Adaptation of a CCR5-Using, Primary Human Immunodeficiency Virus Type 1 Isolate for CD4-Independent Replication

Abstract

The gp120 envelope glycoprotein of the human immunodeficiency virus type 1 (HIV-1) promotes virus entry by sequentially binding CD4 and chemokine receptors on the target cell. Primary, clinical HIV-1 isolates require interaction with CD4 to allow gp120 to bind the CCR5 chemokine receptor efficiently. We adapted a primary HIV-1 isolate, ADA, to replicate in CD4-negative canine cells expressing human CCR5. The gp120 changes responsible for the adaptation were limited to alteration of glycosylation addition sites in the V2 loop–V1-V2 stem. The gp120 glycoproteins of the adapted viruses bound CCR5 directly, without prior interaction with CD4. Thus, a major function of CD4 binding in the entry of primary HIV-1 isolates can be bypassed by changes in the gp120 V1-V2 elements, which allow the envelope glycoproteins to assume a conformation competent for CCR5 binding.

Human immunodeficiency viruses type 1 and type 2 (HIV-1 and HIV-2) are the etiologic agents of AIDS in humans (5, 12, 30). Similarly, simian immunodeficiency virus (SIV) can induce an AIDS-like illness in Old World monkeys (18, 34, 41). AIDS is associated with the depletion of CD4-positive T lymphocytes, which are the major target cells of viral infection in vivo (26).

The entry of primate immunodeficiency viruses into target cells is mediated by the viral envelope glycoproteins, gp120 and gp41, which are organized into trimeric complexes on the virion surface (2, 53). Viral entry usually requires the binding of the exterior envelope glycoprotein, gp120, to the primary receptor CD4 (14, 36, 42). The gp120 glycoprotein is heavily glycosylated and contains protruding variable loops (38, 40), features that are thought to decrease the susceptibility of the virus to host immune responses (73, 75). The interaction between gp120 and CD4 promotes a series of conformational changes in gp120 that result in the formation or exposure of a binding site for particular members of the chemokine receptor family that serve as coreceptors (68, 72). The chemokine receptor CCR5 is the major coreceptor for primary HIV-1 isolates (1, 10, 16, 19, 20) and can be utilized by HIV-2 and SIV isolates as well (9, 43). The CXCR4 chemokine receptor is the predominant coreceptor used by the primary T-cell-tropic and laboratory-adapted HIV-1 strains (27). Binding of gp120 to the coreceptor is thought to induce additional conformational changes that lead to activation of the transmembrane glycoprotein gp41 and subsequent fusion of the viral and cellular membranes (8, 61, 69).

In addition to anchoring and orienting the viral envelope glycoproteins with respect to the target cell membrane, binding to CD4 initiates changes in the conformation of the envelope glycoproteins (3, 4, 17, 22, 55–57, 66, 70, 74). Some of these conformational changes allow high-affinity interaction with CCR5 (68, 72). The binding of soluble CD4 (sCD4) (15, 28, 33, 59, 67, 71) to the envelope glycoprotein complexes of some HIV-1 isolates results in dissociation of gp120 from the gp41 glycoprotein (7, 29, 31, 35, 45–47, 65). This shedding of gp120 has been shown to be dependent upon a conserved stem (the V1-V2 stem) from which the V1 and V2 variable loops of gp120 emerge (74). These variable loops, and the V3 variable loop as well, change conformation or become more exposed upon sCD4 binding (22, 48, 55, 56, 64, 74). The CD4-induced movement of the V1-V2 loops results in the exposure of conserved, discontinuous structures on the HIV-1 gp120 glycoprotein recognized by the 17b and 48d monoclonal antibodies (66, 74). Analysis of a panel of gp120 mutants suggested that this conformational change is functionally important for virus entry (64). The close relationship between the 17b and 48d epitopes and conserved gp120 structures shown to be important for CCR5 binding (52) supports a model in which conformational changes in the V1-V2 stem-loop structure induced by CD4 binding create and/or expose high-affinity binding sites for the CCR5 coreceptor.

Insights into the molecular basis for receptor binding by the primate immunodeficiency virus gp120 glycoproteins have been obtained from analysis of antibody binding, mutagenesis, and X-ray crystallography (39, 48–52, 54, 60, 75). These studies suggest that the major variable loops are well exposed on the surface of the gp120 glycoprotein, whereas the more conserved regions fold into a core structure. This HIV-1 gp120 core has been crystallized in a complex with fragments of the CD4 glycoprotein and the 17b monoclonal antibody (39, 75). The gp120 core is composed of an inner and an outer domain and a four-stranded β-sheet (the bridging sheet). Elements of both domains and the bridging sheet contribute to CD4 binding (39). Of particular interest with respect to the induction of the CCR5-binding site by CD4 is the location of the conserved V1-V2 stem. The V1-V2 stem directly contacts CD4 and contributes two strands to the bridging sheet, which has been implicated in CCR5 binding (39, 52). A plausible model is that CD4 binding repositions the V1-V2 stem, allowing formation of the bridging sheet, one gp120 element important for CCR5 binding. The V3 variable loop of gp120 also contributes to chemokine receptor binding (6, 10, 13, 52, 60, 72). Utilization of CCR5 versus CXCR4 by HIV-1 isolates is largely determined by the sequence of the V3 loop (6, 10, 13). V3-deleted versions of gp120 do not bind CCR5, although CD4 binding occurs at wild-type levels (72). Finally, antibodies against the V3 loop interfere with gp120-CCR5 binding (68, 72). The predicted spatial proximity of the V3 loop and the bridging sheet (39, 52, 75) suggest that these two gp120 elements may contribute to a discontinuous structure that binds the chemokine receptor.

Infection by primate immunodeficiency viruses is generally more efficient when CD4 is expressed on the surface of the target cells. However, some viral isolates are able to achieve reasonably efficient infection of cells lacking CD4. For example, some HIV-2 isolates have been shown to enter CD4-negative cells by using CXCR4 (11, 24). Some SIV strains can infect CD4-negative brain capillary endothelial cells or other cell types by using CCR5 as a primary receptor (23, 57). The gp120 glycoproteins of some SIV isolates can efficiently bind rhesus monkey CCR5 in the absence of sCD4 (44). Naturally occurring, CD4-independent HIV-1 isolates appear to be less common, but a CXCR4-using HIV-1 isolate has been derived by passage on CD4-negative cultured cells (21).

Here, to gain insight into the contribution of CD4 binding to virus entry, we derived and studied a CD4-independent HIV-1 isolate that utilizes the CCR5 coreceptor.

MATERIALS AND METHODS

Cell lines.

HeLa, 293, and Cf2Th cells were obtained from the American Type Culture Collection and maintained as previously described (10).

To generate cells expressing human CCR5 (Cf2Th-CCR5), the canine thymocyte cell line Cf2Th was transfected with Lipofectamine (Gibco BRL) with pcDNA3.1 expressing human CCR5 and a neomycin resistance gene (10). G418-resistant colonies were sorted by fluorescence-activated cell sorting for CCR5 expression, with the 2D7 anti-CCR5 monoclonal antibody (Pharmingen). A Cf2Th cell stably expressing CCR5 was cloned by single-cell sorting with the Vantage flow cytometer (Becton Dickinson).

Cf2Th-CD4/CCR5 cells, which express both human CD4 and human CCR5, were generated as described elsewhere (25) and maintained in medium containing 0.6 mg of G418 per ml and 0.15 mg of hygromycin per ml. CCR5 expression, measured by flow cytometry with the 2D7 antibody, was comparable in the Cf2Th-CD4/CCR5 and Cf2Th-CCR5 cells (data not shown).

For the gp120 binding assays, Cf2Th-synCCR5 cells were used. These cells express a codon-optimized version of human CCR5 containing a C-terminal nonapeptide (TETSQVAPA) tag derived from bovine rhodopsin (44a). The C-terminal bovine rhodopsin sequence can be recognized by the 1D4 monoclonal antibody. The Cf2Th-synCCR5 cells were maintained in medium containing 0.8 mg of G418 per ml.

Virus replication.

The pNL4-ADA plasmid was generated by cloning the Asp718-BamHI env fragment from the pSVIIIenv-ADA plasmid (63) into the pNL4-3 plasmid containing an infectious HIV-1 provirus (62). To generate infectious virus, HeLa cells were transfected with 20 μg of pNL4-ADA. The medium was replaced after 10 h, and HeLa supernatants were harvested 3 days later. After clarification by low-speed (200 × g for 10 min) centrifugation, the supernatants were stored at −70°C. Virus in the supernatant was quantitated by a reverse transcriptase (RT) assay as described elsewhere (51a).

Cells were infected with 30,000 RT units of virus for 10 h at 37°C and then washed twice with growth medium. Every 2 to 3 days, the cell supernatants were removed and used for RT measurements. The cells were trypsinized, diluted 1:5 in fresh medium, and replated.

Analysis of env sequences.

For cloning the envelope genes of adapted viruses, genomic DNA from infected cells was prepared by using the MicroTurboGen genomic DNA isolation kit (Invitrogen, San Diego, Calif.). A 1.7-kb fragment containing a unique BbsI site was generated by PCR with Pfu DNA polymerase and forward ADACD4f (5′-GAAAGAGCAGAAGAGAGTGGCAATGAGAGTG-3′) and reverse ADACD4b (5′-GCCATCCAATCACACTAC-3′) primers. This fragment was gel purified and used as a primer with pSVIIIenv-ADA template DNA in a PCR mutagenesis protocol. Clones were screened for the presence of the BbsI site and sequenced by using the set of GBK96 primers described previously (7a).

Site-directed mutagenesis.

The pSVIIIenv-ADA plasmids expressing ADA envelope glycoproteins with the individual V2 loop–V1-V2 stem changes (190/197 R/S, 190/197 S/N, or 197 K) and/or with gp41 ectodomain changes (539 E, 539/540 E/R) were created by PCR mutagenesis. Complementary pairs of primers were used to introduce the following mutations into pSVIIIenv-ADA by the QuikChange protocol (Stratagene). Only one primer of each pair is given here: 190/197 R/S by primer ADARSf (5′-CCAATAGATAATGATAATACTAGGTATCGATTGATAAATTGTAGTACCTCAACCATTACACAGG-3′), 190/197 N/S by primer ADANSf (5′-GTACCAATAGATAATGATAATACTAACTATCGATTGATAAATTGTAGCACCTCAACCATTACACAGGC-3′), 197 K by primer a-Knewf (5′-CCAATAGATAATGATAATACTAGCTATCGATTGATAAATTGTAAGACCTCAACCATTACACAGG-3′), 539 E by primer ADAEf (5′-CGCAGCG TCAATAACG T TAACGGAACAGGCCAGACTATTAT TG TCTGG-3′), 539/540 E/R by primer ADAERf (5′-CGCAGCGTCAATAACGTTAACGGAACGGGCCAGACTATTATTGTCTGG-3′), and D368R by primer A-D368Rb (5′-CTGTGCATTACAATTTCAGGCCTCCCTCCTGAGGATTGATTAAAGAC-3′). To generate secreted, soluble versions of particular gp120 envelope glycoproteins, the primer ADA stop (5′-GAAGAGTGGTGCAGAGAGAAAAAAGATAAGTGGGAACGATAGGAGCTATGTTCC-3′) was used to introduce a stop codon at a position corresponding to the natural gp120-gp41 cleavage site.

env complementation assay.

Complementation of a single round of replication of the env-deficient chloramphenicol acetyltransferase (CAT)-expressing HIV-1 provirus by the various envelope glycoproteins was performed as previously described (32). Three days after infection, the target cells were lysed and CAT activity was measured as described previously (32).

gp120-CCR5 binding assay.

293T cells were transfected with 20 μg of a plasmid expressing wild-type ADA, 190/197 R/S, or 190/197 R/S D368R soluble gp120 glycoproteins and 2 μg of an HIV-1 Tat-expressing plasmid by the calcium chloride technique. One day after transfection, the cells were labeled for 24 h with [³⁵S]cysteine (100 μCi/ml) and [³⁵S]methionine (100 μCi/ml). The supernatants were harvested and cleared by centrifugation (200 × g for 5 min at 4°C). The amount of gp120 glycoprotein in the supernatants was estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis and densitometry of the crude supernatants. For CCR5 binding assays, the gp120 concentrations in the supernatants were normalized by dilution with Dulbecco modified Eagle medium (DMEM).

Trypsinized Cf2Th-synCCR5 cells were plated on six-well plates at a density of approximately 10⁶ cells/well. Fourteen hours later, the cells were washed once in phosphate-buffered saline, and 500 μl of DMEM containing the labeled gp120 proteins and 0.2% azide was added. In some experiments, sCD4 (6 μg/ml) was incubated with the gp120 solution for 1 h at room temperature prior to addition to the cells. The 2D7 anti-CCR5 monoclonal antibody (6 μg/ml) was included in the binding assay in some cases. Cf2Th cells not expressing CCR5 were used as additional controls.

The gp120-containing supernatants were incubated with the cells for 2 h at 37°C with brief agitation every 15 min. The supernatants were then removed, and the cells were washed four times with cold DMEM. The cells were then lysed in ice-cold solubilization medium [20 mM Tris-HCl (pH 7.5), 100 mM (NH₄)₂ SO₄, 1% Cymal-5, 3 mM CaCl₂, 1 mM MgCl₂, 10% glycerol, and a cocktail of protease inhibitors]. Cell lysates were transferred to Eppendorf tubes and rocked at 4°C for 25 min. The lysates were then cleared by centrifugation at 14,000 × g at 4°C for 30 min. The cleared cell lysates were precipitated with the anti-gp120 monoclonal antibody C11 and protein A-Sepharose beads. Precipitated samples were analyzed under denaturing conditions on a sodium dodecyl sulfate–10% polyacrylamide gel, which was then autoradiographed.

RESULTS

Adaptation of HIV-1 for replication in CD4-negative cells.

To study HIV-1 envelope glycoprotein determinants that are important for CD4-independent replication, a molecularly cloned HIV-1 isolate was adapted to replicate in CD4-negative, CCR5-expressing cells. The starting virus for these studies was derived by transfection of HeLa cells with the pNL4-ADA plasmid (62). The pNL4-ADA plasmid contains an infectious HIV-1 NL4-3 provirus into which the env gene derived from a primary, CCR5-using HIV-1 isolate, ADA, was cloned. The cells used for virus adaptation were Cf2Th canine thymocytes that expressed either human CD4 and CCR5 (Cf2Th-CD4/CCR5) or only human CCR5 (Cf2Th-CCR5). An initial stock of virus, herein referred to as ADA, was prepared by passaging the virus from transfected HeLa cell supernatants in Cf2Th-CD4/CCR5 cells. The ADA virus was then used to infect a 50:50 mixture of Cf2Th-CD4/CCR5 and Cf2Th-CCR5 cells. RT levels indicated that virus replication was efficient in this culture (data not shown). At day 14 of infection, virus-containing supernatants were used to infect either a pure population of Cf2Th-CCR5 cells (culture 1) or a 25:75 mixture of Cf2Th-CD4/CCR5 and Cf2Th-CCR5 cells (culture 2). Although levels of viral RT in the supernatants of culture 1 were barely above background for the first 12 days of infection, these values rose progressively from days 13 to 20 (data not shown). The virus stock from these culture supernatants will be referred to as ADA-P1. In contrast to the lag in replication observed in culture 1, virus replication in culture 2 peaked by day 12 after infection. Supernatants from day 11 after infection were used to infect Cf2Th-CCR5 cells, and efficient virus replication occurred in this culture (data not shown). The virus derived from these Cf2Th-CCR5 cell supernatants is herein referred to as ADA-P2.

To determine whether ADA-P1 and ADA-P2 were capable of CD4-independent replication, Cf2Th-CCR5 and Cf2Th-CD4/CCR5 cells were incubated with these viruses, as well as with the parental ADA virus. Both ADA-P1 and ADA-P2 replicated efficiently in the Cf2Th-CCR5 cells, whereas the ADA virus did not (Fig. 1A). In Cf2Th-CD4/CCR5 cells, all three viruses replicated, with ADA-P1 and ADA-P2 exhibiting faster replication kinetics than the ADA virus (Fig. 1B). Slightly higher levels of virus production and faster replication kinetics of the ADA-P1 and ADA-P2 viruses were observed for the Cf2Th-CD4/CCR5 cells than for the Cf2Th-CCR5 cells. Apparently, the ADA-P1 and ADA-P2 viruses can replicate efficiently in cells lacking human CD4.

FIG. 1.

Replication of adapted viruses in CD4-negative and CD4-positive cells expressing CCR5. Virus production in the supernatants of Cf2Th-CCR5 (A) or Cf2Th-CD4/CCR5 (B) cells following infection by the ADA, ADA-P1, or ADA-P2 virus is shown. The results of a single experiment are shown. The experiment was repeated with similar results.

Analysis of env sequences of CD4-independent viruses.

The env sequences of the ADA-P1 and ADA-P2 viruses were analyzed by PCR with the genomic DNA of the Cf2Th-CCR5 cultures on day 10 after virus infection (Fig. 1A). Three env genes from ADA-P1 and four env genes from ADA-P2 were sequenced in their entirety. Compared with the parental ADA envelope glycoproteins, the predicted envelope glycoprotein sequences of the ADA-P1 and ADA-P2 viruses exhibited changes in two regions (Fig. 2). All seven envelope glycoproteins had alterations in a region encompassing the C-terminal end of the V2 variable loop and the V1-V2 stem. In every instance, these changes involved a loss of one of the two N-linked glycosylation sites in this region (40). When the more N-terminal site at asparagine 188 was lost, the C-terminal site was shifted two residues. Five of the envelope glycoproteins also had changes in a gp41 region just C-terminal to the fusion peptide.

FIG. 2.

Envelope glycoprotein changes observed for the ADA-P1 and ADA-P2 viruses. The organization of the HIV-1 envelope glycoproteins is shown, with the signal peptide (S) and transmembrane region (T) indicated. The gp120 regions conserved among different viral strains are black, and the variable regions are white and labeled. The sequences shown are of the wild-type (w.t.) ADA, ADA-P1, and ADA-P2 envelope glycoproteins in the two regions in which differences were observed. In the wild-type ADA sequence, the relevant residues are underlined and numbered according to the prototypic HXBc2 sequence (37). In the characterized ADA-P1 and ADA-P2 clones, the altered residue is in boldface. The observed frequency of each sequence is indicated at the right. Sites of N-linked glycosylation are indicated by circles atop vertical lines.

The contribution of the observed envelope glycoprotein changes to virus infection of CD4-negative and CD4-positive cells was examined. For this purpose, we constructed plasmids expressing the ADA envelope glycoproteins altered at residues 190 and 197 in the V2 loop–V1-V2 stem. Mutant envelope glycoproteins with each of the three changes (190/197 R/S, 190/197 N/S, and 197 K) observed for the ADA-P1 and ADA-P2 viruses were constructed. Additional mutants in which the glutamic acid change at position 539 in gp41 and the glutamic acid and arginine substitutions at positions 539 and 540 were introduced along with the V2 loop–V1-V2 stem changes were constructed. These gp41 changes were also introduced into the wild-type ADA envelope glycoproteins. The effects of the V2 loop–V1-V2 stem and gp41 changes on HIV-1 infection of cells were examined by an env complementation assay (32). In this assay, an env-defective HIV-1 provirus encoding CAT is transfected into HeLa cells along with a plasmid encoding the envelope glycoproteins of interest. The recombinant viruses produced are incubated with target cells, and the level of CAT activity in the target cells indicates the efficiency with which a single round of infection has occurred. Figure 3A shows the CAT activity in Cf2Th-CCR5 cells incubated with viruses containing the wild-type or mutant ADA envelope glycoproteins. No CAT activity above background was detected in Cf2Th-CCR5 cells incubated with viruses with the wild-type ADA envelope glycoproteins. Infection of the Cf2Th-CCR5 cells was efficiently mediated by the 190/197 R/S envelope glycoproteins. Lower but detectable levels of infection of these cells were observed for viruses containing the 190/197 N/S and 197 K substitutions. These results indicate that changes in the V2 loop–V1-V2 stem region are sufficient to enhance the ability of HIV-1 to infect CD4-negative cells. The V2 loop–V1-V2 stem mutants were unable to infect the CCR5-negative, parental Cf2Th cells (data not shown). Furthermore, the 190/197 R/S mutant was unable to infect cells expressing CXCR4, STRL33, gpr15, or apj, even when the cells expressed CD4 (data not shown). Thus, infection by the V2 loop–V1-V2 stem mutants remained dependent on CCR5.

FIG. 3.

Influence of envelope glycoprotein changes on infection of CD4-negative and CD4-positive cells. An env-deficient, CAT-expressing HIV-1 isolate was complemented by the wild-type (w.t.) ADA or mutant envelope glycoproteins. The envelope glycoproteins exhibit either the VQ, EQ, or ER sequence at gp41 positions 539 and 540, as indicated. CAT activity observed for equivalent amounts of cell lysates derived from Cf2Th-CCR5 (A) or Cf2Th-CD4/CCR5 (B) cells incubated with the recombinant virions is shown. The mean values derived from three independent experiments are shown.

It is possible that CD4-independent infection involves another target cell moiety, present on a variety of cells, that interacts with the HIV-1 gp120 glycoprotein in a manner similar to that of CD4. To investigate this possibility, infection by the 190/197 R/S mutant was compared with that of a virus containing an additional alteration in gp120 residue 368. The substitution of arginine for aspartic acid 368, which contacts CD4 (39), reduces CD4-binding affinity more than 100-fold (50, 51). This change in residue 368 reduced the efficiency with which Cf2Th-CD4/CCR5 cells were infected by the 190/197 R/S mutant but had no negative effect on infection of Cf2Th-CCR5 cells by this V2 loop–V1-V2 stem variant (data not shown). This result suggests that, if a surrogate for CD4 on the target cell promotes CD4-independent infection, it does not bind gp120 in the same manner as CD4.

The wild-type envelope glycoproteins and the V2 loop–V1-V2 stem mutants were able to mediate entry into Cf2Th-CD4/CCR5 cells, although two of the mutants were slightly less efficient than the wild-type glycoproteins (Fig. 3B). The V2 loop–V1-V2 stem mutants infected the Cf2Th-CD4/CCR5 cells more efficiently than they infected the Cf2Th-CCR5 cells. The gp41 substitutions tested resulted in a lowered efficiency of infection of both CD4-negative and CD4-positive Cf2Th cells (Fig. 3).

We investigated whether the 190/197 R/S changes would confer the ability to infect CD4-negative cells on HIV-1 envelope glycoproteins other than that of the ADA isolate. These changes were introduced into the R5 YU2 envelope glycoproteins and the ADA (MMM) glycoproteins. The latter gp120 glycoprotein is a CXCR4-using chimera in which the V3 loop and C4 region of the ADA gp120 have been replaced by the corresponding segments of the MN gp120. In neither instance did the 190/197 R/S change allow CD4-independent infection of target cells expressing the appropriate coreceptor (data not shown).

CCR5 binding of gp120 glycoproteins from CD4-independent viruses.

The ability of HIV-1 gp120 to bind CCR5 is dependent on prior CD4 binding (68, 72). To examine whether the acquisition of CD4 independence influenced CCR5 binding, the ability of wild-type ADA, 190/197 R/S, and 190/197 R/S D368R gp120 glycoproteins to bind to CCR5-expressing Cf2Th cells (Cf2Th-synCCR5) cells was examined. The latter glycoprotein contains the substitution of arginine at position 368 that compromises CD4 binding (50, 51). In some experiments, the gp120 glycoproteins were incubated with saturating concentrations of sCD4. The ADA gp120 glycoprotein bound very weakly to Cf2Th-synCCR5 cells, and incubation with sCD4 enhanced this binding (Fig. 4). In the absence of sCD4, both the 190/197 R/S and 190/197 R/S D368R gp120 glycoproteins bound to Cf2Th-synCCR5 cells better than the wild-type glycoprotein did. Incubation with sCD4 enhanced the binding of the 190/197 R/S glycoprotein but not that of the 190/197 R/S D368R glycoprotein. The addition of the 2D7 monoclonal antibody directed against CCR5 reduced the binding of all of the gp120 variants to Cf2Th-synCCR5 cells (data not shown). None of the gp120 glycoproteins bound to the CCR5-negative, parental Cf2Th cells (Fig. 4). The latter two results indicate that gp120 binding in this assay is specific for CCR5. The results suggest that one of the mechanisms underlying CD4-independent infection is the ability of the gp120 envelope glycoprotein to bind CCR5 directly.

FIG. 4.

CCR5-binding ability of gp120 glycoproteins from the adapted virus. The binding of radiolabeled, soluble gp120 glycoproteins to Cf2Th-synCCR5 cells, in the absence or presence of 6 μg of sCD4 per ml, is shown. The gp120 glycoproteins were also incubated with Cf2Th cells not expressing CCR5, in the presence of 6 μg of sCD4 per ml. Equivalent levels of all three gp120 glycoproteins were incubated with the cells, as described in Materials and Methods. w.t., wild type.

DISCUSSION

CD4-independent HIV-2 and SIV isolates have been reported previously (11, 23, 24, 57), and a CD4-independent HIV-1 isolate that utilizes the CXCR4 chemokine receptor has been previously obtained by in vitro passage (21). Here we report the derivation, by in vitro passage, of a CD4-independent, CCR5-using HIV-1 virus from a primary isolate. The passaged viruses, ADA-P1 and ADA-P2, replicated efficiently in CCR5-expressing, CD4-negative canine cells. In canine cells expressing both CCR5 and CD4, ADA-P1 and ADA-P2 replication was extremely efficient compared with that of the parental ADA virus.

Two sets of changes were consistently observed in the envelope glycoproteins of the uncloned ADA-P1 and ADA-P2 viruses, compared with those of the parental virus. One set of changes involved a 10-residue sequence extending from the C-terminal portion of the gp120 V2 loop into the V1-V2 stem (residues 188 to 197 in the prototypic HXBc2 HIV-1 strain [37]). The second set of changes involved a well-conserved pair of adjacent residues (numbers 539 and 540 in the prototypic HXBc2 HIV-1 sequence) in the gp41 ectodomain. The functional importance of these changes was assessed in assays in which cloned envelope glycoproteins were used to complement an env-defective HIV-1 provirus in 293T cells (32). In these assays, only the V2 loop–V1-V2 stem changes could be shown to contribute to CD4-independent entry into canine cells expressing CCR5. These changes were necessary and sufficient for CD4-independent virus infection and for binding of the gp120 glycoprotein to CCR5 in the absence of CD4.

The location of changes associated with CD4 independence in V2 loop–V1-V2 stem sequences is interesting in light of previous studies suggesting that a repositioning of the V2 loop represents a major gp120 conformational response to CD4 binding (64, 74). The conserved V1-V2 stem contacts CD4 and contributes strands to the bridging sheet, which is implicated in CCR5 binding (39, 52). The V2 loop masks gp120 epitopes near the chemokine receptor-binding site (52, 64, 74). Apparently, the observed changes in V2 loop–V1-V2 stem sequences allow the ADA gp120 glycoprotein to achieve a conformation competent for CCR5 binding without requiring contact with CD4. The contribution of the V1-V2 stem to the induction of chemokine receptor binding applies to T-cell-tropic HIV-1 isolates as well, because changes in the V1-V2 stem, along with V3 loop changes, were shown to contribute to the acquisition of CD4 independence by a CXCR4-using HIV-1 isolate (21). The V2 loop–V1-V2 stem changes observed in our study, however, were apparently quite specific for the ADA envelope glycoprotein and for CCR5, as they did not allow CD4 independence in the context of other envelope glycoproteins or coreceptors.

Two potential mechanisms, which are not mutually exclusive, could explain the observed CD4-independent CCR5 binding. The V2 loop–V1-V2 stem changes associated with CD4 independence might alter the conformation of the V2 loop, thereby better exposing the chemokine receptor-binding site. The gp120 changes could also allow novel initial contacts with CCR5 that promote conformational changes required for high-affinity interaction. The available structural information on gp120 does not allow an unambiguous choice between these possible mechanisms. The V2 loop–V1-V2 stem changes in the CD4-independent viruses involved either loss or movement of two N-linked glycosylation sites 9 residues apart. In most ADA-P1 and ADA-P2 clones, a basic residue appears at or near a lost glycosylation site. Asparagine 188, one of the altered sites of N-linked glycosylation, is within the V2 loop, which is deleted from the crystallized gp120 core (39). The peptide-proximal carbohydrate residue at asparagine 197, in the V1-V2 stem, faces away from the target cell (39, 75) and, therefore, would be expected to influence chemokine receptor binding only indirectly. In addition to possible effects on conformation, the observed changes alter the electrostatic potential of the gp120 surface. In mammalian cells, complex carbohydrates are added to asparagine 188 and asparagine 197 (40), and therefore, loss or movement of these glycosylation sites would be expected to remove a significant amount of sialic acid from this region. This, along with the observed replacement of basic residues, would increase the positive electrostatic potential of this region. The gp120 region thought to interact with CCR5 is relatively basic, a property thought to facilitate electrostatic interactions with the acidic CCR5 amino terminus (25, 39, 52). Removal or repositioning of sialic acid-containing sugars in the V2 loop and/or the V1-V2 stem of gp120 could increase this basicity and encourage interactions with CCR5 even in the absence of CD4.

The ADA-P1 and ADA-P2 viruses replicated more efficiently in CD4-positive cells than in CD4-negative cells. This observation suggests that not all of the functions of CD4 in HIV-1 infection are rendered redundant by the adaptation-associated changes. For example, even for a virus capable of direct CCR5 interaction, CD4 binding would still be expected to facilitate virion attachment and orientation with respect to the target cell membrane.

We were not able to demonstrate a role in CD4 independence for the changes observed in gp41 ectodomain residues valine 539 and glutamine 540. The otherwise well-conserved nature of these gp41 residues (38) and the frequency with which nonconservative substitutions occurred upon viral passage in CD4-negative cells suggest that these changes make some positive contribution to viral adaptation. The phenotypes of these gp41 changes may be subtle or may be manifest only in contexts other than those examined in our assays.

The ADA-P1 and ADA-P2 viruses replicated more efficiently than the parental ADA virus in CD4-positive, CCR5-positive cells (Fig. 1). However, most of the envelope glycoprotein changes responsible for CD4 independence were slightly detrimental to virus entry into CD4-positive cells (Fig. 3). Viral adaptations outside of env likely account for the faster replication kinetics of the ADA-P1 and ADA-P2 viruses in CD4-positive cultures.

The availability of an HIV-1 gp120 glycoprotein that binds CCR5 efficiently in the absence of CD4 may expedite progress in studying the interactions of primate immunodeficiency viruses with their receptors.