viral and target DNA; complexes of integrase associated with a pair of viral DNA are collectively called intasomes. The first intasome on the integration reaction pathway is Stable Synaptic Complex (SSC) intasome that comprises a complex of integrase and the pair of viral DNA ends. Integrase cleaves two nucleotides from each 3' end of the viral DNA (3' end processing) within the SSC and then integrates these 3' ends into target DNA (DNA strand transfer) to form the Strand Transfer Complex (STC) intasome. The FDA has recently approved four drugs, Raltegravir, Elvitegravir, Dolutegravir, and Bictegravir that target HIV-1 integrase and more are in the pipeline. These drugs are highly effective and provide a new class of drugs for combination antiviral therapy. They specifically target the DNA strand transfer step of integration and bind to the assembled SSC intasomes after 3' end processing rather than free integrase protein. High-resolution structural studies of HIV-1 intasomes are therefore required to understand the detailed mechanism of action of inhibitors and mechanisms of escape by mutations that confer resistance. We have established conditions for in vitro assembly of HIV intasomes. The intasomes assembled in vitro mimic all the properties of the association of integrase with viral DNA in preintegration complexes (PICs) isolated from virus-infected cells. Structural studies of HIV-1 intasomes have been frustrated by aggregation of both integrase and intasomes. We have recently overcome these obstacles. Fusing of Sulfolobus solfataricus chromosomal protein (PDB: 1BNZ) to the N-terminus of HIV-1 integrase resulted in a hyperactive protein that assembled intasomes with improved solubility properties. We have also assembled intasomes for our structural studies on branched product DNA, a strategy we have previously validated with the closely related prototype foamy virus integrase. Although the intasomes appeared to be homogeneous as judged by gel filtration, attempts to crystallize were unsuccessful. We there initiated collaboration with Dmitry Lyumkis at the Salk Institute to determine their structure by cryo-EM.The small size of HIV-1 intasomes and the requirement for a high-ionic strength buffer containing glycerol present changes for cryo-EM. Nevertheless, we have obtained cryo-EM structures of HIV STC intasomes with a resolution ranging from 3.5 Angstroms near the core of the intasome to 4.5 Angstroms in peripheral regions. The overall structure is tetrameric and similar to the previously reported PFV intasome structures. The two inner subunits in the tetramer are mainly responsible for interactions with DNA. The C-terminal domains o the other subunits contribute to interactions with viral DNA, while the N-terminal domains of the outer subunits are disordered. In addition to the tetrameric intasomes, there is also a population of higher-order STC intasomes. The best resolved of the higher order intasomes is dodecameric. The dodecameric STC intasome has the same set of positionally conserved domains interacting with DNA, but the subunits to which they belong differ. Our HIV-1 STC intasome structure was facilitated by fusion of an additional domain (Sso7d) to the N-terminus of integrase. Multiple intasome species were observed, including tetramers and dodecamers, all with the same Conserved Intasome Core (CIC) domains around the DNA. In contrast, intasomes of other retroviruses appear to be more homogeneous, although the number of integrase protomers they contain ranges from four to sixteen, depending on the retrovirus. This raises the question of whether the Sso7d domain contributes to the heterogeneity we observe. To address this question we have successfully determined the structures of HIV-1 intasomes assembled with wild-type HIV-1 integrase without the additional Sso7d domain. Although the 4.5 angstrom resolution is too low to study the interaction with drugs, the same multiple species are observed as with the Sso7d fusion protein. This gives us confidence that the Sso7d is not perturbing intasome structures and that intasomes assembled with fusion protein, which are amenable to cryoEM studies at much higher resolution, are a suitable platform to study drug interactions. We have also shown that a fusion of a peptide derived from lens epithelium-derived growth factor (LEDGF) to the N-terminus of HIV-1 integrase greatly improves the biophysical properties of HIV-1 integrase and facilitates high-resolution structural studies. Currently approved drugs that inhibit HIV -1 DNA integration bind to Cleaved Synaptic Complex (CSC) intasomes. Having successfully determined structures of HIV-1 STC intasomes, our focus is now on structural studies of CSC intasomes with and without bound drugs to understand how drugs inhibit DNA integration and how the virus can evolve resistance. For this part of the project we are collaborating with Dmitry Lyumkis at Salk on the cryoEM and with Stephen Hughes and Terrence Burke at NCI on the virology and chemistry, respectively. We now have high-resolution structures of HIV-1 CSC intasomes in complex all the currently FDA approved integrase inhibitors and several potential lead compounds at earlier stages of development. These inhibitors all bind within a well-defined pocket. Our focus is now on understanding the mechanisms by which HIV-1 integrase can develop resistance. Fifteen of the most common resistance mutations are within 10 A of the active site; only six of these are conserved between PFV integrase (the closest conserved integrase for which structures were previously available), highlighting the importance of high-resolution structures of HIV-1 intasomes with bound drugs to understand their detailed mechanism of action. Our current efforts are focused on obtaining high-resolution structures of HIV-1 cleaved synaptic complex (CSC) intasomes with mutations that confer drug resistance, with and without bound drugs. The goal is to understand how these mutations confer drug resistance, how HIV-1 can evolve drug-resistance and aid in the design of new drugs. Some of these mutations have required the development of improved in vitro assembly methods and this has been successfully accomplished. We now have structures of many of the drug-resistant mutant intasomes we have selected for study determined by cryo-EM in the resolution range of 2.5 to 3.0 Angstroms. Prior to these structures, the best available surrogate for studying drug interactions with HIV-1 intasomes was structures of drugs bound to the related prototype foamy virus (PFV) intasomes. However there is considerable divergence between HIV-1 and PFV integrase many the residues that confer drug-resistance to HIV-1 integrase are not conserved in PFV integrase. The results highlight the importance of high-resolution structures of HIV-1 intasomes to understand the detailed mechanism of action drug-resistance mutations and mechanisms of drug resistance.