Rebecca C. Craven

The small RING finger protein Z drives arenavirus budding: Implications for antiviral strategies

By using a reverse genetics system that is based on the prototypic arenavirus lymphocytic choriomeningitis virus (LCMV), we have identified the arenavirus small RING finger Z protein as the main driving force of virus budding. Both LCMV and Lassa fever virus (LFV) Z proteins exhibited self-budding activity, and both substituted efficiently for the late domain that is present in the Gag protein of Rous sarcoma virus. LCMV and LFV Z proteins contain proline-rich motifs that are characteristic of late domains. Mutations in the PPPY motif of LCMV Z severely impaired the formation of virus-like particles. LFV Z contains two different proline-rich motifs, PPPY and PTAP, which are separated by eight amino acids. Mutational analysis revealed that both motifs are required for efficient LFV Z-mediated budding. Both LCMV and LFV Z proteins recruited to the plasma membrane Tsg101, which is a component of the class E vacuolar protein sorting machinery that has been implicated in budding of HIV and Ebola virus. Targeting of Tsg101 by RNA interference caused a strong reduction in Z-mediated budding. These results indicate that Z is the arenavirus functional counterpart of the matrix proteins found in other negative strand enveloped RNA viruses. Moreover, members of the vacuolar protein sorting pathway appear to play an important role in arena-virus budding. These findings open possibilities for antiviral strategies to combat LFV and other hemorrhagic fever arenaviruses.

Visualization of a missing link in retrovirus capsid assembly

For a retrovirus such as HIV to be infectious, a properly formed capsid is needed; however, unusually among viruses, retrovirus capsids are highly variable in structure. According to the fullerene conjecture, they are composed of hexamers and pentamers of capsid protein (CA), with the shape of a capsid varying according to how the twelve pentamers are distributed and its size depending on the number of hexamers. Hexamers have been studied in planar and tubular arrays, but the predicted pentamers have not been observed. Here we report cryo-electron microscopic analyses of two in-vitro-assembled capsids of Rous sarcoma virus. Both are icosahedrally symmetric: one is composed of 12 pentamers, and the other of 12 pentamers and 20 hexamers. Fitting of atomic models of the two CA domains into the reconstructions shows three distinct inter-subunit interactions. These observations substantiate the fullerene conjecture, show how pentamers are accommodated at vertices, support the inference that nucleation is a crucial morphologic determinant, and imply that electrostatic interactions govern the differential assembly of pentamers and hexamers.

Retrovirus-Specific Packaging of Aminoacyl-tRNA Synthetases with Cognate Primer tRNAs

ABSTRACT The tRNAs used to prime reverse transcription in human immunodeficiency virus type 1 (HIV-1), Rous sarcoma virus (RSV), and Moloney murine leukemia virus (Mo-MuLV) are \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(tRNA_{3}^{Lys}\) \end{document} , tRNATrp, and tRNAPro, respectively. Using antibodies to the three cognate human aminoacyl-tRNA synthetases, we found that only lysyl-tRNA synthetase (LysRS) is present in HIV-1, only tryptophanyl-tRNA synthetase (TrpRS) is present in RSV, and neither these two synthetases nor prolyl-tRNA synthetase (ProRS) is present in Mo-MuLV. LysRS and TrpRS are present in HIV-1 and RSV at approximately 25 and 12 molecules/virion, respectively. These results support the hypothesis that, in HIV-1 and RSV, the cognate aminoacyl-tRNA synthetase may be used as the signal for targeting the selective packaging of primer tRNAs into retroviruses. The absence of ProRS in Mo-MuLV is consistent with reports that selective packaging of tRNAPro in this virus is less important for achieving optimum annealing of the primer to Mo-MuLV genomic RNA.

Dynamic interactions of the Gag polyprotein.

It has been known for 20 years that all retroviruses possess a similar genetic organization and that a single gene product, the Gag protein, is responsible for directing particle assembly (DICKSON et al. 1984; COFFIN 1984). However it is only very recently that an understanding of how the Gag protein functions has begun to take shape. This chapter reviews recent efforts to understand the assembly and budding functions of the Gag protein with a primary focus on Rous sarcoma virus (RSV, a member of the avian sarcoma-leukosis virus group of oncoviruses) and the human immunodeficiency virus type 1 (HIV-1). These studies have shown that in spite of a striking divergence in amino acid sequences, the two viruses utilize fundamentally similar mechanisms to direct assembly and budding.

Second-Site Suppressors of Rous Sarcoma Virus CA Mutations: Evidence for Interdomain Interactions

ABSTRACT The capsid (CA) protein, the major structural component of retroviruses, forms a shell that encases the ribonucleoprotein complex in the virion core. The most conserved region of CA, ∼20 amino acids of the major homology region (MHR), lies within the carboxy-terminal domain of the protein. Structural and sequence similarities among CA proteins of retroviruses and the CA-like proteins of hepatitis B virus and various retrotransposons suggest that the MHR is involved in an aspect of replication common to these reverse-transcribing elements. Conservative substitutions in this region of the Rous sarcoma virus protein were lethal due to a severe deficiency in reverse transcription, in spite of the presence of an intact genome and active reverse transcriptase in the particles. This finding suggests that the mutations interfered with normal interactions among these constituents. A total of four genetic suppressors of three lethal MHR mutations have now been identified. All four map to the sequence encoding the CA-spacer peptide (SP) region of Gag. The F167Y mutation in the MHR was fully suppressed by a single amino acid change in the alpha helix immediately downstream of the MHR, a region that forms the major dimer interface in human immunodeficiency virus CA. This finding suggests that the F167Y mutation indirectly interfered with dimerization. The F167Y defect could also be repaired by a second, independent suppressor in the C-terminal SP that was removed from CA during maturation. This single residue change, which increased the rate of SP cleavage, apparently corrected the F167Y defect by modifying the maturation pathway. More surprising was the isolation of suppressors of the R170Q and L171V MHR mutations, which mapped to the N-terminal domain of the CA protein. This finding suggests that the two domains, which in the monomeric protein are separated by a flexible linker, must communicate with each other at some unidentified point in the viral replication cycle.

Critical Role of Conserved Hydrophobic Residues within the Major Homology Region in Mature Retroviral Capsid Assembly

ABSTRACT During retroviral maturation, the CA protein undergoes dramatic structural changes and establishes unique intermolecular interfaces in the mature capsid shell that are different from those that existed in the immature precursor. The most conserved region of CA, the major homology region (MHR), has been implicated in both immature and mature assembly, although the precise contribution of the MHR residues to each event has been largely undefined. To test the roles of specific MHR residues in mature capsid assembly, an in vitro system was developed that allowed for the first-time formation of Rous sarcoma virus CA into structures resembling authentic capsids. The ability of CA to assemble organized structures was destroyed by substitutions of two conserved hydrophobic MHR residues and restored by second-site suppressors, demonstrating that these MHR residues are required for the proper assembly of mature capsids in addition to any role that these amino acids may play in immature particle assembly. The defect caused by the MHR mutations was identified as an early step in the capsid assembly process. The results provide strong evidence for a model in which the hydrophobic residues of the MHR control a conformational reorganization of CA that is needed to initiate capsid assembly and suggest that the formation of an interdomain interaction occurs early during maturation.

Viral DNA Synthesis Defects in Assembly-Competent Rous Sarcoma Virus CA Mutants

ABSTRACT The major structural protein of the retroviral core (CA) contains a conserved sequence motif shared with the CA-like proteins of distantly related transposable elements. The function of this major region of homology (MHR) has not been defined, in part due to the baffling array of phenotypes in mutants of several viruses and the yeast TY3. This report describes new mutations in the CA protein of Rous sarcoma virus (RSV) that were designed to test whether these different phenotypes might indicate distinct functional subdomains in the MHR. A comparison of 25 substitutions at 10 positions in the RSV conserved motif argues against this possibility. Most of the replacements destroyed virus infectivity, although either of two lethal phenotypes was obtained depending on the residue introduced. At most of the positions, one or more replacements (generally the more conservative substitutions) caused a severe replication defect without having any obvious effects on virus assembly, budding, Gag-Pol and genome incorporation, or protein processing. The mutant particles exhibited a defect in endogenous viral DNA synthesis and showed increased sensitivity of the core proteins to detergent, indicating that the mutations interfere with the formation and/or activity of the virion core. The distribution of these mutations across the MHR, with no evidence of clustering, suggests that the entire region is important for a critical postbudding function. In contrast, a second class of lethal substitutions (those that destroyed virus assembly and release) consists of alterations that are expected to cause severe effects on protein structure by disruption either of the hydrophobic core of the CA carboxyl-terminal domain or of the hydrogen bond network that stabilizes the domain. We suggest that this duality of phenotypes is consistent with a role for the MHR in the maturation process that links the two parts of the life cycle.

RNA Dimerization Defect in a Rous Sarcoma Virus Matrix Mutant

ABSTRACT The retrovirus matrix (MA) sequence of the Gag polyprotein has been shown to contain functions required for membrane targeting and binding during particle assembly and budding. Additional functions for MA have been proposed based on the existence of MA mutants in Rous sarcoma virus (RSV), murine leukemia virus, human immunodeficiency virus type 1, and human T-cell leukemia virus type 1 that lack infectivity even though they release particles of normal composition. Here we describe an RSV MA mutant with a surprising and previously unreported phenotype. In the mutant known as Myr1E, the small membrane-binding domain of the Src oncoprotein has been added as an N-terminal extension of Gag. While Myr1E is not infectious, full infectivity can be reestablished by a single amino acid substitution in the Src sequence (G2E), which eliminates the addition of myristic acid and the membrane-binding capacity of this foreign sequence. The presence of myristic acid at the N terminus of the Myr1E Gag protein does not explain its replication defect, because other myristylated derivatives of RSV Gag are fully infectious (e.g., Myr2 [C. R. Erdie and J. W. Wills, J. Virol. 64:5204–5208, 1990]). Biochemical analyses of Myr1E particles reveal that they contain wild-type levels of the Gag cleavage products, Env glycoproteins, and reverse transcriptase activity when measured on an exogenous template. Genomic RNA incorporation appears to be mildly reduced compared to the wild-type level. Unexpectedly, RNA isolated from Myr1E particles is monomeric when analyzed on nondenaturing Northern blots. Importantly, the insertional mutation does not lie within previously identified dimer linkage sites. In spite of the dimerization defect, the genomic RNA from Myr1E particles serves efficiently as a template for reverse transcription as measured by an endogenous reverse transcriptase assay. In marked contrast, after infection of avian cells, the products of reverse transcription are nearly undetectable. These findings might be explained either by the loss of a normal function of MA needed in the formation or stabilization of RNA dimers or by the interference in such events by the mutant MA molecules. It is possible that Myr1E viruses package a single copy of viral RNA.

A two-pronged structural analysis of retroviral maturation indicates that core formation proceeds by a disassembly-reassembly pathway, rather than a displacive transition

ABSTRACT Retrovirus maturation involves sequential cleavages of the Gag polyprotein, initially arrayed in a spherical shell, leading to formation of capsids with polyhedral or conical morphology. Evidence suggests that capsids assemble de novo inside maturing virions from dissociated capsid (CA) protein, but the possibility persists of a displacive pathway in which the CA shell remains assembled but is remodeled. Inhibition of the final cleavage between CA and spacer peptide SP1/SP blocks the production of mature capsids. We investigated whether retention of SP might render CA assembly incompetent by testing the ability of Rous sarcoma virus (RSV) CA-SP to assemble in vitro into icosahedral capsids. Capsids were indeed assembled and were indistinguishable from those formed by CA alone, indicating that SP was disordered. We also used cryo-electron tomography to characterize HIV-1 particles produced in the presence of maturation inhibitor PF-46396 or with the cleavage-blocking CA5 mutation. Inhibitor-treated virions have a shell that resembles the CA layer of the immature Gag shell but is less complete. Some CA protein is generated but usually not enough for a mature core to assemble. We propose that inhibitors like PF-46396 bind to the Gag lattice where they deny the protease access to the CA-SP1 cleavage site and prevent the release of CA. CA5 particles, which exhibit no cleavage at the CA-SP1 site, have spheroidal shells with relatively thin walls. It appears that this lattice progresses displacively toward a mature-like state but produces neither conical cores nor infectious virions. These observations support the disassembly-reassembly pathway for core formation.

Lysines Close to the Rous Sarcoma Virus Late Domain Critical for Budding

ABSTRACT The release of retroviruses from the plasma membrane requires host factors that are believed to be recruited to the site of budding by the late (L) domain of the virus-encoded Gag protein. The L domain of Rous sarcoma virus (RSV) has been shown to interact with a ubiquitin (Ub) ligase, and budding of this virus is dependent on Ub. RSV is similar to other retroviruses in that it contains ∼100 molecules of Ub, but it is unique in that none of these molecules has been found to be conjugated to Gag. If transient ubiquitination of RSV Gag is required for budding, then replacement of the target lysine(s) with arginine should prevent the addition of Ub and reduce budding. Based on known sites of ubiquitination in other viruses, the important lysines would likely reside near the L domain. In RSV, there are five lysines located just upstream of the L domain in a region of the matrix (MA) protein that is dispensable for membrane binding, and replacement of these with arginine (mutant 1-5KR) reduced budding 80 to 90%. The block to budding was found to be on the plasma membrane; however, the few virions that were released had normal size, morphology, and infectivity. Budding was restored when any one of the residues was changed back to lysine or when lysines were inserted in novel positions, either within this region of MA or within the downstream p10 sequence. Moreover, the 1-5KR mutant could be rescued into particles by coexpression of budding-competent Gag molecules. These data argue that the phenotype of mutant 1-5KR is not due to a conformational defect. Consistent with the idea that efficient budding requires a specific role for lysines, human T-cell leukemia virus type 1, which does not bud well compared to RSV and lacks lysines close to its L domain, was found to be released at a higher level upon introduction of lysines near its L domain. This report strongly supports the hypothesis that ubiquitination of the RSV Gag protein (and perhaps those of other retroviruses) is needed for efficient budding.