HIV-1 reverse transcriptase stability correlates with Gag cleavage efficiency: reverse transcriptase interaction implications for modulating protease activation.

Proteolytic processing of human immunodeficiency virus type 1 particles mediated by viral protease (PR) is essential for acquiring virus infectivity. Activation of PR embedded in Gag-Pol is triggered by Gag-Pol dimerization during virus assembly. We previously reported that amino acid substitutions at the RT tryptophan repeat motif destabilize virus-associated RT and attenuate the ability of efavirenz (EFV, an RT dimerization enhancer) to increase PR-mediated Gag cleavage efficiency. Furthermore, a single amino acid change at RT significantly reduces virus yields due to enhanced Gag cleavage. These data raise the possibility of the RT domain contributing to PR activation by promoting Gag-Pol dimerization. To test this hypothesis, we investigated the putative involvement of a hydrophobic leucine repeat motif (LRM) spanning RT L282 to L310 in RT/RT interactions. We found that LRM amino acid substitutions led to RT instability and that RT is consequently susceptible to degradation by PR. The LRM mutants exhibited reduced Gag cleavage efficiencies while attenuating the EFV enhancement of Gag cleavage. In addition, an RT dimerization-defective mutant, W401A, reduced enhanced Gag cleavage via a leucine zipper (LZ) motif inserted at the deleted Gag-Pol region. Importantly, the presence of RT and integrase domains failed to counteract the LZ enhancement of Gag cleavage. A combination of the Gag cleavage enhancement factors EFV and W402A markedly impaired Gag cleavage, indicating a disruption of W402A Gag-Pol dimerization following EFV binding to W402A Gag-Pol. Our results support the idea that RT modulates PR activation by affecting Gag-Pol/Gag-Pol interaction. IMPORTANCE A stable reverse transcriptase (RT) p66/51 heterodimer is required for HIV-1 genome replication in host cells following virus entry. The activation of viral protease (PR) to mediate virus particle processing helps viruses acquire infectivity following cell release. RT and PR both appear to be major targets for inhibiting HIV-1 replication. We found a strong correlation between impaired p66/51RT stability and deficient PR-mediated Gag cleavage, suggesting that RT/RT interaction is critical for triggering PR activation via the promotion of adequate Gag-Pol dimerization. Accordingly, RT/RT interaction is a potentially advantageous method for anti-HIV/AIDS therapy if it is found to simultaneously block PR and RT enzymatic activity.

Effects of reduced gag cleavage efficiency on HIV-1 Gag-Pol package.

BACKGROUND: HIV-1 pol, which encodes enzymes required for virus replication, is initially translated as a Gag-Pol fusion protein. Gag-Pol is incorporated into virions via interactions with Gag precursor Pr55gag. Protease (PR) embedded in Gag-Pol mediates the proteolytic processing of both Pr55gag and Gag-Pol during or soon after virus particle release from cells. Since efficient Gag-Pol viral incorporation depends on interaction with Pr55gag via its N-terminal Gag domain, the prevention of premature Gag cleavage may alleviate Gag-Pol packaging deficiencies associated with cleavage enhancement from PR. RESULTS: We engineered PR cleavage-blocking Gag mutations with the potential to significantly reduce Gag processing efficiency. Such mutations may mitigate the negative effects of enhanced PR activation on virus assembly and Gag-Pol packaging due to an RT dimerization enhancer or leucine zipper dimerization motif. When co-expressed with Pr55gag, we noted that enhanced PR activation resulted in reduced Gag-Pol cis or trans incorporation into Pr55gag particles, regardless of whether or not Gag cleavage sites within Gag-Pol were blocked. CONCLUSIONS: Our data suggest that the amount of HIV-1 Gag-Pol or Pol viral incorporation is largely dependent on virus particle production, and that cleavage blocking in the Gag-Pol N-terminal Gag domain does not exert significant impacts on Pol packaging.

Amino acid substitutions at the HIV-1 transframe region significantly impair virus infectivity.

A transframe region within HIV-1 Gag-Pol (referred to as p6* or p6pol), directly linked to the protease (PR) N-terminus, plays a pivotal role in modulating PR activation. To identify specific p6* residues involved in PR activation, we created a series of p6* mutants by making substitutions for conserved p6* residues. Our results indicate that some p6* mutants were defective in terms of virus infectivity, despite displaying a wild-type virus particle processing pattern. Mutations at p6* F8 reduced virus infectivity associated with insufficient virus processing, due in part to impaired PR maturation and RT packaging. Our data strongly suggest that conserved Phe (F) residues at position 8 of p6* are involved in the PR maturation process.

Conditional activation of an HIV-1 protease attenuated mutant by a leucine zipper dimerization motif.

Mature HIV-1 protease (PR) functions as a dimer. Changes in HIV-1 PR activation can block virus assembly via premature or enhanced Gag cleavage. HIV-1 PR precursor contains N terminal-linked p6*, a possible modulating factor in PR activation. We found that p6* replacement with a leucine zipper (LZ) dimerization motif (creating a DWzPR construct) or an LZ insertion at the PR C-terminus significantly reduced virus yields due to enhanced Gag cleavage, suggesting that an LZ insertion promotes PR activation by facilitating PR dimer formation. However, introducing T26S (a PR activity-attenuated mutation) into DWzPR strongly impaired Gag cleavage, except when the native C-terminal p6* tetrapeptide remained at the LZ/PR junction. LZ insertion at the PR C-terminus still strongly enhanced PR T26S Gag cleavage. Our data suggest that in addition to p6* mutations, a single amino acid substitution within PR can impair PR activation, likely due to conformational changes triggered by the PR precursor.

HIV-1 Mutant Assembly, Processing and Infectivity Expresses Pol Independent of Gag.

The pol retrovirus gene encodes required enzymes for virus replication and maturation. Unlike HIV-1 Pol (expressed as a Gag-Pol fusion protein), foamy virus (described as an ancient retrovirus) expresses Pol without forming Gag-Pol polyproteins. We placed a "self-cleaving" 2A peptide between HIV-1 Gag and Pol. This construct, designated G2AP, is capable of producing virions with the same density as a wild-type (wt) HIV-1 particle. The 2A peptide allows for Pol to be packaged into virions independently from Gag following co-translationally cleaved from Gag. We found that G2AP exhibited only one-third the virus infectivity of the wt, likely due, at least in part, to defects in Pol packaging. Attenuated protease (PR) activity, or a reduction in Pol expression due to the placement of 2A-mediated Pol in a normal Gag-Pol frameshift context, resulted in significant increases in virus yields and/or titers. This suggests that reduced G2AP virus yields were largely due to increased PR activity associated with overexpressed Pol. Our data suggest that HIV-1 adopts a gag/pol ribosomal frameshifting mechanism to support virus assembly via the efficient modulation of Gag-Pol/Gag expression, as well as to promote viral enzyme packaging. Our results help clarify the molecular basis of HIV-1 gene expression and assembly.

Severe acute respiratory syndrome coronavirus spike protein counteracts BST2-mediated restriction of virus-like particle release.

BST2/tetherin, an interferon-inducible antiviral factor, can block the cellular release of various enveloped viruses. We previously reported that human coronavirus 229E (HCoV-229E) infection can alleviate the BST2 tethering of HIV-1 virions by downregulating cell surface BST2, suggesting that coronaviruses are capable of encoding anti-BST2 factors. Here we report our new finding that severe acute respiratory syndrome coronavirus (SARS-CoV) spike (S) glycoprotein, similar to Vpu, is capable of antagonizing the BST2 tethering of SARS-CoV, HCoV-229E, and HIV-1 virus-like particles via BST2 downregulation. However, unlike Vpu (which downmodulates BST2 by means of proteasomal and lysosomal degradation pathways), BST2 downregulation is apparently mediated by SARS-CoV S through the lysosomal degradation pathway only. We found that SARS-CoV S colocalized with both BST2 and reduced cell surface BST2, suggesting an association between SARS-CoV S and BST2 that targets the lysosomal degradation pathway. According to one recent report, SARS-CoV ORF7a antagonizes BST2 by interfering with BST2 glycosylation1 . Our data provide support for the proposal that SARS-CoV and other enveloped viruses are capable of evolving supplementary anti-BST2 factors in a manner that requires virus replication. Further experiments are required to determine whether the BST2-mediated restriction of authentic SARS-CoV virions is alleviated by the SARS-CoV spike protein.

C-Terminal HIV-1 Transframe p6* Tetrapeptide Blocks Enhanced Gag Cleavage Incurred by Leucine Zipper Replacement of a Deleted p6* Domain.

HIV-1 protease (PR) functions as a homodimer mediating virus maturation following virus budding. Gag-Pol dimerization is believed to trigger embedded PR activation by promoting PR dimer formation. Early PR activation can lead to markedly reduced virus yields due to premature Gag cleavage. The p6* peptide, located between Gag and PR, is believed to ensure virus production by preventing early PR maturation. Studies aimed at finding supporting evidence for this proposal are limited due to a reading frame overlap between p6* and the p6gag budding domain. To determine if p6* affects virus production via the modulation of PR activation, we engineered multiple constructs derived from Dp6*PR (an assembly- and processing-competent construct with Pol fused at the inactivated PR C terminus). The data indicated that a p6* deletion adjacent to active PR significantly impaired virus processing. We also observed that the insertion of a leucine zipper (LZ) dimerization motif in the deleted region eliminated virus production in a PR activity-dependent manner, suggesting that the LZ insertion triggered premature PR activation by facilitating PR dimer formation. As few as four C-terminal p6* residues remaining at the p6*/PR junction were sufficient to restore virus yields, with a Gag processing profile similar to that of the wild type. Our study provides supporting evidence in a virus assembly context that the C-terminal p6* tetrapeptide plays a role in preventing premature PR maturation.IMPORTANCE Supporting evidence for the assumption that p6* retards PR maturation in the context of virus assembly is lacking. We found that replacing p6* with a leucine zipper peptide abolished virus assembly due to the significant enhancement of Gag cleavage. However, as few as four C-terminal p6* residues remaining in the deleted region were sufficient for significant PR release, as well as for counteracting leucine zipper-incurred premature Gag cleavage. Our data provide evidence that (i) p6* ensures virus assembly by preventing early PR activation and (ii) four C-terminal p6* residues are critical for modulating PR activation. Current PR inhibitor development efforts are aimed largely at mature PR, but there is a tendency for HIV-1 variants that are resistant to multiple protease inhibitors to emerge. Our data support the idea of modulating PR activation by targeting PR precursors as an alternative approach to controlling HIV-1/AIDS.

p6gag domain confers cis HIV-1 Gag-Pol assembly and release capability.

During virus assembly, HIV-1 Gag-Pol is packaged into virions via interaction with Pr55gag. Studies suggest that Gag-Pol by itself is incapable of virus particle assembly or cell release, perhaps due to the lack of a budding domain in the form of p6gag, which is truncated within Gag-Pol because of a ribosomal frameshift during Gag translation. Additionally (or alternatively), large molecular size may not support Gag-Pol assembly into virus-like particles (VLPs) or release from cells. To test these hypotheses, we constructed Gag-Pol expression vectors retaining and lacking p6gag, and then reduced Gag-Pol molecular size by removing various lengths of the Pol sequence. Results indicate that Gag-Pol constructs retaining p6gag were capable of forming VLPs with a WT HIV-1 particle density. Gag-Pol molecular size reduction via partial removal of the Pol sequence mitigated the Gag-Pol assembly defect to a moderate degree. Our results suggest that the Gag-Pol assembly and budding defects are largely due to a lack of p6gag, but also in part due to size limitation.

Gag-Pol Transframe Domain p6* Is Essential for HIV-1 Protease-Mediated Virus Maturation.

HIV-1 protease (PR) is encoded by pol, which is initially translated as a Pr160gag-pol polyprotein by a ribosomal frameshift event. Within Gag-Pol, truncated p6gag is replaced by a transframe domain (referred to as p6* or p6pol) located directly upstream of PR. p6* has been proposed as playing a role in modulating PR activation. Overlapping reading frames between p6* and p6gag present a challenge to researchers using genetic approaches to studying p6* biological functions. To determine the role of p6* in PR activation without affecting the gag reading frame, we constructed a series of Gag/Gag-Pol expression vectors by duplicating PR with or without p6* between PR pairs, and observed that PR duplication eliminated virus production due to significant Gag cleavage enhancement. This effect was mitigated when p6* was placed between the two PRs. Further, Gag cleavage enhancement was markedly reduced when either one of the two PRs was mutationally inactivated. Additional reduction in Gag cleavage efficiency was noted following the removal of p6* from between the two PRs. The insertion of a NC domain (wild-type or mutant) directly upstream of PR or p6*PR did not significantly improve Gag processing efficiency. With the exception of those containing p6* directly upstream of an active PR, all constructs were either noninfectious or weakly infectious. Our results suggest that (a) p6* is essential for triggering PR activation, (b) p6* has a role in preventing premature virus processing, and

HIV-1 matrix domain removal ameliorates virus assembly and processing defects incurred by positive nucleocapsid charge elimination.

Human immunodeficiency virus type 1 nucleocapsid (NC) basic residues presumably contribute to virus assembly via RNA, which serves as a scaffold for Gag-Gag interaction during particle assembly. To determine whether NC basic residues play a role in Gag cleavage (thereby impacting virus assembly), Gag processing efficiency and virus particle production were analyzed for an HIV-1 mutant NC15A, with alanine serving as a substitute for all NC basic residues. Results indicate that NC15A significantly impaired virus maturation in addition to significantly affecting Gag membrane binding and assembly. Interestingly, removal of the matrix (MA) central globular domain ameliorated the NC15A assembly and processing defects, likely through enhancement of Gag multimerization and membrane binding capacities.

SARS-CoV envelope protein palmitoylation or nucleocapid association is not required for promoting virus-like particle production.

BACKGROUND: Coronavirus membrane (M) proteins are capable of interacting with nucleocapsid (N) and envelope (E) proteins. Severe acute respiratory syndrome coronavirus (SARS-CoV) M co-expression with either N or E is sufficient for producing virus-like particles (VLPs), although at a lower level compared to M, N and E co-expression. Whether E can release from cells or E/N interaction exists so as to contribute to enhanced VLP production is unknown. It also remains to be determined whether E palmitoylation or disulfide bond formation plays a role in SARS-CoV virus assembly. RESULTS: SARS-CoV N is released from cells through an association with E protein-containing vesicles. Further analysis suggests that domains involved in E/N interaction are largely located in both carboxyl-terminal regions. Changing all three E cysteine residues to alanines did not exert negative effects on E release, E association with N, or E enhancement of VLP production, suggesting that E palmitoylation modification or disulfide bond formation is not required for SARS-CoV virus assembly. We found that removal of the last E carboxyl-terminal residue markedly affected E release, N association, and VLP incorporation, but did not significantly compromise the contribution of E to efficient VLP production. CONCLUSIONS: The independence of the SARS-CoV E enhancement effect on VLP production from its viral packaging capacity suggests a distinct SARS-CoV E role in virus assembly.

BST2/CD317 counteracts human coronavirus 229E productive infection by tethering virions at the cell surface.

Bone marrow stromal antigen 2 (BST2), an interferon-inducible antiviral factor, has been shown to block the release of various enveloped viruses from cells. It has also been identified as an innate immune system component. Most enveloped viruses subject to BST2 restriction bud at the plasma membrane. Here we report our findings that (a) the production of human coronavirus 229E (HCoV-229E) progeny viruses, whose budding occurs at the ER-Golgi intermediate compartment (ERGIC), markedly decreases in the presence of BST2; and (b) BST2 knockdown expression results in enhanced HCoV-229E virion production. Electron microscopy analyses indicate that HCoV-229E virions are tethered to cell surfaces or intracellular membranes by BST2. Our results suggest that BST2 exerts a broad blocking effect against enveloped virus release, regardless of whether budding occurs at the plasma membrane or intracellular compartments.

Identifying SARS-CoV membrane protein amino acid residues linked to virus-like particle assembly.

Severe acute respiratory syndrome coronavirus (SARS-CoV) membrane (M) proteins are capable of self-assembly and release in the form of membrane-enveloped vesicles, and of forming virus-like particles (VLPs) when coexpressed with SARS-CoV nucleocapsid (N) protein. According to previous deletion analyses, M self-assembly involves multiple M sequence regions. To identify important M amino acid residues for VLP assembly, we coexpressed N with multiple M mutants containing substitution mutations at the amino-terminal ectodomain, carboxyl-terminal endodomain, or transmembrane segments. Our results indicate that a dileucine motif in the endodomain tail (218LL219) is required for efficient N packaging into VLPs. Results from cross-linking VLP analyses suggest that the cysteine residues 63, 85 and 158 are not in close proximity to the M dimer interface. We noted a significant reduction in M secretion due to serine replacement for C158, but not for C63 or C85. Further analysis suggests that C158 is involved in M-N interaction. In addition to mutations of the highly conserved 107-SWWSFNPE-114 motif, substitutions at codons W19, W57, P58, W91, Y94 or F95 all resulted in significantly reduced VLP yields, largely due to defective M secretion. VLP production was not significantly affected by a tryptophan replacement of Y94 or F95 or a phenylalanine replacement of W19, W57 or W91. Combined, these results indicate the involvement of specific M amino acids during SARS-CoV virus assembly, and suggest that aromatic residue retention at specific positions is critical for M function in terms of directing virus assembly.

Placement of leucine zipper motifs at the carboxyl terminus of HIV-1 protease significantly reduces virion production.

Natural HIV-1 protease (PR) is homodimeric. Some researchers believe that interactions between HIV-1 Gag-Pol molecules trigger the activation of embedded PR (which mediates Gag and Gag-Pol cleavage), and that Gag-Pol assembly domains outside of PR may contribute to PR activation by influencing PR dimer interaction in a Gag-Pol context. To determine if the enhancement of PR dimer interaction facilitates PR activation, we placed single or tandem repeat leucine zippers (LZ) at the PR C-terminus, and looked for a correlation between enhanced Gag processing efficiency and increased Gag-PR-LZ multimerization capacity. We found significant reductions in virus-like particles (VLPs) produced by HIV-1 mutants, with LZ fused to the end of PR as a result of enhanced Gag cleavage efficiency. Since VLP production can be restored to wt levels following PR activity inhibition, this assembly defect is considered PR activity-dependent. We also found a correlation between the LZ enhancement effect on Gag cleavage and enhanced Gag-PR multimerization. The results suggest that PR dimer interactions facilitated by forced Gag-PR multimerization lead to premature Gag cleavage, likely a result of premature PR activation. Our conclusion is that placement of a heterologous dimerization domain downstream of PR enhances PR-mediated Gag cleavage efficiency, implying that structural conformation, rather than the primary sequence outside of PR, is a major determinant of HIV-1 PR activation.

Mutations in the HIV-1 reverse transcriptase tryptophan repeat motif affect virion maturation and Gag-Pol packaging.

Our goal was to determine the contribution of HIV-1 reverse transcriptase tryptophan repeat motif residues to virion maturation. With the exception of W402A, we found none of the single substitution mutations exerted major impacts on virus assembly or processing. However, all mutants except for W410A exhibited significant decreases in virus-associated RT, presumably a result of unstable RT mutant degradation. Mutations W398A, W401A and W406A decreased the enhancement effect of efavirenz on PR-mediated Gag processing efficiency, which is in agreement with their destabilizing RT effects. Furthermore, combined double or triple W398, W401 and W406 mutations significantly affected virus processing and Gag-Pol packaging. Further analyses suggest that inefficient PR-mediated Gag cleavage partly accounts for the virion processing defect. Our results support the idea that in addition to playing a role in RT heterodimer stabilization, the RT Trp repeat motif in the Gag-Pol context is also involved in PR activation via Gag-Pol/Gag-Pol interaction.

Efficient in silico assay of inhibitors of hepatitis C Virus RNA-dependent RNA polymerase by structure-based virtual screening and in vitro evaluation.

To identify a new protective or therapeutic intervention for hepatitis C virus (HCV) infection, we performed efficient structure-based virtual screening to identify novel inhibitory agents for HCV. To this end, we selected NS5B, an RNA-dependent RNA polymerase (RdRp), as the target for the treatment of HCV infection. To decipher the dockable nature of various RdRp X-ray crystals, we docked the crystal ligand (inhibitor) to the crystal receptor (enzyme). The accuracy of regeneration of the crystal pose indicates the amenability of the RdRp binding pocket for structure-based virtual screening. We also utilized a consensus scoring scheme to reduce false positives, thereby ensuring efficient virtual screening. In this study, each molecule that ranked in the top 1% among all screening molecules gained 1 consensus point in a scoring function. Thus, after virtual screening of 57,177 chemicals from the Maybridge Screening collection, 14 molecules gained 8 points across 11 scoring functions. One of them, an isoxazole, showed significant dose-dependent inhibition of HCV RdRp activity and replication. In this study, we have developed a structure-based virtual screening method using HCV RdRp for efficient identification of novel inhibitors.

A cell-based reporter assay for inhibitor screening of hepatitis C virus RNA-dependent RNA polymerase.

The hepatitis C virus (HCV) NS5B, a RNA-dependent RNA polymerase (RdRp), is an attractive target for anti-HCV agents. The major disadvantages of the commonly used polymerase inhibitor screening involving the assessment of in vitro RNA synthesis are that it is incapable of demonstrating the cellular permeability and the cytotoxicity of compounds. To overcome these limitations, we created the BHK-NS5B-FRLuc reporter cell line that carries stably transfected NS5B and a bicistronic reporter gene, (+)FLuc-(-)UTR-RLuc, which can be used to simultaneously measure cellular toxicity and intracellular RdRp activity. The (+)FLuc-(-)UTR-RLuc construct comprises the firefly luciferase (FLuc) gene and the Renilla luciferase (RLuc) gene in reverse orientation flanked by both negative strands of the HCV 5'- and 3'-untranslated regions (UTRs), in which FLuc and RLuc reporter proteins are regulated by host polymerase and functional NS5B polymerase, respectively. The reporter system was validated with specific agents against NS5B polymerization. Additionally, this assay was placed in 96-well plates and had a Z'-factor value of approximately 0.75, which is amenable for facilitating high-throughput screening operations. Notably, in combination with the structured-based virtual screening, an imidazole derivative compound was evaluated as a candidate HCV RdRp inhibitor.

Self-assembly of severe acute respiratory syndrome coronavirus membrane protein.

Coronavirus membrane (M) protein can form virus-like particles (VLPs) when coexpressed with nucleocapsid (N) or envelope (E) proteins, suggesting a pivotal role for M in virion assembly. Here we demonstrate the self-assembly and release of severe acute respiratory syndrome coronavirus (SARS-CoV) M protein in medium in the form of membrane-enveloped vesicles with densities lower than those of VLPs formed by M plus N. Although efficient N-N interactions require the presence of RNA, we found that M-M interactions were RNA-independent. SARS-CoV M was observed in both the Golgi area and plasma membranes of a variety of cells. Blocking M glycosylation does not appear to significantly affect M plasma membrane labeling intensity, M-containing vesicle release, or VLP formation. Results from a genetic analysis indicate involvement of the third transmembrane domain of M in plasma membrane-targeting signal. Fusion proteins containing M amino-terminal 50 residues encompassing the first transmembrane domain were found to be sufficient for membrane binding, multimerization, and Golgi retention. Surprisingly, we found that fusion proteins lacking all three transmembrane domains were still capable of membrane binding, Golgi retention, and interacting with M. The data suggest that multiple SARS-CoV M regions are involved in M self-assembly and subcellular localization.

A single amino acid substitution in HIV-1 reverse transcriptase significantly reduces virion release.

HIV-1 protease (PR) mediates the proteolytic processing of virus particles during or after virus budding. PR activation is thought to be triggered by appropriate Gag-Pol/Gag-Pol interaction; factors affecting this interaction either enhance or reduce PR-mediated cleavage efficiency, resulting in markedly reduced virion production or the release of inadequately processed virions. We previously showed that a Gag-Pol deletion mutation involving the reverse transcriptase tryptophan (Trp) repeat motif markedly impairs PR-mediated virus maturation and that an alanine substitution at W401 (W401A) or at both W401 and W402 (W401A/W402A) partially or almost completely negates the enhancement effect of efavirenz (a nonnucleoside reverse transcriptase inhibitor) on PR-mediated virus processing efficiency. These data suggest that the Trp repeat motif may contribute to the PR activation process. Here we demonstrate that due to enhanced Gag cleavage efficiency, W402 alanine or leucine substitution significantly reduces virus production. However, W402 replacement with phenylalanine does not significantly affect virus particle assembly or processing, but it does markedly impair viral infectivity in a single-cycle infection assay. Our results demonstrate that a single amino acid substitution at HIV-1 RT can radically affect virus assembly by enhancing Gag cleavage efficiency, suggesting that in addition to contributing to RT biological function during the early stages of virus replication, the HIV-1 RT tryptophan repeat motif in a Gag-Pol context may play an important role in suppressing the premature activation of PR during late-stage virus replication.