Kinetic studies of HIV-1 and HIV-2 envelope glycoprotein-mediated fusion
© Gallo et al. 2006
Received: 13 July 2006
Accepted: 04 December 2006
Published: 04 December 2006
Skip to main content
© Gallo et al. 2006
Received: 13 July 2006
Accepted: 04 December 2006
Published: 04 December 2006
HIV envelope glycoprotein (Env)-mediated fusion is driven by the concerted coalescence of the HIV gp41 N-helical and C-helical regions, which results in the formation of 6 helix bundles. Kinetics of HIV Env-mediated fusion is an important determinant of sensitivity to entry inhibitors and antibodies. However, the parameters that govern the HIV Env fusion cascade have yet to be fully elucidated. We address this issue by comparing the kinetics HIV-1IIIB Env with those mediated by HIV-2 from two strains with different affinities for CD4 and CXCR4.
HIV-1 and HIV-2 Env-mediated cell fusion occurred with half times of about 60 and 30 min, respectively. Binding experiments of soluble HIV gp120 proteins to CD4 and co-receptor did not correlate with the differences in kinetics of fusion mediated by the three different HIV Envs. However, escape from inhibition by reagents that block gp120-CD4 binding, CD4-induced CXCR4 binding and 6-helix bundle formation, respectively, indicated large difference between HIV-1 and HIV-2 envelope glycoproteins in their CD4-induced rates of engagement with CXCR4.
The HIV-2 Env proteins studied here exhibited a significantly reduced window of time between the engagement of gp120 with CD4 and exposure of the CXCR4 binding site on gp120 as compared with HIV-1IIIB Env. The efficiency with which HIV-2 Env undergoes this CD4-induced conformational change is the major cause of the relatively rapid rate of HIV-2 Env mediated-fusion.
The origins of Human Immunodeficiency Virus (HIV) can be traced to zoonotic transmissions of Simian Immunodeficiency Virus (SIV) to humans from at least two different kinds of non-human primates : HIV-1, which came from chimpanzees, and HIV-2, which came from sooty mangabeys. While similar in many ways, there are important differences between HIV-1 and HIV-2 that provide insights into virus evolution, tropism and pathogenesis . Major differences include reduced pathogenicity of HIV-2 relative to HIV-1, enhanced immune control of HIV-2 infection and often some degree of CD4-independence. Despite considerable sequence and phenotypic differences between HIV-1 and 2 envelopes, structurally they are quite similar. Both membrane-anchored proteins eventually form the 6-helix bundles from the N-terminal and C-terminal regions of the ectodomain , which is common to many viral and cellular fusion proteins and which seems to drive fusion . HIV-1IIIB gp41 helical regions can form more stable 6-helix bundles than HIV-2SBL gp41 helical regions [3, 5]; however HIV-2 fusion occurs at a lower threshold temperature (25°C), does not require Ca2+ in the medium, is insensitive to treatment of target cells with cytochalasin B , and is not affected by target membrane glycosphingolipid composition .
In order to elucidate mechanisms of HIV envelope glycoprotein-mediated fusion we have kinetically resolved steps in the pathway of HIV-1 membrane fusion . To gain a better understanding of the molecular mechanisms underlying these steps, we compared kinetic parameters of HIV-1IIIB with two strains of HIV-2. We found a significant difference in fusion kinetics, which appears to be related to the CD4-induced rate of engagement of HIV gp120 with its coreceptor. Since the CD4-induced binding of gp120 proteins to CXCR4 is not very different between the different strains, we surmise that in the intact Env other regions (e.g. the cytoplasmic tail) may have a profound influence on the conformational changes in the surface-exposed portions of the envelope glycoproteins.
Inhibition of HIV-1 and HIV-2 Env-mediated fusion. AMD3100 or Leu3A were added to cocultures of HIV-expressing cells and target cells at the time of co-culture and fusion was measured as described in Methods. IC50 values were derived from dose-response by fitting the data to a hyperbolic decay function.
IC50 (μg/ml) for fusion at 37°C
0.53 ± 0.07
0.3 ± 0.1
0.03 ± 0.02
0.14 ± 0.02
0.09 ± 0.01
The current model of HIV viral entry involves the binding of the trimeric viral Env glycoprotein gp120/gp41 to cell surface receptor CD4, which triggers conformational changes in the envelope proteins. Gp120 is then re-positioned allowing gp41 to undergo conformational changes that result in the formation of the gp41 "pre-hairpin" [12–14]. Upon engagement with chemokine co-receptors CXCR4 or CCR5 [15, 16], the C-terminal heptad repeat region and the leucine/isoleucine zipper region form the thermostable 6-helix bundle, which drives membrane merger and eventual fusion . In the case of HIV-1IIIB it appears that the pre-hairpin conformation is quite a long lasting state [11, 18]. Previously, we had attributed the relatively slow kinetics of HIV-1 Env-mediated fusion to the stochastic nature of HIV Env triggering giving rise to a relatively low probability of 6-helix bundle formation and fusion . We had invoked the "harpoon" model  according to which the HIV-1 gp120 is searching to engage its co-receptor following initial conformational changes induced by CD4 binding. We had shown that HIV-1 gp41 6-helix bundle formation occurs rapidly after the engagement of gp120 by CXCR4 in the HIV-1 Env-mediated fusion process . We reasoned that higher gp120-coreceptor affinities may result in more rapid gp120-coreceptor binding leading to enhanced rates of 6-helix formation and fusion. This hypothesis has been born out in studies showing that envelope:coreceptor affinity correlates with fusion kinetics [9, 10].
In this study we show that fusion mediated by the HIV-1IIIB Env exhibits slower kinetics as compared to that of two different HIV-2 Envs (Figure 1). Since we observed little CD4-induced binding of gp120 from HIV-1IIIB and HIV-2SBL to CXCR4 under steady-state condition (figure 2), and similar AMD3100 susceptibilities (Table 1) that are in contrast to efficient CXCR4 binding and markedly reduced AMD3100 susceptibility for HIV-2ROD, the difference in fusion rates mediated by these HIV-1 and HIV-2 Envs are not likely due to differences in gp120-coreceptor affinity per se. However, the time of addition experiments (figure 3) indicate that the major difference in kinetics between HIV-1IIIB and HIV-2SBL lays in the efficiency of conformational changes that occur after CD4 is bound and that result in the formation of the coreceptor-binding site. In the case of HIV-1, exposure of the coreceptor binding site has not yet occurred immediately after the stage that Leu3A can block. It has been shown that binding of a single CD4 molecule to an envelope trimer leads to conformational changes in all three gp120 molecules . This process may take a relatively long time in the case of HIV-1 Env. By contrast, the coreceptor binding site is rapidly exposed following CD4 binding in the case of HIV-2 Env consistent with the observation that HIV-2 strains (in contrast to HIV-1) are often able to infect cells CD4-independently. Since CD4-induced binding of soluble gp120 molecules to CXCR4 was not very different between the HIV-1IIIB and HIV-2SBL strains, we surmise that in the intact Env other regions may have a profound influence on this conformational change.
Another region of HIV-1 gp41 that has profound effects on fusion rates is the cytoplasmic tail . HIV-1, HIV-2 and SIV CTs are remarkably long and contain domains that likely interact with host cell components, such as calmodulin [21, 22], α-catenin , p115-RhoGEF , Prenylated Rab acceptor protein , or AP-1 clathrin adaptor proteins . Three alpha helical "lentivirus lytic peptide" domains (LLP-1, LLP-2 and LLP-3) highly conserved in HIV-1 have been implicated in interacting with the cytoplasmic leaflet of plasma membrane, decreasing bilayer stability, altering membrane ionic permeability, and mediating cell killing [26–28]. Poorly defined regions of gp41 have also been implicated in interacting with viral matrix proteins during virion assembly [29, 30]. This interaction also modulates Env function in that gp41 is more stably associated with immature rather than mature viral particles , and cleavage of the p55 Gag precursor protein by the viral protease is required to generate Envs with maximal fusogenicity [31, 32].
Truncations of gp41 proximal to the most N-terminal alpha helix, LLP-2, produced a significant increase in the rate of HIV-1 Env-mediated cell fusion [20, 33]. These effects were not seen with a truncation distal to this domain and before LLP1 . These results were observed for X4-, R5-, and dual-tropic Envs on CXCR4- and CCR5-expressing target cells. Sequence comparisons of gp41 derived from HIV-1, HIV-2 and SIV indicate high homology between the three viral Envs in the LPP-1 region but not in the LLP-2 region of the cytoplasmic tail (the sequence comparisons can be obtained from Brian Foley upon request). In the case of certain HIV-1 Env strains it has been shown that CT truncations prior to LLP-2 profoundly affect the exposure of CD4-induced epitopes on gp120 [20, 34], indicating that "inside-out" signalling changes the conformation in the Env ectodomain. Although we have performed this analysis by comparing only one HIV-1 and two HIV-2 strains we believe that the methodology developed here can be expanded to include comparisons between various strains of HIV-1 and HIV-2. Future studies that involve domain swapping between HIV-1 and HIV-2 gp41 CT's will determine whether our LLP-2 hypothesis turns out to be correct.
We find that the differences in fusion mediated by HIV-1 and HIV-2 Env are due to the rates by which the Envs assume a conformation that expose the binding sites on gp120 to the CXCR4 following engagement with CD4. In spite of the fact that CD4-induced binding to CXCR4 of isolated gp120 derived from HIV-1IIIB and HIV-2SBL were similar, their rate of binding to the co-receptor was markedly different in the context of the complete Envs. We speculate that sequences in the cytoplasmic tail of the Env may have a profound influence upon the rate by which the extracytoplasmic portion of gp120 assumes its CXCR4-engageable conformation.
QT6, 293T and HeLa cell lines were cultured in DMEM supplemented with 10% fetal calf serum, 60 mg/ml of penicillin and 100 mg/ml streptomycin. Sup-T1 cells (Non-Hodgkin's T-cell lymphoma cell line) were cultured in RPMI supplemented with 10% serum, 60 mg/ml of penicillin and 100 mg/ml streptomycin.
The fusion inhibitor SIV C34  was dissolved in PBS at a stock concentration of 500 uM. The CXCR4 antagonist AMD3100 , a kind gift from Anormed, Inc. (Langley, Canada), was dissolved in PBS at a stock concentration of 1 μg/ul. Leu3A, an antibody against the gp120 binding site of CD4 and a highly effective fusion inhibitor  (BD Biosciences, San Jose, CA), was dissolved in 0.1% Azide at 25 ug/ml. All inhibitors were stored at 4°C.
The 3' half proviral clone of HIV-2SBL (KF-3; kindly provided by G. Franchini) was used as a template for PCR amplification of the envelope gene using 5' CACCATGAGTGGTAAAATTCAGCTGC 3' and 5' CTCCTTGCTGATATCTCTGTCCCTCA 3' oligonucleotides. The PCR product was cloned into the pcDNA3.1 D/V5-His-TOPO expression vector (Invitrogen, Carlsbad, CA) to generate an sbl/isy gp160 expression construct. A stop codon was introduced at the gp120/gp41 cleavage junction of the sbl/isy gp160 expression construct using the 'Quikchange' site directed mutagenesis kit (Stratagene, La Jolla, CA) and 5' GGGAGACATAAGAGATGAAAGCTTGTGCTAGGGTTC 3' and 5' GAACCCTAGCACAAGCTTTCATCTCTTATGTCTCCC 3' oligonucleotides to generate a gp120 expression construct. These oligonucleotides also introduce an HindIII restriction enzyme site (for screening purposes) down stream of the stop codon. HIV-2ROD/A and HIV-1IIIB gp120 expression vectors have been described previously [37, 38].
HIV-1 and HIV-2 Env proteins were expressed in HeLa cells using the following vaccinia recombinants;vPE16, for the IIIB strain of HIV-1 , vvROD, for the ROD strain of HIV-2 , and vSC50 for the SBL/ISY strain of HIV-2 . These recombinants were incubated with cells overnight at a ratio of 10:1 infectious virions to cells. HeLa cells were plated on 12 or 24 well plates overnight before infection (Costar, Cambridge, MA). The recombinant vTF1.1 vaccinia virus encoding T7 polymerase was used to drive expression of plasmids with the T7 promoter.
For dye transfer assays , HeLa cells infected with recombinant vaccinia viruses expressing Env proteins and labeled with CMTMR (494/517, red) and target SupT1 cells labeled with Calcein (541/565, green) in suspension were co-cultured (Molecular Probes, Eugene, OR). Kinetic experiments were conducted in which target SupT1 cells in suspension in 37°C media were added to plated effector cells at various times during a two hour period. The cells were then examined by fluorescence microscopy for dye transfer between Env and receptor expressing cells indicating cell-cell fusion. Phase and fluorescent images were collected using an Olympus IX70 coupled to a CCD camera (Princeton Instruments, Trenton, NJ) with a 10× objective lens. An 82000 optical filter cube (Chroma Technology Corp., Brattleboro, VT) was used for the excitation of calcein and CMTMR. Three images per well were collected and then analyzed using Metamorph software (Universal Imaging, West Chester, PA) for dye transfer from the donor to the acceptor cell. The scoring of fusion events was conducted as previously described . The results were normalized by the control fusion and expressed as a percentage. Curves for fusion inhibition assays were fitted, using a hyperbolic decay function, and the IC50 was extracted. Fusion kinetics curves were fit to a sigmoidal function and the half time, at which 50% fusion was achieved, was extracted.
gp120s were produced from 293T cells calcium phosphate transfected with gp120 expression constructs and infected with vTF1.1 vaccinia virus. Cell culture supernatants were harvested 24 hours post transfection/infection and gp120 concentrations determined by ELISA as previously described  with the exception that different antibodies were utilized. Receptor binding efficiencies of gp120s were determined using a cell surface-binding assay in which bound protein is detected by Western blot analysis [38, 43]. Briefly, 2 × 106 QT6 target cells were calcium phosphate transfected with 6 ug pcDNA3.1 (control), CD4 or CXCR4 expression plasmids in 25 cm2 culture flasks and infected with vTF1.1 to boost expression. 24 hrs post-transfection/infection, cells were incubated with gp120 for 2 hours at room temperature with and without 5 ug/ml soluble CD4 (sCD4) to trigger coreceptor binding site exposure. Cells were washed 3× with cold PBS to remove unbound gp120 then lysed with NP40 lysis buffer (0.5% NP40, 150 mM NaCl, 50 mM Tris pH 8) on ice for 10 minutes. Clarified lysates were assayed for gp120 content by SDS-PAGE and Western blotting and detected with an HIV-1 Env specific rabbit serum and an HRP-conjugated anti-rabbit antibody (Amersham Life Science, Piscataway, NJ) or a HIV-2 Env reactive MAb and an HRP-conjugated anti-mouse antibody (Promega, Madison, WI) followed by Supersignal chemiluminescent substrate (Pierce, Rockford, IL). Band intensity was quantitated using Kodak 1D software.
We thank Aimee Kessler for technical assistance. We are grateful to the NIH AIDS Research and Reference Reagent Program for supply of Sup-T1 cells and VSC50, VVROD and VPE16 recombinants. We are grateful to AnorMED Inc. for a gift of AMD3100. We thank Geneveffa Franchini for the supply of HIV-2 envelope plasmids. We thank the members of the Blumenthal lab for their helpful suggestions. This research was supported [in part] by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.