During FY18, we continued to focus our efforts on (1) the assembly of HIV virions, and (2) the structure of the Rev regulatory protein and its interactions with other macromolecules. 1) Retrovirus capsids are unusual in that they are assembled inside a specialized microcompartment - the maturing virion - not in the cytoplasm or nucleus of the infected cell. Capsid protein is incorporated into the precursor particle, the provirion, as part of the Gag polyprotein. After the provirion has budded off from the infected cell, maturation ensues whereby the viral protease dissects Gag into its matrix (MA), capsid (CA), and nucleocapsid (NC) domains and two spacer peptides, SP1 and SP2. CA assembles into the viral capsid, housing the RNA and NC. Evidence suggests that a correctly formed conical core is essential for HIV infectivity. There are three ways in which interference with maturation can inhibit the virus. One is Protease Inhibitors (PI), the first drugs to be used successfully against HIV. More recently, maturation inhibitors (MI) have been discovered that act by blocking the protease from access to its cleavage site in the SP1 spacer peptide. The class member MI is a compound called Beviramat (BVM). We used cryo-electron tomography to show that virions isolated from HIV-infected cells treated with BVM mostly lack capsids but have an incomplete shell of protein underlying the viral envelope, with a honeycomb structure resembling the Gag lattice of immature HIV but lacking the innermost layer that is associated with NC/RNA. These findings were published in 2011. Subsequently, we obtained data that strongly support the view that the conical capsids of wild-type HIV virions are assembled de novo inside maturing virions and not by a displacive transition. Virions produced in the presence of a second MI, PF-46396, resemble BVM-treated virions although this MI is less stringent in both CA-SP1 cleavage inhibition and in blocking infectivity. MI-treated virions have a shell that resembles the CA layer of the immature Gag shell but is less complete. We concluded that inhibitors like PF-46396 and BVM bind to the partially processed Gag lattice where they deny the protease access to the CA-SP1 cleavage site and prevent the release of CA. These findings were published in 2013. Next, we investigated the mode of action of integrase (IN) inhibitors (ALLINIs allosteric IN inhibitors) and certain mutants in the integrase gene that they phenocopy. Both kinds of virions were found to contain eccentric condensates, located outside empty and often malformed capsids. We were able to show that eccentric condensates have a high NC content by tomo-bubblegram imaging, a novel labeling technique that exploits NC's susceptibility to radiation damage. Tomo-bubblegrams also localized NC inside wild-type cores and lining the spherical Gag shell in immature virions. Based on these observations, we have proposed a role for IN in initiating core assembly and the incorporation of vRNP into the mature core. These findings were published in 2015. In FY18, we pursued two projects. In one, a continuing collaboration with E. Freed (NCI), we analyzed the role of a conserved Pro-Pro-Ile-Pro (PPIP) motif (CA residues 122-125 in the loop connecting helices 6 and 7 of CA) in HIV-1 assembly. Prior data indicated that the mutations P122A and I124A impair release, infectivity, and replication, and the T58S/T107I/P122A mutant restores wild-type-like (WT-like) infectivity. We used cryo-ET with subtomogram averaging to assess how these mutations affect assembly of the immature (protease-minus, PR-) virion. Our data show that the P122A and I124A mutations impair Gag lattice coordination and T58S/T107I/P122A reverts to a WT-like lattice. We concluded that the CA PPIP loop comprises a structural element critical for formation of the immature Gag lattice. A paper has been submitted for publication. The second project, a collaboration with G. Kalpana (Albert Einstein School of Medicine), addresses the role played by a host protein, INI1, in interacting with the viral integrase to guide capsid assembly. Integrase is known to enter nascent viral particles as the distal component of the Gag-Pol polyprotein. In earlier work (see above), we showed that mutations in integrase or exposure to certain integrase-inhibiting compounds result in a high percentage of malformed capsids. Our more recent studies have detected similar effects involving IN1. These results are part of a paper recently submitted for publication. 2) HIV Rev is a small regulatory protein that mediates the nuclear export of genomic viral mRNAs, an essential step in the HIV replication cycle. In the process, Rev oligomerizes in association with a structured RNA molecule, the Rev response element (RRE). The resulting complex engages with the nuclear export machinery of the host cell. Detailed structural information on this interaction is essential for the purpose of designing Rev-inhibiting antiviral drugs. For many years, crystallographic studies were thwarted by Rev's tendency to aggregate. However, we were able to construct a hybrid monoclonal antibody whose Fab forms a stable complex with Rev, and solved these co-crystals at 0.32 nm resolution. These results were published in FY 11. Our research on HIV Rev has continued with further exploitation of this antibody. In particular, a single-chain version (scFv) was found to co-crystallize with Rev, giving crystals that diffracted to significantly higher resolution. These crystals, coming in four different space groups, revealed essentially the same structure of the monomer, although the crossing angle of the Rev dimer varies widely from 90 to 140 degrees. We also performed cryo-EM studies of helical tubes that Rev assembles into in vitro. They exhibited polymorphism, with the tube diameter varying between 11 nm and 13 nm. These variations in tube width correlated with the variations in crossing-angle seen in the crystals. Our data also revealed a third interface between Revs that explains how the arrangement of Rev subunits can be matched to the A-shaped architecture of the RRE in export-active complexes. This work was published late in 2016. Over the current reporting period, activity on this project, carried out in collaboration with NIAMS Protein Expression Laboratory, has focused on crystallographic studies of Rev constructs alone or complexed with interaction partners. Rev has two domains an alfa-helical hairpin NTD, called the assembly domain, followed by a disordered CTD. We determined a crystal structure for the assembly domain at 2.25 resolution without resort to mutations or chaperones. It reveals a subunit arrangement which suggests how four molecules of Rev can assemble at two interacting sites on the RRE to form a specificity check-point that can be further stabilized by the binding of additional copies of Rev. This study has recently been published.
Showing the most recent 10 out of 14 publications
