Previous Article | Next Article ![]()
Journal of Virology, January 2003, p. 726-731, Vol. 77, No. 1
0022-538X/03/$08.00+0 DOI: 10.1128/JVI.77.1.726-731.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute,1 Department of Pathology, Division of AIDS, Harvard Medical School, and Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, Massachusetts 021152
Received 29 July 2002/ Accepted 24 September 2002
|
|
|---|
|
|
|---|
CD4 and the chemokine receptors of several Old World monkey species are able to support HIV-1 entry (10, 39). However, HIV-1 replication is greater than 100-fold lower in the peripheral blood mononuclear cells (PBMC) of rhesus macaques, an Old World monkey species, than SIVmac replication (27). The products of HIV-1 reverse transcription are significantly reduced in rhesus macaque cells relative to those in human PBMC, suggesting that a major block to productive HIV-1 replication in these monkey cells occurs following entry of the viral core into the host cell but prior to completion of reverse transcription (5a, 11a, 27). Studies of HIV-SIV chimeras indicate that HIV-1 regions in the 5' half of the genome are responsible for the observed restricted replication in macaque cells (35, 46). However, HIV-SIV chimeras containing the HIV-1 reverse transcriptase replicate efficiently in rhesus macaques, indicating that this viral protein is not sufficient to mediate the postentry restriction (4, 50). Additionally, because the viral Vif protein exerts its phenotypic effects in a manner dependent on the virus-producing cell (21, 23), Vif is less likely to be involved in this species-specific block to HIV-1 infection, which is determined by the target cell rather than the virus-producing cell. Thus, by process of elimination, components of the HIV-1 Gag protein may be responsible for determining the species-specific restrictions (17, 44-46). The Gag proteins of HIV-1 and other retroviruses function at several different levels of the retrovirus life cycle, including assembly, maturation, uncoating of the virion core following entry, and reverse transcription (22). HIV-1 Gag is initially synthesized as a 55-kDa polyprotein, p55, and is subsequently cleaved by the viral protease into the four mature Gag proteins p17 matrix, p24 capsid (CA), p7 nucleocapsid (NC), and p6 (22). The CA and NC domains of primate immunodeficiency virus Gag polyproteins are joined by a p2 linker region (1). Each of the four Gag proteins interacts with viral and host cell factors at various points in the viral life cycle (21, 22, 33, 37, 47, 48, 53). For example, efficient viral replication requires the incorporation of cyclophilin A into HIV-1 but not SIVmac virions (7-9, 20, 37, 49). It is thought that the prolyl isomerase function of cyclophilin A is necessary for inducing a conformational change in the CA protein, facilitating the uncoating of the viral core following entry (6, 25, 36, 37). However, cyclophilin A does not appear to be responsible for the species-specific block to HIV-1 infection (52).
Cell type-specific factors influence the infectivity of other retroviruses, such as the murine leukemia virus (MuLV). The early phase of MuLV infection can be dominantly blocked by a host cell factor, the product of the Fv1 allele (41-43). Restriction by Fv1 abrogates MuLV infection after reverse transcription but prior to entry of the preintegration complex into the nucleus (30, 31). Fv1-mediated resistance is not absolute and can be at least partially overcome by high multiplicities of infection (26). The Fv1 allele is derived from a murine endogenous retroviral gag gene, the product of which can influence the function of the incoming MuLV CA protein during postentry events (26, 34).
In this study, we tested the hypothesis that the HIV-1 and SIVmac CA proteins determine the early restrictions on infection by these viruses in Old World and New World monkey cells, respectively. This hypothesis is consistent with the previous observation that an SIVmac mutant expressing the HIV-1 CA-p2 domain replicated in human cells but failed to replicate in macaque PBMC, suggesting that this Gag region might play an important role in viral tropism (17). In the present study, we generated several HIV-1/SIVmac Gag CA recombinants and mutants in the context of HIV-1 and SIVmac vectors expressing green fluorescent protein (GFP). These recombinant viruses are capable of a single round of infection and allow assessment of the efficiency of the early phase of retrovirus replication. These viruses were analyzed for infection efficiency on a panel of cells, of which some are able to be infected comparably by HIV-1 and SIVmac and others are restricted for infection by one of the viruses.
Construction of viruses with chimeric Gag proteins. To investigate the contribution of the HIV-1 and SIVmac CA protein to infection of human and simian cells, viruses containing chimeric Gag proteins with HIV-1 (HXBc2 strain) and SIVmac239 sequences were created. In the initial set of recombinants, the CA domain of either HIV-1 or SIVmac was precisely substituted for the CA domain in the heterologous proviral vector. However, these chimeric vectors were found to be replication defective, even in human cell lines that support efficient HIV-1 and SIVmac replication (data not shown). Further analysis revealed that the proteolytic processing of these chimeric Gag polyproteins at the CA-p2 junction was inefficient (data not shown). Therefore, additional chimeras were created or adapted for use in this study; in these chimeras, the C-terminal junction between HIV-1 and SIVmac sequences is located either N terminal or C terminal to the CA-p2 cleavage site (Fig. 1). In the HIV(SCA) chimera, residues 1 to 204 of the HIV-1 CA were replaced by the corresponding SIVmac CA residues, in the context of an HIV-1 vector. The HIV (H/SCA) construct contains most of the C-terminal CA domain of SIVmac, replacing HIV-1 CA residues 146 to 204 in the HIV-1 vector. In both the HIV(SCA) and HIV(H/SCA) chimeras, the C-terminal SIVmac/HIV-1 junction was positioned 26 amino acids N terminal to the CA-p2 cleavage site. The HIV(S/HCA) chimera contains the SIVmac amino-terminal CA domain, replacing HIV-1 CA residues 1 to 146, in the HIV-1 vector. The SIV(HCA-p2) chimera contains the complete HIV-1 CA and p2 sequences in the context of an SIV vector, as previously described (17). Finally, an HIV-1 vector containing a single residue change (T58C) in CA was generated. Cysteine is found at the corresponding residue in the SIV, but not the HIV-1, CA protein. The T58C mutant was included in the study for comparison.
![]() View larger version (22K): [in a new window] |
FIG. 1. HIV-1/SIVmac recombinants. The name of the construct reflects the HIV-1 or SIV proviral backbone, and the names in parentheses designate the source of the CA or CA-p2 sequences (H = HIV-1; S = SIVmac). The amino-terminal portions of the Gag polyproteins of the recombinant viruses used in this study are shown beneath the wild-type HIV-1 (white) and SIVmac239 (black) sequences. The locations of the matrix (MA), CA, p2, and NC proteins are shown. The numbers represent the boundaries of the HIV-1 CA sequence that was replaced by the SIVmac sequence. The locations of the N- and C-terminal domains of the CA proteins are shown. The asterisk represents the approximate location of a threonine-to-cysteine change at residue 58 of the HIV-1 CA protein.
|
P1
envpA (40) or SIVmac GFP-expressing proviral vector pSIvec1.GFP (29) by using the restriction enzyme combination BssHII and PspOMI or DraIII and SbfI, respectively. All constructs were sequenced to verify the position of the HIV-1/SIVmac junction and to rule out unwanted changes. Single-round replication-competent HIV-1 virions pseudotyped by the vesicular stomatitis virus (VSV) G glycoprotein were made as previously described (29, 51), by cotransfecting 293T cells with the GFP-expressing HIV-1 vector (pHIvec2.GFP), the appropriate HIV-1 gag/pol construct, pHCMV-G (a VSV G envelope glycoprotein-expressing plasmid), and a Rev-expressing plasmid (psRev) in a 10:10:2:1 ratio. Single-round SIVmac virions were made by cotransfecting pSIvec1.GFP (or chimeras), pHCMV-G, and psRev in a 20:2:1 ratio as previously described (29). 293T cells were plated at a density of 106/100-mm-diameter plate 24 h prior to transfection. Cells were transfected by calcium phosphate precipitation and incubated for 10 h before washing and replacement of the medium with RPMI 1640 medium containing 10% fetal bovine serum and 1% penicillin-streptomycin. Approximately 36 h following transfection, the supernatant containing recombinant viruses was harvested, filtered (0.45-µm pore size), and frozen in aliquots. To determine the amount of virus in the medium, an aliquot of frozen supernatant was thawed and analyzed by the reverse transcriptase assay (35). All of the constructs represented in Fig. 1 produced readily detectable quantities of virus (data not shown). Virion proteins of HIV-1/SIVmac chimeras. To examine the protein composition of virions produced by the chimeric gag constructs, the above-described transfections were repeated with 293T cells without the addition of an envelope glycoprotein-expressing plasmid. At approximately 20 h following transfection, the medium was removed and the cultures were metabolically labeled with [35S]methionine (50 µCi/ml) for 16 h. Virions released into the supernatant were pelleted through a 20% sucrose cushion as described previously (17). The pelleted viral particles were lysed and resuspended in radioimmunoprecipitation assay buffer and analyzed directly by sodium dodecyl sulfate-12.5% polyacrylamide gel electrophoresis and autoradiography.
The virion components consisted of the CA protein (p24 for HIV-1 and p27 for SIVmac) and smaller Gag products (Fig. 2). Proteolytic processing of the chimeric Gag precursor proteins was qualitatively and quantitatively similar to that of the wild-type HIV-1 or SIVmac protein, except for the presence of an additional 23-kDa protein that appeared in the HIV(H/SCA) chimera. The presence of this additional band may reflect aberrant processing of this chimeric Gag protein. Comparison of the migration of the HIV(S/HCA) CA protein (lane 2) with that of the HIV(H/SCA) (lane 3) CA protein suggests that the slower mobility of the SIVmac CA protein compared with that of the HIV-1 CA protein is determined by the amino-terminal SIVmac CA domain. Given the sequence similarities and equivalent expected sizes of the HIV-1 and SIVmac amino-terminal CA domains, the basis for this migration difference is not apparent.
![]() View larger version (63K): [in a new window] |
FIG. 2. Virion proteins produced by HIV-1/SIVmac chimeras. 293T cells were transfected with plasmids HIV.GFP (lane 1), HIV(S/HCA).GFP (lane 2), HIV(H/SCA).GFP (lane 3), HIV(SCA).GFP (lane 4), HIV(T58C).GFP (lane 5), SIV.GFP (lane 6), and SIV(HCA-p2).GFP (lane 7). Following transfection, the cells were labeled with [35S]methionine. Virions released into the cell supernatants were pelleted through 20% sucrose and resuspended in radioimmunoprecipitation assay buffer before analysis by sodium dodecyl sulfate-12.5% polyacrylamide gel electrophoresis and autoradiography. The location of the CA proteins is shown on the left. The values on the right are molecular sizes in kilodaltons.
|
The infectivity of HIV(H/SCA).GFP was negligible in all of the cells tested (data not shown). The aberrant proteolytic processing of the HIV(H/SCA) Gag precursor protein (Fig. 2, lane 3) may contribute to this poor level of infectivity. Figure 3A illustrates the results obtained with the other recombinant viruses in this panel of permissive cells. All four cell lines were infected efficiently by the HIV.GFP and SIV.GFP vectors. In general, the infectivities of the Gag chimeras were lower than that of either the HIV.GFP or the SIV.GFP vector. Apparently, some aspect of the postentry function of Gag is deleteriously affected in these chimeras, despite efficient virion production and processing of the Gag polyprotein. None of the chimeras was consistently more defective than the others in the permissive cell lines studied.
![]() View larger version (32K): [in a new window] |
FIG. 3. Infection of cells by HIV-1/SIVmac recombinants. Supernatants containing the indicated single-round HIV-1 and SIVmac vectors, which encode GFP, were assessed for reverse transcriptase (RT) activity. The indicated number of RT units of virus was added to cells, which were incubated for 3 days prior to fluorescence-activated cell sorting analysis for GFP expression. The cells in the top row (A) are permissive for both HIV-1 and SIVmac, the cells in the second row (B) are permissive for SIVmac but restricted for HIV-1, and the cells in panel C are permissive for HIV-1 but restricted for SIVmac. The symbols and lines used for the viruses are as follows: viruses with HIV-1 CA sequences, open symbols and dotted lines; viruses with SIVmac CA sequences, black symbols and solid lines; virus with CA sequences derived from SIVmac and HIV-1, gray symbol and dashed line.
|
The infectivity of the chimeric viruses in a cell line restricted for SIVmac but permissive for HIV-1 was examined. A cell line, Pindak, derived from squirrel monkeys, a New World species, was used as a representative target cell. As expected, HIV.GFP was significantly more efficient than SIV.GFP at infecting Pindak cells (Fig. 3C). The viruses with the HIV-1 CA domain, HIV(T58C).GFP and SIV(HCA-p2).GFP, infected Pindak cells almost as efficiently as did HIV.GFP. By contrast, viruses with SIVmac CA components, HIV(SCA).GFP and HIV(S/HCA).GFP, inefficiently infected these cells. Thus, the CA protein appears to be a major viral determinant of the early-phase SIVmac restriction in Pindak cells. The amino-terminal domain of the SIVmac CA protein is apparently sufficient to render the virus susceptible to the Pindak cell restriction.
In this study, we created HIV-1 and SIVmac vectors containing chimeric Gag proteins. The data obtained with the subset of these chimeric viruses competent for a single round of infection indicate that the CA protein is a major determinant of the early-phase, postentry HIV-1 or SIV block observed in nonhuman primates. HIV(SCA).GFP, which contains the SIVmac CA protein in an HIV-1 background, exhibited some decrease in the ability to infect permissive human and COS-1 cells, compared with HIV.GFP or SIV.GFP. Despite this generalized decrease in infectivity, HIV(SCA).GFP efficiently infected three cell types, including PRL cells, that are restricted for HIV-1 infection. In each case, HIV(SCA).GFP infected the cells more efficiently than did wild-type HIV.GFP. In these same cells, SIV(HCA-p2).GFP, which contains the HIV-1 CA-p2 region in an SIVmac background, was as inefficient as HIV.GFP. Thus, even though SIV(HCA-p2).GFP exhibited a reasonable ability to infect cells permissive for HIV-1 infection, infection by this virus was inefficient in Old World monkey cells in which HIV-1 is restricted. These results strongly implicate the viral CA as a determinant of this restriction.
SIVmac, but not HIV-1, encounters early-phase, postentry restrictions in many New World monkey cells. The chimeric viruses with the HIV-1 CA domain efficiently infected a New World monkey cell line, Pindak. By contrast, chimeric viruses with the SIVmac CA domain exhibited much less efficient infection of these cells. Thus, it appears that viral CA determinants mediate early-phase blocks to primate immunodeficiency viruses in cells derived from Old World and New World monkeys.
Further research is required to determine whether the mechanisms of viral restriction are similar in Old World and New World monkeys. This line of inquiry will be assisted by a precise definition of the CA region that specifies the restriction. The HIV-1 and SIVmac CA proteins are composed of N-terminal and C-terminal domains. Only one of the chimeras containing heterologous CA domains was replication competent. HIV(S/HCA).GFP, in contrast to the HIV-1 variant containing the complete SIVmac CA protein, behaved more like HIV-1 in cells restricted for HIV-1 infection and more like SIVmac in a cell line restricted for SIVmac infection. This could indicate differences in the precise CA determinants for the restrictions in these cell types. Alternatively, the CA structural requirements that govern the molecular interactions relevant to the restriction may involve both the N- and C-terminal domains. The results reported herein should assist efforts to elucidate the basis of the species-specific blocks to primate immunodeficiency virus infection.
This work was supported by a grant from the National Institutes of Health (HL54785) and by a Center for AIDS Research award (AI28691). We also thank William F. McCarty-Cooper and the Bristol-Myers Squibb Foundation for support.
|
|
|---|
, MIP-1ß receptor as a fusion cofactor for macrophage-tropic HIV-1. Science 272:1955-1958.[Abstract]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»