The authors have declared that no competing interests exist.
Conceived and designed the experiments: SHX CB. Performed the experiments: CB DB JT LA SGR. Analyzed the data: SHX CW YM JS. Wrote the paper: SHX JS.
The V1 and V2 variable regions of the primate immunodeficiency viruses contribute to the trimer association domain of the gp120 exterior envelope glycoprotein. A pair of V2 cysteine residues at 183 and 191 (“twin cysteines”) is present in several simian immunodeficiency viruses, human immunodeficiency virus type 2 (HIV-2) and some SIVcpz lineages, but not in HIV-1. To examine the role of this potentially disulfide-bonded twin-cysteine motif, the cysteine residues in the SIVmac239 envelope glycoproteins were individually and pairwise substituted by alanine residues. All of the twin-cysteine mutants exhibited decreases in gp120 association with the Env trimer, membrane-fusing activity, and ability to support virus entry. Thus, the twin-cysteine motif plays a role in Env trimer stabilization in SIV and may do so in HIV-2 and some SIVcpz as well. This implies that HIV-1 lost the twin-cysteines, and may have relatively unstable Env trimers compared to SIV and HIV-2.
It has been well established that human immunodeficiency viruses (HIV) are derived from simian immunodeficiency viruses (SIV) through cross-species transmission
The PLV, which include HIV-1, HIV-2 and SIV, are enveloped retroviruses. The trimeric envelope glycoprotein (Env) spikes on the virion surface are the only viral molecules making direct contacts with host cell receptors (CD4 and a chemokine receptor like CCR5). PLV Env evolution is driven by requirements to mediate host cell entry and to evade host-generated neutralizing antibodies. HIV and SIV Envs evade host immunity by employing mechanisms such as rapid alteration of surface loops, glycan shielding and presentation of multiple conformers, which may act as decoys
We are particularly interested in the participation of the V2 region in the trimer association domain (TAD) because so little is known about its structure and function. However, the V2 region is immunogenic, and in some cases serves as a target for broadly cross-reactive neutralizing antibodies
Alignment of the PLV V2 sequences revealed the presence of two conserved cysteine residues in HIV-2 and SIV strains (
The Phylogenetic tree (unrooted, left panel) was constructed based on whole genome nucleotide sequences, using a neighbor-joining method
The 293T cells used for SIV envelope expression and recombined viruses production were grown on Dulbecco’s modified Eagle medium (DMEM) containing 10% fetal bovine serum and 100 µg/ml of penicillin/streptomycin at 37°C and 5% CO2. The cell lines based on Cf2Th-CD4 cells were grown on complete DMEM with 150 µg/ml of hygromycin medium but supplemented with additional antibiotics, 50 µg/ml G418 for Cf2Th-CD4/CCR5 cells, 300 ug/ml Zeocin for CfTh2-CD4/CXCR4 cells. The TZM-bl reporter cell line (NIH AIDS Research and Reagent Program) was used for viral infection experiments as it expresses CD4 and CCR5, and also contains both beta-galactosidase and luciferase reporter genes; the TZM-bl cell line was grown in normal complete DMEM medium.
The twin-cysteine substitution mutants (183A/C and 191C/A, 183C/A+191C/A) as well as other related mutants (192Y/A and 190R/E) were made in the pcDNA3.1(+) vector using site-directed mutagenesis (QuikChange II XL kit from Stratagene). All mutagenesis primers were synthesized by Integrated DNA Technologies (IDT). DNA sequencing was used to verify all mutations generated by the designed primers.
Gel electrophoresis was used to check the envelope glycoprotein expression, processing and stability, and also used to separate gp160 and gp120 Env glycoproteins. This method is based on radioactive (35S methionine/cysteine) metabolic labeling of the Env-expressing cells and immunoprecipitation
The transfected 293T cells were cultured in methionine/cysteine-free medium for 30 minutes followed by pulse labeling for 1 hour with 300 uCi of [35S]-methionine/cysteine (>1000 Ci/mmol; NEN). The labeling medium was removed and the cells were washed with complete medium and chased in complete medium for 2, 4, and 8 hours. At the appropriate times, medium was collected, clarified by micro-centrifugation, and adjusted to 1X lysis buffer containing protease inhibitors. The cells were lysed in 1X lysis buffer containing protease inhibitors and clarified by micro-centrifugation. Lysates were pre-cleared with normal monkey serum and viral proteins were immunoprecipitated with SIVmac251-infected monkey serum, separated by SDS-PAGE, and analyzed by phosphoimaging using the Discovery Series Quantity One software (Bio-Rad).
The wild-type or a mutant envelope plasmid was transfected into 293T cells using the Effectene transfection reagent (Qiagen). After 2 days, the transfected cells were metabolically labeled using 35S-methionine/cysteine (1000 Ci/mmol, Perkin-Elmer) overnight in DMEM medium free of methionine and cysteine. The medium was collected and the cells were lysed with buffer containing a lower NP-40 concentration. After centrifugation to remove the cell debris, the supernatant and the collected medium were mixed for the ligand binding experiments. The CD4 binding was carried out by using human Ig-tagged CD4 protein (CD4-IgG2 from the NIH AIDS Research and Reagent Program) with the radiolabeled viral envelope glycoproteins. The bound envelope proteins were analyzed by SDS-PAGE and exposure to X-ray film. The assessment of gp120 binding to the co-receptor CCR5 was done by using CF2Th-CCR5 cells in the presence of sCD4 (10 µg/ml). The CCR5-expressing cells were incubated with the radiolabeled gp120 envelope glycoproteins in the supernatants of Env-expressing cells. The CCR5-expressing cells were washed, lysed, and precipitated with HIV-1 positive pertinent sera and Protein-A beads. The immunoprecipitates were applied to an SDS-PAGE gel, and then visualized and quantified after exposing to X-ray film or using a phosphoimager.
The fusion abilities of all the Env mutants were measured using TZM-bl target cells that express the HIV/SIV receptors CD4 and coreceptor CCR5 and CXCR4. The 293T cells in 6-well plates were co-transfected with the Env-expressing plasmids and a Tat-expressing plasmid using Fugene6
The viral infection analysis was carried out using TZM-bl as target cells. The wild-type and twin-cysteine or other Env mutants were pseudotyped on viruses using the viral genomic backbone plasmid pNL-4-3_GFPΔEnv in transfected 293T cells. The viral titers were measured by RT activity using liquid scintillation to detect incorporation of H3 into newly synthesized DNA; input virus for infections was normalized as CPM reflecting functional reverse transcriptase in virus stocks. The infectivity of the wild-type and mutant viruses was determined by incubating the recombinant virions with TZM-bl target cells for 48 hours at 37°C. The GFP intensity was measured using flow cytometry.
To investigate the role of the twin cysteines during viral infection, we altered them in the prototypic SIV strain SIVmac239 and tested the effect of these changes on virus infectivity. The twin cysteines were changed singly (183C/A, 191C/A) or together (183C/A+191C/A). Additionally, we altered a conserved tyrosine (192Y) located adjacent to the second cysteine (192Y/A) (
The relative infectivity of the SIVmac239 V2 mutants is shown in
Viruses pseudotyped with the wild-type (wt) and mutant SIVmac239 Envs were made with a GFP-expressing vector and used to infect TZM-bl target cells. The infectivity was evaluated by measuring the fluorescence intensity of GFP by flow cytometry.
We tested the ability of the SIVmac239 mutants to mediate cell-cell fusion. The twin-cysteine mutants were significantly reduced in their cell-cell fusion ability, along with the 192Y/A mutant (
The cell-cell fusion assay was performed by measuring the fluorescence intensity of Tat-inducible luciferase resulting from fusion of the SIVmac239 Env-expressing cells with target TZM-bl cells.
To investigate the mechanism of the observed decreases in SIVmac239 Env function, we examined the binding of radiolabeled gp120 to the receptors, CD4 and CCR5. The CCR5 binding experiments were carried out in the presence of soluble CD4. Neither CD4 nor CCR5 binding were affected significantly by the introduced changes (
The assay measuring gp120 binding to human CD4 and human CCR5 utilized 35S-radiolabeled gp120 precipitated with CD4-Ig or incubated with CCR5-expressing cells in the presence of soluble CD4.
Because of the association of the V2 region with the gp120 TAD, we hypothesized that the twin cysteines may contribute to stabilization of the Env trimer. We examined Env trimer stability by determining the amount of the Env glycoprotein precursor (gp160) and mature gp120 glycoprotein in the cells, as well as shed gp120 in the medium of Env-expressing cells. As shown in
Cells expressing the SIVmac239 Env variants were labeled with 35S-(methioine and cysteine) either continuously overnight (
To confirm the above results, we conducted a pulse-chase experiment on cells expressing the wild-type and mutant SIVmac239 Envs (
Here we show that alteration of the twin-cysteine motif disrupts the stable and non-covalent association of SIVmac239 gp120 with the Env trimer. Presumably as a result of the instability of the mutant Env trimers, these Env mutants exhibit very poor function in cell-cell fusion or virus entry. We suggest that the twin cysteines may form a disulfide-bond to stabilize the Env trimer. This disulfide bond could be formed either within the gp120 subunit or between adjacent subunits. Although there is currently no evidence for inter-protomer disulfide bonds in the SIVmac239 Env trimer, the close association of the V2 regions in the gp120 TAD raises this possibility. Future studies will explore the formation and biological relevance of disulfide bonds between Env protomers. The adjacent tyrosine 192 of the twin-cysteine motif is markedly conserved in SIV, HIV-2 and HIV-1, and change of this residue also interrupts the trimer association. It is indicated that this tyrosine 192 may also play an important role in the Env trimer stabilization.
The twin cysteines in the V2 regions of the gp120 trimer association domain are very conserved in SIV and HIV-2, but are not present in any HIV-1 strains. The majority of SIVcpz strains lack these two cysteine residues, whereas some strains have two cysteine residues in the same vicinity. The twin cysteines appear to have evolved in several SIV lineages, but were minimally retained in SIVcpz and then lost completely in HIV-1. Of interest, during zoonotic transmission from monkeys and establishment in hominoids, PLVs became progressively more dependent on CD4. This increased dependence on CD4 was accompanied by changes in the Phe 43 cavity and inner domain of gp120
Modulation of Env trimer stability may have implications for viral evasion of the host immune response and pathogenicity. The added TAD stability that SIV and HIV-2 gain from structural elements like the twin cysteines may influence the elicitation and efficiency of neutralizing antibodies. Although SIV and HIV-2 elicit a host immune response in the form of neutralizing antibodies, both viruses persist, achieving a balance between virus infection and the host immune reaction. As a result, SIV does not typically cause serious disease in monkey hosts. Similarly, HIV-2 infection in humans is much less likely to progress to full-blown AIDS than HIV-1 infection. As CD4-dependence is associated with greater resistance to neutralization antibodies, HIV-1 may have evolved greater CD4 dependence to escape host immune responses. Poorer host control of HIV-1 is consistent with its pathogenicity. Investigating the functional phenotypes associated with Env diversity in the HIV-1 and HIV-2/SIV lineages should lead to a better understanding of PLV pathogenesis in primate species.
We thank Ms. Yvette McLaughlin for help with manuscript preparation. We thank Dr. Dan Barouch and Dr. Michael Seaman at Harvard Medical School for providing anti-SIV sera.