In Vitro Assembly Properties of Human Immunodeficiency Virus Type 1 Gag Protein Lacking the p6 Domain

Human immunodeficiency virus type 1 (HIV-1) normally assembles into particles of 100 to 120 nm in diameter by budding through the plasma membrane of the cell. The Gag polyprotein is the only viral protein that is required for the formation of these particles. We have used an in vitro assembly system to examine the assembly properties of purified, recombinant HIV-1 Gag protein and of Gag missing the C-terminal p6 domain (Gag Δp6). This system was used previously to show that the CA-NC fragment of HIV-1 Gag assembled into cylindrical particles. We now report that both HIV-1 Gag and Gag Δp6 assemble into small, 25- to 30-nm-diameter spherical particles in vitro. The multimerization of Gag Δp6 into units larger than dimers and the formation of spherical particles required nucleic acid. Removal of the nucleic acid with NaCl or nucleases resulted in the disruption of the multimerized complexes. We conclude from these results that (i) N-terminal extension of HIV-1 CA-NC to include the MA domain results in the formation of spherical, rather than cylindrical, particles; (ii) nucleic acid is required for the assembly and maintenance of HIV-1 Gag Δp6 virus-like particles in vitro and possibly in vivo; (iii) a wide variety of RNAs or even short DNA oligonucleotides will support assembly; (iv) protein-protein interactions within the particle must be relatively weak; and (v) recombinant HIV-1 Gag Δp6 and nucleic acid are not sufficient for the formation of normal-sized particles.     

Analysis of the Assembly Function of the Human Immunodeficiency Virus Type 1 Gag Protein Nucleocapsid Domain

Previous studies have shown that in addition to its function in specific RNA encapsidation, the human immunodeficiency virus type 1 (HIV-1) nucleocapsid (NC) is required for efficient virus particle assembly. However, the mechanism by which NC facilitates the assembly process is not clearly established. Formally, NC could act by constraining the Pr55^gag polyprotein into an assembly-competent conformation or by masking residues which block the assembly process. Alternatively, the capacity of NC to bind RNA or make interprotein contacts might affect particle assembly. To examine its role in the assembly process, we replaced the NC domain in Pr55^gag with polypeptide domains of known function, and the chimeric proteins were analyzed for their abilities to direct the release of virus-like particles. Our results indicate that NC does not mask inhibitory domains and does not act passively, by simply providing a stable folded monomeric structure. However, replacement of NC by polypeptides which form interprotein contacts permitted efficient virus particle assembly and release, even when RNA was not detected in the particles. These results suggest that formation of interprotein contacts by NC is essential to the normal HIV-1 assembly process.Human immunodeficiency virus type 1 (HIV-1) encodes three major genes, gag, pol, and env, which are commonly found in all mammalian retroviruses. It also encodes accessory genes whose protein products are important for regulation of its life cycle (6, 30, 35). However, of all the genes encoded by HIV-1, only the protein product of the gag gene has been found to be necessary and sufficient for the assembly of virus-like particles (11, 13, 17, 22, 32, 33). The HIV-1 Gag protein initially is expressed as a 55-kDa polyprotein precursor (Pr55^gag), but during or shortly after particle release, Pr55^gag ordinarily is cleaved by the viral protease (PR). The products of the protease action are the four major viral proteins matrix (MA), capsid (CA), nucleocapsid (NC), and p6, and the two spacer polypeptides p2 and p1, which represent sequences between CA and NC and between NC and p6, respectively (15, 19, 23, 30).The HIV-1 nucleocapsid proteins have two Cys-X₂-Cys-X₄-His-X₄-Cys (Cys-His) motifs, reminiscent of the zinc finger motifs found in many DNA binding proteins, and NC has been shown to facilitate the specific encapsidation of HIV-1 genomic RNAs. In addition to its encapsidation function, NC influences virus particle assembly (7, 10, 17, 21, 40). In particular, Gag proteins lacking the NC domain fail to assemble virus particles efficiently. Nevertheless, some chimeric Gag proteins which carry foreign sequences in place of NC have been shown to assemble and release virus particles at wild-type (wt) levels (2, 37, 40). Thus, it appears that in some circumstances, the role that NC plays in virus particle assembly can be replaced. To date, it is not clear how NC affects particle assembly, although several possibilities might be envisioned. One possibility is that deletion of NC unmasks inhibitory sequences in p2 or the C terminus of CA. Alternatively, NC may simply provide a stable monomeric folded structure which locks CA or other Gag domains into an assembly-competent conformation. Another possibility is that NC facilitates assembly by forming essential protein-protein contacts between neighbor Pr^gag molecules, as suggested in cross-linking studies (21). Finally, the assembly role of NC may stem from its RNA binding capabilities, a hypothesis supported by studies of Campbell and Vogt (5), which have shown that RNA facilitates the in vitro assembly of retroviral Gag proteins into higher-order structures.To distinguish among possible mechanisms by which NC facilitates HIV-1 assembly, we replaced NC with polypeptides having known structural characteristics and examined particle assembly directed by these chimeric proteins. Using this approach, we have found that NC does not play a passive role in HIV-1 assembly as either a mask to assembly inhibitor domains or a nonspecific, stably folded structure. Rather, sequences known to form strong interprotein contacts were observed to enhance assembly, suggesting a similar role for the NC domain itself. With several assembly-competent chimeric proteins, we detected no particle-associated RNAs. These results suggest that while RNA may be essential to virus assembly in the context of the wt Pr55^gag protein, it is dispensable for formation of virus-like particles from chimeric proteins.     

Ubiquitin Is Covalently Attached to the p6Gag Proteins of Human Immunodeficiency Virus Type 1 and Simian Immunodeficiency Virus and to the p12Gag Protein of Moloney Murine Leukemia Virus

Host proteins are incorporated into retroviral virions during assembly and budding. We have examined three retroviruses, human immunodeficiency virus type 1 (HIV-1), simian immunodeficiency virus (SIV), and Moloney murine leukemia virus (Mo-MuLV), for the presence of ubiquitin inside each of these virions. After a protease treatment to remove exterior viral as well as contaminating cellular proteins, the proteins remaining inside the virion were analyzed. The results presented here show that all three virions incorporate ubiquitin molecules at approximately 10% of the level of Gag found in virions. In addition to free ubiquitin, covalent ubiquitin-Gag complexes were detected, isolated, and characterized from all three viruses. Our immunoblot and protein sequencing results on treated virions showed that approximately 2% of either HIV-1 or SIV p6^Gag was covalently attached to a single ubiquitin molecule inside the respective virions and that approximately 2 to 5% of the p12^Gag in Mo-MuLV virions was monoubiquitinated. These results show that ubiquitination of Gag is conserved among these retroviruses and occurs in the p6^Gag portion of the Gag polyprotein, a region that is likely to be involved in assembly and budding.     

The Late Domain of Human Immunodeficiency Virus Type 1 p6 Promotes Virus Release in a Cell Type-Dependent Manner

The p6 domain of human immunodeficiency virus type 1 (HIV-1) is located at the C terminus of the Gag precursor protein Pr55(Gag). Previous studies indicated that p6 plays a critical role in HIV-1 particle budding from virus-expressing HeLa cells. In this study, we performed a detailed mutational analysis of the N terminus of p6 to map the sequences required for efficient virus release. We observed that the highly conserved P-T/S-A-P motif located near the N terminus of p6 is remarkably sensitive to change; even conservative mutations in this sequence imposed profound virus release defects in HeLa cells. In contrast, single and double amino acid substitutions outside the P-T/S-A-P motif had no significant effect on particle release. The introduction of stop codons one or two residues beyond the P-T/S-A-P motif markedly impaired virion release, whereas truncation four residues beyond P-T/S-A-P had no effect on particle production in HeLa cells. By examining the effects of p6 mutation in biological and biochemical analyses and by electron microscopy, we defined the role of p6 in particle release and virus replication in a panel of T-cell and adherent cell lines and in primary lymphocytes and monocyte-derived macrophages. We demonstrated that the effects of p6 mutation on virus replication are markedly cell type dependent. Intriguingly, even in T-cell lines and primary lymphocytes in which p6 mutations block virus replication, these changes had little or no effect on particle release. However, p6-mutant particles produced in T-cell lines and primary lymphocytes exhibited a defect in virion-virion detachment, resulting in the production of tethered chains of virions. Virus release in monocyte-derived macrophages was markedly inhibited by p6 mutation. To examine further the cell type-specific virus release defect in HeLa versus T cells, transient heterokaryons were produced between HeLa cells and the Jurkat T-cell line. These heterokaryons display a T-cell-like phenotype with respect to the requirement for p6 in particle release. The results described here define the role of p6 in virus replication in a wide range of cell types and reveal a strong cell type-dependent requirement for p6 in virus particle budding.     

Particle Size Determinants in the Human Immunodeficiency Virus Type 1 Gag Protein

The retroviral Gag protein plays the central role in the assembly process and can form membrane-enclosed, virus-like particles in the absence of any other viral products. These particles are similar to authentic virions in density and size. Three small domains of the human immunodeficiency virus type 1 (HIV-1) Gag protein have been previously identified as being important for budding. Regions that lie outside these domains can be deleted without any effect on particle release or density. However, the regions of Gag that control the size of HIV-1 particles are less well understood. In the case of Rous sarcoma virus (RSV), the size determinant maps to the CA (capsid) and adjacent spacer sequences within Gag, but systematic mapping of the HIV Gag protein has not been reported. To locate the size determinants of HIV-1, we analyzed a large collection of Gag mutants. To our surprise, all mutants with defects in the MA (matrix), CA, and the N-terminal part of NC (nucleocapsid) sequences produced dense particles of normal size, suggesting that oncoviruses (RSV) and lentiviruses (HIV-1) have different size-controlling elements. The most important region found to be critical for determining HIV-1 particle size is the p6 sequence. Particles lacking all or small parts of p6 were uniform in size distribution but very large as measured by rate zonal gradients. Further evidence for this novel function of p6 was obtained by placing this sequence at the C terminus of RSV CA mutants that produce heterogeneously sized particles. We found that the RSV-p6 chimeras produced normally sized particles. Thus, we present evidence that the entire p6 sequence plays a role in determining the size of a retroviral particle.     

The C-Terminal Half of the Human Immunodeficiency Virus Type 1 Gag Precursor Is Sufficient for Efficient Particle Assembly

Human immunodeficiency virus type 1 particle assembly is directed by the Gag polyprotein Pr55^gag, the precursor for the matrix (MA), capsid (CA), and nucleocapsid proteins of the mature virion. We now show that CA sequences N terminal to the major homology region (MHR), which form a distinct domain, are dispensable for particle formation. However, slightly larger deletions which extend into the MHR severely impair particle production. Remarkably, a deletion which removed essentially all MA and CA sequences between the N-terminal myristyl anchor and the MHR reduced the yield of extracellular particles only moderately. Particle formation even exceeded wild-type levels when additional MA sequences, either from the N or the C terminus of the domain, were retained. We conclude that no distinct region between the myristyl anchor and the MHR is required for efficient particle assembly or release.     

Cytosolic Gag p24 as an Index of Productive Entry of Human Immunodeficiency Virus Type 1

We have investigated the cellular uptake of Gag p24 shortly after exposure of cells to human immunodeficiency virus (HIV) particles. In the absence of envelope glycoprotein on virions or of viral receptors or coreceptors at the cell surface, p24 was incorporated in intracellular vesicles but not detected in the cytosolic subcellular fraction. When appropriate envelope-receptor interactions could occur, the nonspecific vesicular uptake was still intense and cytosolic p24 represented 10 to 40% of total intracellular p24. The measurement of cytosolic p24 early after exposure to HIV type 1 is a reliable assay for investigating virus entry and early events leading to authentic cell infection.The entry of human immunodeficiency virus type 1 (HIV-1) into target cells follows receptor-mediated attachment of viral particles to the cell surface. The cell surface receptor for HIV-1 is the CD4 molecule (7, 15), which promotes attachment of the particle to the cell surface. Fusion between the viral and plasma membranes leading to virus entry into the cytoplasm also requires interaction with a coreceptor. Various chemokine receptors ensure this function. The CXCR4 receptor is used by lymphotropic virus strains (10), whereas the entry of macrophage-tropic and of most primary isolates is processed through interaction with the CCR5 receptor (8, 9). Interactions with CD4 and with a coreceptor expose highly hydrophobic epitopes at the N terminus of the gp41 transmembrane component of envelope, leading to subsequent fusion between viral and cell membranes (6, 17, 34, 35).Several observations have suggested that the fusion process takes place at the cell surface: (i) HIV infection is pH independent, whereas infection by most viruses entering through the endocytic pathway is inhibited by weak bases and ionophore agents (20, 32); (ii) HIV fusion images have been observed at the cell surface (11); (iii) endocytosis of CD4 is not required for entry (18, 20, 25, 28, 32); and (iv) mutant CXCR5 receptors which are not endocytosed in response to ligand binding still function as HIV coreceptors (2). However, other considerations led to the assumption that although HIV entry is clearly pH independent, it may not necessarily be endocytosis independent: (i) images of HIV particles internalized in endocytic vesicles and undergoing fusion with endosomal membranes have been observed (11, 27), (ii) pH-independent entry via endosomal vesicles has been reported for poliovirus (29), (iii) binding and cross-linking by multivalent virus particles may induce endocytic behavior of cell surface receptors different from that induced by their natural ligands, and (iv) endocytosis of CD4 and that of coreceptors have not been simultaneously examined after HIV exposure. Moreover, since studies of virus entry have been performed with cells where the endocytic pathway is active, it is difficult to determine whether particular fusion events at the cell surface or in endosomal vesicles give rise to productive infection.With the aim of examining the role of endosomal HIV particle uptake, infection was synchronized by exposing cells to the virus at 4°C, cells were warmed at 37°C, and p24 was measured in the vesicular and cytosolic fractions of cell extracts. p24 was detected in intracellular vesicles regardless of whether exposure to virus particles could give rise to authentic infection or not. On the other hand, the detection of p24 in cytosolic fractions was strictly associated with authentic infectious events. However, it represented a minor fraction of intracellular p24. Thus, although vesicular uptake is quantitatively the main route of virus particle internalization, it is essentially a dead end with respect to cell infection.     

Nonreciprocal Packaging of Human Immunodeficiency Virus Type 1 and Type 2 RNA: a Possible Role for the p2 Domain of Gag in RNA Encapsidation

The ability of human immunodeficiency virus types 1 (HIV-1) and 2 (HIV-2) to cross-package each other’s RNA was investigated by cotransfecting helper virus constructs with vectors derived from both viruses from which the gag and pol sequences had been removed. HIV-1 was able to package both HIV-1 and HIV-2 vector RNA. The unspliced HIV-1 vector RNA was packaged preferentially over spliced RNA; however, unspliced and spliced HIV-2 vector RNA were packaged in proportion to their cytoplasmic concentrations. The HIV-2 helper virus was unable to package the HIV-1 vector RNA, indicating a nonreciprocal RNA packaging relationship between these two lentiviruses. Chimeric proviruses based on HIV-2 were constructed to identify the regions of the HIV-1 Gag protein conferring RNA-packaging specificity for the HIV-1 packaging signal. Two chimeric viruses were constructed in which domains within the HIV-2 gag gene were replaced by the corresponding domains in HIV-1, and the ability of the chimeric proviruses to encapsidate an HIV-1-based vector was studied. Wild-type HIV-2 was unable to package the HIV-1-based vector; however, replacement of the HIV-2 nucleocapsid by that of HIV-1 generated a virus with normal protein processing which could package the HIV-1-based vector. The chimeric viruses retained the ability to package HIV-2 genomic RNA, providing further evidence for a lack of reciprocity in RNA-packaging ability between the HIV-1 and HIV-2 nucleocapsid proteins. Inclusion of the p2 domain of HIV-1 Gag in the chimera significantly enhanced packaging.     

Mutations within Four Distinct Gag Proteins Are Required To Restore Replication of Human Immunodeficiency Virus Type 1 after Deletion Mutagenesis within the Dimerization Initiation Site

Human immunodeficiency virus type 1 (HIV-1) genomic RNA segments at nucleotide (nt) positions +240 to +274 are thought to form a stem-loop secondary structure, termed SL1, that serves as a dimerization initiation site for viral genomic RNA. We have generated two distinct deletion mutations within this region, termed BH10-LD3 and BH10-LD4, involving nt positions +238 to +253 and +261 to +274, respectively, and have shown that each of these resulted in significant diminutions in levels of viral infectiousness. However, long-term culture of each of these viruses in MT-2 cells resulted in a restoration of infectiousness, due to a series of compensatory point mutations within four distinct proteins that are normally cleaved from the Gag precursor. In the case of BH10-LD3, these four mutations were MA1, CA1, MP2, and MNC, and they involved changes of amino acid Val-35 to Ile within the matrix protein (MA), Ile-91 to Thr within the capsid (CA), Thr-12 to Ile within p2, and Thr-24 to Ile within the nucleocapsid (NC). The order in which these mutations were acquired by the mutated BH10-LD3 was MNC > CA1 > MP2 > MA1. The results of site-directed mutagenesis studies confirmed that each of these four substitutions contributed to the increased viability of the mutated BH10-LD3 viruses and that the MNC substitution, which was acquired first, played the most important role in this regard. Three point mutations, MP2, MNC, and MA2, were also shown to be sequentially acquired by viruses that had emerged in culture from the BH10-LD4 deletion. The first two of these were identical to those described above, while the last involved a change of Val-35 to Leu. All three of these substitutions were necessary to restore the infectiousness of mutated BH10-LD4 viruses to wild-type levels, although the MP2 mutation alone, but neither of the other two substitutions, was able to confer some viability on BH10-LD4 viruses. Studies of viral RNA packaging showed that the BH10-LD4 deletion only marginally impaired encapsidation while the BH10-LD3 deletion caused a severe deficit in this regard.     

Actin Associates with the Nucleocapsid Domain of the Human Immunodeficiency Virus Gag Polyprotein

Recently, it was shown that actin molecules are present in human immunodeficiency virus type 1 (HIV-1) particles. We have examined the basis for incorporation and the location of actin molecules within HIV-1 and murine retrovirus particles. Our results show that the retroviral Gag polyprotein is sufficient for actin uptake. Immunolabeling studies demonstrate that actin molecules localize to a specific radial position within the immature particle, clearly displaced from the matrix domain underneath the viral membrane but in proximity to the nucleocapsid (NC) domain of the Gag polyprotein. When virus or subviral Gag particles were disrupted with nonionic detergent, actin molecules remained associated with the disrupted particles. Actin molecules remained in a stable complex with the NC cleavage product (or an NC-RNA complex) after treatment of the disrupted HIV-1 particles with recombinant HIV-1 protease. In contrast, matrix and capsid molecules were released. The same result was obtained when mature HIV-1 particles were disrupted with detergent. Taken together, these results indicate that actin molecules are associated with the NC domain of the viral polyprotein.     

The I Domain Is Required for Efficient Plasma Membrane Binding of Human Immunodeficiency Virus Type 1 Pr55Gag

The interaction of the human immunodeficiency virus type 1 (HIV-1) Pr55^Gag molecule with the plasma membrane of an infected cell is an essential step of the viral life cycle. Myristic acid and positively charged residues within the N-terminal portion of MA constitute the membrane-binding domain of Pr55^Gag. A separate assembly domain, termed the interaction (I) domain, is located nearer the C-terminal end of the molecule. The I domain is required for production of dense retroviral particles, but has not previously been described to influence the efficiency of membrane binding or the subcellular distribution of Gag. This study used a series of Gag-green fluorescent protein fusion constructs to define a region outside of MA which determines efficient plasma membrane interaction. This function was mapped to the nucleocapsid (NC) region of Gag. The minimal region in a series of C-terminally truncated Gag proteins conferring plasma membrane fluorescence was identified as the N-terminal 14 amino acids of NC. This same region was sufficient to create a density shift in released retrovirus-like particles from 1.13 to 1.17 g/ml. The functional assembly domain previously termed the I domain is thus required for the efficient plasma membrane binding of Gag, in addition to its role in determining the density of released particles. We propose a model in which the I domain facilitates the interaction of the N-terminal membrane-binding domain of Pr55^Gag with the plasma membrane.     

Retention of the Human Immunodeficiency Virus Type 1 Envelope Glycoprotein in the Endoplasmic Reticulum Does Not Redirect Virus Assembly from the Plasma Membrane

The envelope glycoprotein (Env) of human immunodeficiency virus type 1 (HIV-1) has been shown to redirect the site of virus assembly in polarized epithelial cells. To test whether localization of the glycoprotein exclusively to the endoplasmic reticulum (ER) could redirect virus assembly to that organelle in nonpolarized cells, an ER -retrieval signal was engineered into an epitope-tagged variant of Env. The epitope tag, attached to the C terminus of Env, did not affect the normal maturation and transport of the glycoprotein or the incorporation of Env into virions. The epitope-tagged Env was also capable of mediating syncytium formation and virus entry with a similar efficiency to that of wild-type Env. When the epitope was modified to contain a consensus K(X)KXX ER retrieval signal, however, the glycoprotein was no longer proteolytically processed into its surface and transmembrane subunits and Env could not be detected at the cell surface by biotinylation. Endoglycosidase H analysis revealed that the modified Env was not transported to the Golgi apparatus. Immunofluorescent staining patterns were also consistent with the exclusion of Env from the Golgi. As expected, cells expressing the modified Env failed to form syncytia with CD4⁺ permissive cells. Despite this tight localization of Env to the ER, when the modified Env was expressed in the context of virus, virions continued to be produced efficiently from the plasma membrane of transfected cells. However, these virions contained no detectable glycoprotein and were noninfectious. Electron microscopy revealed virus budding from the plasma membrane of these cells, but no virus was seen assembling at the ER membrane and no assembled virions were found within the cell. These results suggest that the accumulation of Env in an intracellular compartment is not sufficient to redirect the assembly of HIV Gag in nonpolarized cells.     

Translation Elongation Factor 1-Alpha Interacts Specifically with the Human Immunodeficiency Virus Type 1 Gag Polyprotein

Human immunodeficiency virus type 1 (HIV-1) gag-encoded proteins play key functions at almost all stages of the viral life cycle. Since these functions may require association with cellular factors, the HIV-1 matrix protein (MA) was used as bait in a yeast two-hybrid screen to identify MA-interacting proteins. MA was found to interact with elongation factor 1-alpha (EF1alpha), an essential component of the translation machinery that delivers aminoacyl-tRNA to ribosomes. EF1alpha was then shown to bind the entire HIV-1 Gag polyprotein. This interaction is mediated not only by MA, but also by the nucleocapsid domain, which provides a second, independent EF1alpha-binding site on the Gag polyprotein. EF1alpha is incorporated within HIV-1 virion membranes, where it is cleaved by the viral protease and protected from digestion by exogenously added subtilisin. The specificity of the interaction is demonstrated by the fact that EF1alpha does not bind to nonlentiviral MAs and does not associate with Moloney murine leukemia virus virions. The Gag-EF1alpha interaction appears to be mediated by RNA, in that basic residues in MA and NC are required for binding to EF1alpha, RNase disrupts the interaction, and a Gag mutant with undetectable EF1alpha-binding activity is impaired in its ability to associate with tRNA in cells. Finally, the interaction between MA and EF1alpha impairs translation in vitro, a result consistent with a previously proposed model in which inhibition of translation by the accumulation of Gag serves to release viral RNA from polysomes, permitting the RNA to be packaged into nascent virions.     

The Late-Domain-Containing Protein p6 Is the Predominant Phosphoprotein of Human Immunodeficiency Virus Type 1 Particles

The Gag-derived protein p6 of human immunodeficiency virus type 1 (HIV-1) plays a crucial role in the release of virions from the membranes of infected cells. It is presumed that p6 and functionally related proteins from other viruses act as adapters, recruiting cellular factors to the budding site. This interaction is mediated by so-called late domains within the viral proteins. Previous studies had suggested that virus release from the plasma membrane shares elements with the cellular endocytosis machinery. Since protein phosphorylation is known to be a regulatory mechanism in these processes, we have investigated the phosphorylation of HIV-1 structural proteins. Here we show that p6 is the major phosphoprotein of HIV-1 particles. After metabolic labeling of infected cells with [ortho-32P]phosphate, we found that phosphorylated p6 from infected cells and from virus particles consisted of several forms, suggesting differential phosphorylation at multiple sites. Apparently, phosphorylation occurred shortly before or after the release of p6 from Gag and involved only a minor fraction of the total virion-associated p6 molecules. Phosphoamino acid analysis indicated phosphorylation at Ser and Thr, as well as a trace of Tyr phosphorylation, supporting the conclusion that multiple phosphorylation events do occur. In vitro experiments using purified virus revealed that endogenous or exogenously added p6 was efficiently phosphorylated by virion-associated cellular kinase(s). Inhibition experiments suggested that a cyclin-dependent kinase or a related kinase, most likely ERK2, was involved in p6 phosphorylation by virion-associated enzymes.