Proteolytic refolding of the HIV-1 capsid protein amino-terminus facilitates viral core assembly
1998; Springer Nature; Volume: 17; Issue: 6 Linguagem: Inglês
10.1093/emboj/17.6.1555
ISSN1460-2075
AutoresUta K. von Schwedler, Timothy L. Stemmler, Victor Y. Klishko, Sam Li, Kurt H. Albertine, Darrell R. Davis, Wesley I. Sundquist,
Tópico(s)Biochemical and Molecular Research
ResumoArticle16 March 1998free access Proteolytic refolding of the HIV-1 capsid protein amino-terminus facilitates viral core assembly Uta K. von Schwedler Uta K. von Schwedler Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Timothy L. Stemmler Timothy L. Stemmler Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Victor Y. Klishko Victor Y. Klishko Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Su Li Su Li Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Kurt H. Albertine Kurt H. Albertine Department of Pediatrics, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Darrell R. Davis Darrell R. Davis Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Department of Medicinal Chemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Wesley I. Sundquist Corresponding Author Wesley I. Sundquist Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Uta K. von Schwedler Uta K. von Schwedler Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Timothy L. Stemmler Timothy L. Stemmler Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Victor Y. Klishko Victor Y. Klishko Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Su Li Su Li Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Kurt H. Albertine Kurt H. Albertine Department of Pediatrics, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Darrell R. Davis Darrell R. Davis Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Department of Medicinal Chemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Wesley I. Sundquist Corresponding Author Wesley I. Sundquist Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA Search for more papers by this author Author Information Uta K. von Schwedler1, Timothy L. Stemmler1, Victor Y. Klishko1, Su Li1, Kurt H. Albertine2, Darrell R. Davis1,3 and Wesley I. Sundquist 1 1Department of Biochemistry, University of Utah, Salt Lake City, UT, 84132 USA 2Department of Pediatrics, University of Utah, Salt Lake City, UT, 84132 USA 3Department of Medicinal Chemistry, University of Utah, Salt Lake City, UT, 84132 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:1555-1568https://doi.org/10.1093/emboj/17.6.1555 Correction(s) for this article Proteolytic refolding of the HIV-1 capsid protein amino-terminus facilitates viral core assembly15 May 2000 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info After budding, the human immunodeficiency virus (HIV) must 'mature' into an infectious viral particle. Viral maturation requires proteolytic processing of the Gag polyprotein at the matrix–capsid junction, which liberates the capsid (CA) domain to condense from the spherical protein coat of the immature virus into the conical core of the mature virus. We propose that upon proteolysis, the amino-terminal end of the capsid refolds into a β-hairpin/helix structure that is stabilized by formation of a salt bridge between the processed amino-terminus (Pro1) and a highly conserved aspartate residue (Asp51). The refolded amino-terminus then creates a new CA–CA interface that is essential for assembling the condensed conical core. Consistent with this model, we found that recombinant capsid proteins with as few as four matrix residues fused to their amino-termini formed spheres in vitro, but that removing these residues refolded the capsid amino-terminus and redirected protein assembly from spheres to cylinders. Moreover, point mutations throughout the putative CA–CA interface blocked capsid assembly in vitro, core assembly in vivo and viral infectivity. Disruption of the conserved amino-terminal capsid salt bridge also abolished the infectivity of Moloney murine leukemia viral particles, suggesting that lenti- and oncoviruses mature via analogous pathways. Introduction Retroviral assembly is initially driven by polymerization of the Gag polyprotein, which forms a spherical shell associated with the inner membrane of the freshly budded particle (Figure 1). Concomitant with budding, the viral protease cleaves Gag into a series of smaller, discrete proteins. These processed proteins then rearrange to form the mature, infectious viral particle (reviewed in Kräusslich, 1996). Gag processing thereby permits the orderly transformation from a virion that is competent to assemble and bud from one cell into a virion that can disassemble and replicate in a new host cell. Figure 1.Structure and maturation of the HIV-1 virion. (A) Domain structure of the HIV-1 Gag polyprotein. Locations of the capsid Pro1 and Asp51 residues. Color coding for the different domains of Gag is the same throughout this figure. (B) Schematic structures of immature and mature HIV-1 particles. The figure summarizes current models for the locations of the major virion components and emphasizes the dramatic structural rearrangements that accompany viral maturation. (C) Model for structural rearrangement of the HIV-1 capsid protein upon proteolytic processing at the MA–CA junction of Gag. Ribbon diagrams of the MA and CA domains of the unprocessed Gag protein (left) are based upon the crystal structures of MA (Hill et al., 1996) and the amino- and carboxy-terminal domains of capsid (Gamble et al., 1996, 1997). An expanded model of the processed capsid protein is shown on the right. Two molecules of capsid are shown to illustrate how folding of the capsid β-hairpin (light blue, small arrows) could be coupled to formation of a new CA–CA interface. The amino-terminal capsid interface shown is the major CA–CA interface in the co-crystal structure of CA151 in complex with cyclophilin A. Ribbon diagrams in Figures 1, 2 and 5 were made with the program MidasPlus (Ferrin et al., 1988). Download figure Download PowerPoint The human immunodeficiency virus type 1 (HIV-1) Gag protein is proteolytically processed into the following discrete proteins and spacer peptides: matrix (MA, residues 1–132), capsid (CA, 133–363), p2 (364–376), nucleocapsid (NC, 377–432), p1 (433–448) and p6 (449–500) (Figure 1A). Upon maturation, the matrix protein remains associated with the inner viral membrane, while capsid, nucleocapsid and the viral RNA condense into the center of the virus (Figure 1B). The diploid RNA genome and associated NC proteins form an electron-dense ribonucleoprotein complex, with a concomitant increase in the thermal stability of the dimeric RNA (Fu et al., 1994; Feng et al., 1996). The processed capsid protein then forms a conical shell that encapsidates the RNA–NC copolymer. In the absence of Gag proteolysis, neither capsid core formation (Göttlinger et al., 1989; Peng et al., 1989; Schatzl et al., 1991; Kaplan et al., 1993; Kräusslich et al., 1995) nor RNA stabilization occur (Fu et al., 1994), indicating that the driving force underlying maturation is the creation of new interactions between the processed domains of Gag. Three-dimensional structures are now available for the three major Gag-derived proteins of HIV-1, either as full-length proteins [matrix (Massiah et al., 1994; Matthews et al., 1994; Rao et al., 1995; Hill et al., 1996) and nucleocapsid (Morellet et al., 1992; South and Summers, 1993)] or as a compilation of the two independent folding domains of capsid (Gamble et al., 1996, 1997; Gitti et al., 1996; Momany et al., 1996). It is not clear, however, how accurately these proteins mimic their analogous domains within the unprocessed Gag polyprotein. Significant local structural changes may occur when Gag is cleaved, particularly given the dramatic global changes that accompany viral maturation. Indeed, the detailed structure of the processed amino-terminal domain of capsid strongly suggests that the conformation surrounding the MA–CA junction of Gag changes significantly upon proteolysis (Gamble et al., 1996; Gitti et al., 1996; Momany et al., 1996). The amino-terminal end of the processed capsid protein forms a β-hairpin that spans residues 1–13. The hairpin folds the charged amino-terminus of Pro1 back into the protein, where it forms a buried salt bridge with the carboxylate of Asp51 (Figures 1C and 2). Several observations suggest that the amino-terminal capsid β-hairpin forms after proteolytic cleavage at the MA–CA junction. First, the capsid amino-terminus is uncharged prior to proteolysis and thus cannot form the stabilizing salt bridge with Asp51. Moreover, the buried amino-terminus of the processed capsid protein appears sterically incompatible with a matrix protein extension or with processing by the viral protease, which recognizes the MA–CA junction in an extended conformation (Wlodawer and Erickson, 1993). Figure 2.Ribbon diagram of the capsid amino-terminal interface. Amino acids mutated in this study are shown in red. Download figure Download PowerPoint The proposed proteolytic refolding of capsid is analogous to zymogen activation in the serine proteases, where precursor processing also causes the new amino-terminus to rearrange into a salt bridge with a buried aspartate residue (Sigler et al., 1968). The energetics of the trypsinogen to trypsin folding transition have been studied in detail (Hedstrom et al., 1996). In that case, the salt bridge between the amino-terminus and the buried Asp residue contributes significant stabilization energy (∼3 kcal/mol), as does the packing of the amino-terminal isoleucine side chain into a hydrophobic binding site in the protein (∼5 kcal/mol). Refolding of the HIV-1 capsid amino-terminus appears to be driven by a series of analogous interactions: (i) a salt bridge between the amino-terminus and Asp51, (ii) a second hydrogen bond between the amino-terminus and Gln13 O, (iii) van der Waals contacts between the invariant Pro1 ring and the Cα atoms of Ile15 and Gly46 and (iv) hydrogen bonding interactions between the two strands of the β-hairpin. We propose that the functional consequence of capsid refolding is the creation of a new CA–CA interface in the mature capsid core (the 'amino-terminal capsid interface'). This model is depicted schematically in Figure 1C and is based upon the crystal structure of the amino-terminal domain of capsid in complex with cyclophilin A (Gamble et al., 1996). The major CA–CA interface in these crystals is created by intermolecular packing of capsid helices 1 and 2 into a four-helix bundle (with their symmetry-related pairs; Figures 1C and 2). The four-helix bundle buries a total of 570 Å2/subunit and exhibits a packed hydrophobic core ringed by hydrophilic interactions. Additional intermolecular interactions are formed between the two amino-terminal β-hairpins that project above the four-helix bundle (burying 230 Å2), suggesting how folding of the capsid β-hairpin could be coupled to formation of this interface. In summary, we propose that proteolysis at the MA–CA junction of Gag allows retroviral maturation by refolding the capsid amino-terminus and thereby facilitating the protein's rearrangement into the central conical core. Experiments described here are aimed at testing this mechanistic model for a simple developmental switch. Results Mutations in the capsid amino-terminal interface inhibit capsid assembly in vitro The importance of the amino-terminal CA–CA interface for capsid assembly was tested initially using pure recombinant capsid proteins (Figure 3). In vitro, the HIV-1 capsid protein can assemble into long hollow cylinders (Figure 3A) which presumably utilize many of the same CA–CA interactions as the viral capsid core (Campbell and Vogt, 1995, 1997; Groß et al., 1997). Although the precise relationship between the viral core and the capsid cylinders remains to be elucidated, the two structures appear to share at least a subset of similar CA–CA interactions. For example, as shown in Figure 3B, capsid cylinder assembly is blocked by a point mutation (M185A) that disrupts the well-characterized carboxy-terminal capsid dimer interface and blocks viral replication in culture (Gamble et al., 1997). Instead of cylinders, CA M185A forms long strings of protein, as though destabilizing the repeating carboxy-terminal capsid dimer interface prevents the protein from winding up into a cylinder. Figure 3.Aberrant in vitro assembly of mutant HIV-1 capsid proteins. The figure shows representative thin-section transmission electron microgaphs of the structures formed by wild-type CA (positive control) (A), CA M185A (negative control) (B), CA M39D (C) and CA D51A (D). Each protein was incubated for 1 h at 37°C in the assembly buffer (400 μM protein). Large aggregates (if any) were collected by centrifugation, fixed, stained and analyzed by thin-section TEM. The small circular structures in the micrograph of the wild-type capsid protein are hollow cylinders that were sectioned perpendicular to the cylinder axis. Note that the amorphous protein aggregates formed by CA D51A occasionally also contained imbedded cylinders. Nevertheless, the selected field shows an unusually high concentration of D51A cylinders. Cylinder widths were 32 ± 2 nm (wt CA, n=31) and 33 ± 3 nm (CA D51A, n=20). Scale bars are 100 nm. Download figure Download PowerPoint As described above, the amino-terminal capsid interface is composed of two distinct structural elements: the four-helix bundle and the packed β-hairpins (Figure 2). Two different capsid point mutations, M39D and D51A, were used to test the importance of each of these secondary structural elements for cylinder formation. Met39 is buried within the core of the four-helix bundle, and mutation to Asp was therefore expected to disrupt this hydrophobic core. Asp51 forms a salt bridge with the amino-terminal capsid proline residue, and mutation to Ala was expected to break the salt bridge and thereby destabilize the β-hairpin. The wild-type and mutant capsid proteins were expressed and purified to homogeneity as described in Materials and methods. Concentrated solutions of each protein were incubated at high ionic strength for 1 h at 37°C [400 μM protein, 1 M NaCl, 50 mM Tris–HCl (pH 8.0)]. Oligomeric assemblies were collected by centrifugation, fixed and observed in thin sections by transmission electron microscopy (TEM). As shown in Figure 3A, the wild-type capsid protein efficiently assembled into hollow cylinders averaging 32 ± 2 nm in diameter (n=20) and up to 2000 nm in length. In contrast, both amino-terminal capsid interface mutants were defective in cylinder assembly, although the two mutant proteins behaved somewhat differently. Incubation of CA M39D under assembly conditions produced very little material that could be collected by centrifugation or discerned in thin section by TEM (Figure 3C). Oligomeric CA M39D complexes were also absent in negatively stained samples deposited directly on Formvar carbon-coated copper grids (not shown). Thus, the M39D mutation entirely abrogated the ability of capsid to form large arrays under our assembly conditions. In contrast, the CA D51A protein did form large arrays that were collected by centrifugation. However, electron micrographs revealed that the insoluble material consisted mainly of large amorphous protein aggregates. In some preparations, we observed cylinders amongst the aggregated protein (Figure 3D). Even when they formed, however, these CA D51A cylinders were shorter and far less prevalent than the wild-type CA cylinders. Thus, the D51A mutation inhibited, but did not abolish, capsid cylinder formation. Overall, the data are consistent with a role for helix 2 in cylinder formation, and suggest that the capsid β-hairpin, although not absolutely essential, also contributes to capsid assembly in vitro. Matrix residues redirect capsid assembly from cylinders to spheres The role of the MA–CA junction in capsid assembly was also examined in vitro, using proteins in which the final 28, 6 or 4 amino acids of matrix were fused onto the amino-terminus of CA. The first of these proteins, designated MA28–CA, was designed to initiate at the first amino acid beyond the globular domain of the mature matrix protein (Massiah et al., 1994; Matthews et al., 1994; Hill et al., 1996). The other two proteins, designated MA6–CA and MA4–CA, were designed to have minimal matrix extensions because NMR spectroscopic studies revealed that most (or all) of the matrix residues of MA28–CA were disordered in solution (see below) and because the MA28–CA protein was partially insoluble when expressed in Escherichia coli, presumably owing to aggregation of the disordered matrix residues. Unlike the CA protein, the MA–CA fusion proteins did not form cylinders in vitro. Instead, the fusion proteins assembled into spherical particles as well as into amorphous aggregates (Figure 4). These particles were deposited on Formvar carbon-coated copper grids and visualized by negative staining (Figure 4B). The spheres formed by MA4–CA were readily distinguishable from the CA cylinders (compare Figure 4A and B, note scale changes), and in no case were cylinders observed for MA4–CA. The MA4–CA particles were sometimes highly spherical, but could also appear faceted and/or irregular, as has been observed for cryo-EM images of immature HIV-1 particles (Fuller et al., 1997). The average diameter of the negatively stained MA4–CA particles was 55 ± 13 nm (n= 20), which is smaller than estimates for the diameter (∼110 nm) of the capsid ring within the immature virion (Fuller et al., 1997). The inner and outer edges of the spherical shells were often clearly defined by uranyl acetate staining of thin sections of the MA–CA fusion proteins (Figure 4D), and the thickness of the shell was 5.6 ± 0.6 nm (n=31). This thickness is compatible with models for the intact capsid protein derived from crystal structures of its composite domains, which indicate that CA spans ∼60 Å in its longest dimension (Gamble et al., 1997). In the three constructs tested, the efficiency of sphere formation varied inversely with the length of the matrix tail, and the MA28–CA protein formed spheres very inefficiently, with most of the protein simply aggregating (data not shown). Figure 4.Spherical assembly of the MA4–CA protein. Left: wild-type CA (A and C), cylinders; right: MA4–CA spheres (B and D). Direct transmission electron micrographic images of negatively stained particles are shown above (A and B), and thin-section transmission electron micrographs of positively stained particles are shown below (C and D). Particle assembly conditions are as in Figure 3. Amorphous protein aggregates were also observed in thin-section transmission electron micrographs of the MA4–CA preparations (not shown). Scale bars are 100 nm. Download figure Download PowerPoint These experiments demonstrate that fusing as few as four matrix residues onto the amino-terminus of capsid redirects protein assembly from cylinders to spheres. This transformation is strikingly reminiscent of the morphological transformation that accompanies viral maturation, where the unprocessed capsid protein initially participates in forming the spherical protein shell of the immature virus, but then rearranges into the conical core following proteolysis at the MA–CA junction. NMR spectroscopic characterization of capsid protein constructs NMR spectroscopy was used to test whether the protein constructs described above altered the structure of the amino-terminal end of capsid. The NMR studies were performed on monomeric proteins encompassing the first 146 or 151 residues of capsid (i.e. lacking the protein's carboxy-terminal dimerization domain). Complete proton chemical shift assignments have been reported previously for CA151 (Gitti et al., 1996), and it was therefore possible to use amide proton NMR chemical shift perturbations as a sensitive probe for localizing structural changes in the MA4–CA, MA28–CA, CA M39D and CA D51A proteins. The initial chemical shift analysis was performed on the MA28–CA151 fusion protein (not shown). We identified 36 amide protons within the amino-terminal capsid domain of this protein that shifted by >0.2 p.p.m. versus the processed CA151 protein. The shifted residues were clustered throughout the β-hairpin and the helices against which the hairpin normally packs (1, 2, 3 and 6). As expected, all of the shifted resonances returned to their CA151 positions upon cleavage at the MA–CA junction with recombinant HIV-1 protease (not shown). We therefore attribute the chemical shift changes within the capsid domain to structural perturbations introduced by the additional matrix residues. Interestingly, the chemical shifts of at least 17 of the matrix residues did not change significantly upon proteolysis, indicating that the majority of matrix residues in the MA28–CA151 protein were disordered both before and after proteolysis. Thus, this analysis suggested that the matrix residues caused refolding of the amino-terminal end of capsid, but did not themselves adopt a defined structure. To investigate the refolding event further, we analyzed the backbone amide chemical shift changes in the shorter MA4–CA146 fusion protein (capsid residues 147–151, which are disordered in CA151, were deleted from this construct). The 1H, 15N heteronuclear single quantum coherence (HSQC) NMR spectra of CA151 (blue) and MA4–CA146 (red) are shown superimposed in Figure 5A. The majority (91/134) of backbone amide proton resonances within the amino-terminal domain of capsid were not significantly shifted in the MA4–CA protein. However, 43 amide protons were shifted by >0.2 p.p.m. in the 1H dimension or 0.3 p.p.m. in the 15N dimension upon addition of the four matrix residues. Locations of the shifted (red) and unshifted (blue) MA4–CA146 residues were mapped back onto the structure of the amino-terminal domain of capsid (Figure 5B). Strikingly, the backbone amide protons of at least 26 of the first 29 capsid residues, spanning the β-hairpin and helix 1, were significantly shifted in the fusion protein. In addition, 11 of 12 residues spanning helix 3 were also significantly shifted, as were several residues in helices 2 and 6. Amide protons throughout the remainder of the protein remained unshifted (except at the very C-terminal end of the domain, presumably owing to the absence of residues 147–151), indicating that the mutation does not significantly alter the structure of helices 4, 5 and 7 (i.e. the left half of the molecule in Figure 5B). Taken together, these chemical shift analyses strongly support the model that the amino-terminal end of capsid adopts significantly different structures before and after proteolysis. We are currently determining the three-dimensional structure of the MA4–CA fusion protein in order to define precisely how the additional matrix residues refold this region of capsid. Figure 5.NMR spectral mapping of structural perturbations in MA4–CA146 (A and B), CA151 D51A (C and D) and CA151 M39D (E and F). Left: superimpositions of the 1H,15N HSQC NMR spectrum of wild-type CA151 (blue) upon those of MA4–CA146 (A), CA151 D51A (C) and CA151 M39D (E) (red). Right: locations of backbone amide protons that exhibit significant chemical shift changes versus wild-type CA151 in MA4–CA146 (B), CA151 D51A (D) and CA151 M39D (F). Amide protons shifted by >0.2 p.p.m. in the 1H dimension or >0.3 p.p.m. in the 15N dimension (red) are shown mapped back onto the structure of residues 1–146 of the wild-type CA151 protein (blue). Proline residues are depicted in the same color as their immediately adjacent residues. Download figure Download PowerPoint A similar approach was used to localize the structural changes caused by the CA M39D and D51A mutations. As shown in Figure 5C, the D51A mutation caused significant shifts in 51 of the 138 CA151 backbone amide protons versus the native protein. Although Asp51 is located near the amino-terminus of helix 3, chemical shift changes were again propagated throughout the β-hairpin and its adjacent helices (Figure 5D). The perturbed amide proton chemical shifts generally moved toward random coil values, indicating that disruption of the Pro1–Asp51 salt bridge altered the equilibrium to favor unfolding of the amino-terminal end of capsid. In contrast, only 19 of the 138 amide proton residues were significantly shifted in the CA M39D mutant (Figure 5E). In this case, the shifted residues clustered about the mutation site and the structural perturbations were not propagated to the amino-terminal β-hairpin (Figure 5F). This mutation therefore does not significantly disrupt the capsid structure beyond helices 1 and 2. Point mutations within the capsid amino-terminal interface render HIV-1 non-infectious Site-directed mutagenesis was used to examine whether the capsid amino-terminal interface is essential for viral replication. Five different point mutations designed to disrupt the interface were tested for their effects on HIV-1 replication (Figure 2). As described above, the capsid D51A mutation was designed to unfold the β-hairpin and thereby disrupt its intermolecular contacts. Capsid A22D, E28,29A, M39D and A42D point mutations were designed to disrupt intermolecular packing interactions within the hydrophobic core of the four-helix bundle at the CA–CA interface. Each of the mutated residues was selected to lie on the surface of the monomeric CA protein in order to minimize intramolecular structural perturbations. A final mutation (Q7,9A) was used as a control to test the predictive power of the CA151 crystal structure. Gln7 and Gln9 reside on the outside of the β-hairpin loop and make no intermolecular contacts in the CA151 crystal structure. The Q7,9A mutation was therefore not expected to interfere significantly with capsid core assembly. Proviral DNA constructs encoding wild-type and mutant HIV-1NL4−3 genomes were transfected into 293T producer cells, and viral particles were harvested from the supernatant after 2 days. Particle production was analyzed initially by assaying reverse transcriptase activity and capsid (p24) levels in the supernatant. These levels were up to 3-fold lower than wild-type for several of the mutant viruses (Table I). In order to determine the fraction of released capsid protein present in intact viral particles, the virions were purified and concentrated by centrifugation through a 20% sucrose cushion. As shown in Table I, on average 77% of the released wild-type capsid protein was pelletable, whereas this value ranged from 15 to 40% for the various mutants. Similar values were obtained for viruses produced in transfected COS-7 cells (not shown). Hence, the production of intact, stable virions was reduced 4- to 20-fold by the various amino-terminal interface mutations. Nevertheless, all of the mutants produced virions, and we therefore characterized their protein composition, infectivity and morphology. Table 1. Phenotypes of HIV-1 capsid mutants Capsid mutation Virus production Pelletable virusb Infectivity Cone formation RT/p24 assaysa T cell linesc MAGId TEMe Q7,9A 78 ± 8% 65% 4 days delayed 10% yes A22D 30 ± 18% 39% no 1.6% no E28,29A 66 ± 10% 32% no 0.04% no M39D 37 ± 17% 15% no 0.04% no A42D 53 ± 8% 20% no 0.05% no D51A 61 ± 45% 25% no 0.04% no WT 100% 77% yes 100% yes a Virus production from transfected 293T cells, assayed as reverse transcriptase activity and p24 CA antigen levels in the supernatant. Values from both measurements were converted to percentages of wild-type levels in each transfection, averaged, and the values are reported ± 1 SD (n=3–7). Absolute levels of wild-type virus were (19 ± 5)×10 c.p.m./10 μl (n=7) in RT assays and 3.5 ± 0.9 μg p24/ml (n=7) in p24 ELISA assays. b Virion particles were pelleted through 20% sucrose and quantitated by p24 ELISA. The reported values are the percentage of total p24 in the supernatant that pelleted through the sucrose cushion, averaged from two experiments. c Viral replication detected in SupT1, CEM and/or H9 human T cells by assaying reverse transcriptase activity in the supernatant of infected cells (growth curves). d Infectivity in P4 (HeLa.CD4.LTR-β-gal) cells in a single round of infection. Blue cells per ng of p24 in the inoculum are reported as a percentage of wild-type (infectious titer 620 ± 238). Values are an average of three experiments. Note that others have also reported measurable background staining for non-infectious HIV-1 in this assay (e.g. Wu et al., 1997), perhaps because Tat protein is sometimes synthesized from extrachromosomal DNA even in non-productive infections. e Conical viral cores detected by TEM in thin sections of concentrated virions or of cells producing virus. The protein composition of the viral partic
Referência(s)