Novel drugs targeting virus entry and maturation have recently been added to the growing armory of antiviral agents aimed at stemming the continuing spread of HIV infection. In addition, host factors participating in key steps in HIV replication are under consideration as targets for therapeutic intervention. Despite these promising approaches, the viral enzymes PR, RT, and IN remain primary targets of highly active antiretroviral therapy, with almost 20 drugs in use against PR and RT, and IN inhibitors undergoing clinical trials. With respect to HIV-1 RT, our section combines biochemistry and biophysics with structural, molecular, and chemical biology to provide molecular details of its role in converting RNA of the invading virus into integration-competent double-stranded DNA. Using nucleoside and amino acid analogs to provide novel acumen into the structures of complexes representing replication intermediates complements traditional nucleic acid and protein mutagenesis methods. Chemical nucleic acid footprinting is now complemented with direct mass spectrometric analysis, while single-molecule fluorescence studies have been introduced to probe the orientation of RT on a variety of nucleic acid duplexes.Site-specific introduction of unnatural amino acids with novel biophysical properties into proteins by translational suppression permits a structure/function analysis with an exceptionally high level of resolution. Our previous studies concentrated on analyzing HIV-1 RT function in response to targeted insertion of unnatural amino acids with different biophysical properties. We are now extending this technology into a more biological context by developing methods to study the interaction of viral and host proteins. The ability of an HIV-1 CA-NC fusion protein to assemble, in the presence of RNA, into structures that morphologically resemble authentic cores was reported by Ganser et al. (Science 1999). Subsequently, CA-NC tubes have been used to understand host factor-mediated restriction of core dissembly. Using the translational suppression technology developed by Chin et al. (PNAS 2002), p-benzoyl-l-phenylalanine (Bpa) was site-specifically inserted into CA at several positions of recombinant CA-NC tubes expressed in E. coli. In the presence of RNA, these mutants assemble into photoactive tubes that are reasonable mimics of unmodified structures. In contrast to traditional methods such as immunoprecipitation, photocrosslinking has the advantage that it immediately identifies the binding region on CA-NC and accelerates identification of the binding partner and its point of contact by mass spectrometry. In addition, weak and transient interactions are more likely to be detected by photocrosslinking. As proof-of-principle, we will insert Bpa at known interaction sites for host proteins such as cyclophilin and CPSF6. Thereafter, Bpa-containing CA-NC tubes will be incubated with cell lysates and irradiated to search for additional host restriction factors using a combination of immunoprecipitation and mass spectrometry. This project will be performed in collaboration with Dr. Vineet KewalRamani, whose group will provide cell culture facilities and training.Although we can reconstitute Bpa-containing CA-NC tubes that are reasonable morphological mimics of the wild-type structure, there are potential pitfalls to this approach: (1) Bpa insertion, while site specific, may induce a conformational change that inhibits host factor binding;(2) the site of Bpa insertion may itself overlap with a binding site;and (3) Bpa insertion may preferentially induce crosslinked CA-NC dimers. We therefore propose a complementary strategy that exploits the photoactive amino acids photo-Leu and photo-Met. Similarity of these diazirine-containing amino acids to their natural counterparts overcomes identity control elements of the eukaryotic cell, allowing them to be incorporated into proteins by the unmodified translational machinery. Photo-Leu and photo-Met, which are commercially available, have been used by other groups to examine protein:protein interactions in living cells. Our strategy will therefore be to culture the appropriate eukaryotic cell lines in medium depleted of Leu and Met but containing their photoactive counterparts. These cell lysates will be incubated with wild-type CA-NC tubes and irradiated at 365 nm to identify interacting partners. CA-NC tubes will be recovered either by immunoprecipitation or rate, zone sedimentation and analyzed by mass spectrometry for the interacting partner and site of interaction.Despite significant effort, we have not obtained vinylogous urea-containing cocrystals of HIV-1 RT. Although HIV-1 RT mutants or related lentiviral enzymes (EIAV) with increased vinylogous urea sensitivity may facilitate crystallization, we have implemented an alternative solution strategy to provide information on conformational changes in the immediate vicinity of the vinylogous urea binding site following inhibitor binding. This strategy exploits site-specific incorporation of the fluorinated Phe derivative OCF3-Phe. We replaced p66 residue Tyr501 with OCF3-Phe without loss of DNA polymerase and RNase H activity and obtained the 19F spectrum of the mutant at an enzyme concentration of 50 uM. Experimentally, this required purifying 3 mg of reconstituted, selectively mutated HIV-1 RT, which is well within our capacity. Mutant enzymes containing aromatic ->OCF3-Phe substitutions in the p51 thumb (e.g., Trp266, Tyr271, Tyr318, Tyr319) or the p66 RNase H domain in the vicinity of catalytic residues (e.g., Tyr441, Tyr532, Tyr483) will be prepared, analyzed for retention of DNA polymerase and RNase H activity, and examined by 19F-NMR in the absence and presence of inhibitor. We will perform NMR studies in collaboration with Dr. Angela Gronenborn, where a 19F cryoprobe that provides spectra at protein concentrations of 10-20 uM is available.Another strategy that we are pursuing in this project is designed in conjunction with single-molecule spectroscopy studies of Project ZIA BC 011443, the goal of which is to retain the natural cysteines of p66 and p51 HIV-1 RT (Cys38 and Cys280) and introduce Phe->azido-Phe (Az-Phe) substitutions in order to introduce an alkynyl-fluorophore (Kolb et al., Angew. Chem. Int. Ed. Engl. 2001). Given the abundance of aromatic residues in RT, this approach gives considerably more flexibility than targeted Cys insertions into a Cys-free enzyme. Secondly, we propose extending single-molecule spectroscopy studies to include other lentiviral and gammaretroviral RTs, where elimination of a significantly larger number of cysteines (8 in MLV/XMRV RT, 7 in EIAV RT) becomes impractical. As an alternative strategy Az-Phe will be site-specifically introduced into p66 or p51 RT via translational suppression for attachment of either alkynyl Cy3 or Cy5 by Click chemistry, which involves coupling of azide and alkyne functions to produce a 1,2,3-triazole. In response to a call for proposals, alkynyl Cy3 and alkynyl Cy5 were synthesized for us by Dr. Gary Griffiths, Image Probe Development Center, NIH. As proof-of-principle, we will introduce Az-Phe at the N- or C-terminus of p66 and compare conformational dynamics of the singly labeled RT with the Cys-labeled counterpart, for which we have considerable data. This strategy opens the possibility of establishing three-color FRET, with fluorophores on each RT subunit and a third on the nucleic acid substrate, to examine conformational changes within the enzyme that accompany translocation on its nucleic acid, akin to studies performed with E. coli RNA polymerase by Lee et al. (Biophys. J. 2007).[Corresponds to Le Grice Project 3 in the October 2011 site visit report of the HIV Drug Resistance Program]
