![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article 
Journal of Virology, June 2008, p. 5118-5126, Vol. 82, No. 11
0022-538X/08/$08.00+0 doi:10.1128/JVI.00305-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
Department of Pathology and Laboratory Medicine, Emory Vaccine Center, and Yerkes National Primate Research Center, Emory University, Atlanta, Georgia 30329
Received 11 February 2008/ Accepted 8 March 2008
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-helical conformation on membrane contact. We have investigated the membrane-disruptive requirements for this region using a panel of mutants that replace most of the MPER with antibacterial, membrane-active peptides. The results demonstrate that the mutant Envs were processed, transported, and expressed on the cell surface similar to wild type. Some of the mutant Envs induced moderate levels of cell-cell fusion, demonstrating that the region can accommodate the substitution of proline-rich foreign peptides while retaining significant biological function. In contrast, the incorporation into and stability of the mutated Envs in virions was reduced, consistent with the severely impaired viral entry observed for all the mutants. These data suggest that both structural (for Env incorporation) and functional (membrane disruption) constraints may contribute to the highly conserved nature of this region. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Previous studies in this laboratory have shown that the MPER is important for both viral fusion and incorporation of the Env complex into virions (30). Its function appears to be dependent on the five highly conserved tryptophan residues that make up more than 25% of the region since even single tryptophan substitutions interfered with Env incorporation and virus infectivity (30). The significance of this conserved region is also emphasized by the fact that it contains epitopes to three neutralizing monoclonal antibodies, 2F5, 4E10, and Z13 (25, 34, 46), and that it overlaps the C terminus of the fusion-inhibiting peptide T20 (40). Moreover, both neutralizing and masking antibodies directed at this region can be identified in HIV-1-infected individuals (2, 18, 19).
The conformation of the MPER has been an area of speculation. It has been unclear what conformation the region adopts in the mature Env trimer and at the various stages of viral fusion. Recently, Sun et al. (37) suggested that the MPER can assume a kinked-helix conformation with a majority of the tryptophans embedded in the lipid bilayer. This conclusion is different from the straight -helical structure proposed by Schibli et al. (31), who suggested that most of the aromatic residues in this region form a "collar" along the helix. This collar refers to the way in which four of the tryptophan residues and the single tyrosine residue are positioned on one face of the helix, allowing the MPER to establish "velcro-like" interactions with the viral membrane. Suarez et al. (35) demonstrated that a peptide corresponding to the MPER had the potential to permeabilize membranes and induce lipid mixing. This property could contribute to membrane destabilization during viral fusion.
The goal of this study was to determine the importance of the membrane-perturbing properties of the MPER region. In order to test this, we designed chimeras that replaced part or all of the MPER with a sequence of a membrane-perturbing cationic peptide, indolicidin (32, 36), or its helical alanine analog, CP10A (17). Indolicidin is a 13-amino acid, naturally occurring antimicrobial peptide (32) isolated from bovine neutrophils. It aggregates or stacks up at the membrane (14), thereby destabilizing lipid layers by forming channels. When mixed with erythrocytes, the indolicidin peptides displayed hemolytic activity (36). This peptide (ILPWKWPWWPWRR) is similar to the MPER in that it is rich in aromatic residues, is capable of disrupting membranes, and has an unusually high number of tryptophans. It has been reported that indolicidin, in a manner similar to the MPER, also forms an aromatic collar (29) at the membrane interface. CP10A, an alanine analog of indolicidin (ILAWKWAWWAWRR), has more potent hemolytic activity but, in contrast to the latter, has a strong tendency to form an -helix. We hypothesized that if one of the key functions of the MPER were to destabilize the viral lipid bilayer during fusion, insertion of these foreign peptides would yield a biologically functional Env.
Our results demonstrate that the MPER can be replaced by a proline-rich membrane-disruptive peptide while retaining Env fusion activity, suggesting that it is the membrane perturbing properties of the region that are key to its membrane fusion function. In contrast, none of the indolicidin-Env chimeras were efficiently or stably incorporated into virions, suggesting that structural features beyond just the tryptophan-rich nature of the native MPER are critical to this process.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mutagenesis and plasmid vectors.
An NL4.3 (laboratory-adapted X4 strain) KpnI-BamHI envelope fragment was cloned into pSP72 (Promega, Madison, WI) to facilitate mutagenesis. All mutations in the MPER, except the more complex indolicidin chimeras, were introduced using QuikChange (Stratagene, La Jolla, CA). Indolicidin mutants that required substitutions of up to 36 bases were constructed using site-directed ligation-independent mutagenesis as described by Chiu et al. (7) with the following modifications: the PCR was carried out in a volume of 50 µl, the number of PCR cycles was reduced to 18, and the annealing temperature was reduced to 44°C (for 50 s). Betaine was not added to the PCR, and Pfu polymerase (Stratagene) was substituted for a mixture of Taq and Pfx. The DpnI digestion was done in NEB buffer 4 (50 mM KAc, 20 mM Tris-Ac, 10 mM MgAc, 1 mM dithiothreitol [pH 7.9]; New England Biolabs, Ipswich, MA). After digestion, the mixture was concentrated to 25 µl in water, the product DNA was hybridized using two cycles of 65°C for 5 min and 30°C for 15 min, and then 10 µl of this DNA was used to transform DH5 competent cells (Invitrogen, Carlsbad, CA).
The mutated envelope fragment was subcloned into the simian virus 40-based vector pSRHS (30). An NheI (nucleotide 7250)-XhoI (nucleotide 8887) envelope fragment from pSRHS was then subcloned back into provirus pNL4.3 (1) and pNLd. The plasmid pNLd is a replication-defective NL4.3 proviral vector analogous to a pFN variant of HXB2 described previously with a 300-bp deletion in the pol gene (11).
Cell-cell fusion assays. 293T cells were transfected with pNLd proviral plasmid vectors expressing wild-type (WT) and mutant envelope proteins (Env) using Fugene 6 transfection reagent (Roche Diagnostics Corporation, Indianapolis, IN). At 20 h posttransfection, the transfected 293T cells were mixed in a 1:5 ratio with TZM-BL cells and then replated in 24-well plates. After 18 to 20 h, the cells were microscopically examined for fusion following the addition of 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (Fisher Scientific) to stain Env- and Tat-expressing cells blue. Fusion was also quantitated by measuring luciferase activity (Luciferase Assay System; Promega) according to the manufacturer's protocol.
Flow cytometry assay for cell surface expression of Env.
The pNLd proviral vectors encoding WT and mutant Envs were transfected into 293T cells in six-well plates. At 36 to 40 h posttransfection, the cells were trypsinized and washed in flow cytometry buffer (phosphate-buffered saline [PBS] with 0.1% bovine serum albumin [BSA] and 0.1% NaN3). Cells were resuspended in 500 µl of buffer. Cell surface-associated Env molecules were fluorescently labeled by incubating 100 µl of harvested 293T cells with 
0.5 mg/ml of AlexaFluor 647-conjugated gp120-specific b12 monoclonal antibody at room temperature for 1 h. The b12 antibody (a gift from Dennis Burton, Scripps Research Institute, La Jolla, CA) was tagged with the Alexa dye using the manufacturer's (Invitrogen) instructions. In order to control for transfection efficiency, cells were then permeabilized using BD Cytofix/Cytoperm (BD Biosciences, San Diego, CA) and fluorescently stained with fluorescein isothiocyanate-conjugated anti-Gag KC57 (Beckman Coulter, Fullerton, CA). The Env staining was evaluated by flow cytometry with a FACSCalibur (Becton Dickinson, Franklin Lakes, NJ) instrument; the data were analyzed using FloJo software, version 8.2 (Tree Star Inc., Ashland, OR), and the mean fluorescence intensity was calculated from the acquisition of at least 50,000 gated events.
Radioimmunoprecipitation. The pNLd proviral vectors encoding WT and mutant Envs were transfected into 293T cells in 60-mm-diameter plates. At 42 to 48 h posttransfection, the cells were starved in methionine- and cysteine-free DMEM for 15 min and pulse labeled in 500 µl of methionine-cysteine-free DMEM supplemented with [35S]methionine-cysteine (0.20 mCi/well; Invitrogen) for 30 min. The labeled cells were then chased for 2 or 4 h in complete medium. The cells were washed with PBS and lysed on ice using lysis buffer A (25 mM Tris, pH 8.0, 50 mM NaCl, 1.0% Triton X-100 in PBS). The cellular debris and nuclei were removed by centrifugation in a microcentrifuge at 2,000 x g for 10 min at room temperature (RT). The medium and cell lysate were precleared by incubating at RT with fixed Staphylococcus aureus cells for 30 min. Env was immunoprecipitated from the cell lysate and medium by incubating with a mixture of AIDS patient serum (kindly provided by Jeffery Lennox, Clinical Core, Emory Center for AIDS Research) and gp120 monoclonal antibody (F105; NIH AIDS Research and Reference Reagent Program) at 4°C overnight. The antibody complexes were incubated with S. aureus cells for 30 min at RT and pelleted by centrifugation at 13,000 rpm in a Beckman microcentrifuge for 1 min at RT. The pellets were washed three times in lysis buffer B (LBB; lysis buffer A containing 0.1% sodium dodecyl sulfate [SDS]) and once in 20 mM Tris-HCl (pH 6.8) and analyzed by SDS-polyacrylamide gel electrophoresis (PAGE).
For the analysis of viral proteins, the pNLd vectors encoding WT or mutant Envs were introduced into 293T cells by Fugene 6 transfection. At 48 h posttransfection, the cells were radiolabeled as described above. The culture medium was filtered through a 0.4-µm-pore-size filter to remove any cell debris and centrifuged in a type 50.4 Ti fixed-angle rotor (Beckman Coulter, Fullerton, CA) at 285,000 x g for 2 h at 4°C through a 1-ml sucrose cushion (20% sucrose [wt/vol] in PBS). The supernatant was removed, and an appropriate volume of 5x LBB was added prior to immunoprecipitation. The viral pellet was lysed in 0.5 ml of LBB, precleared, and then immunoprecipitated as described above. Following PAGE, the dried radioactive gels were exposed to X-ray film for 48 h at –80°C or to an OptiQuant screen overnight at RT. The band intensities were measured from the exposed screen using a Cyclone phosphorimager with OptiQuant software (Perkin Elmer, Shelton, CT).
Immunoblotting. The pNLd proviral vectors encoding WT and mutant Envs were transfected into 293T cells in 60-mm-diameter plates. Forty-eight hours after transfection, the cell supernatant was clarified, and the virus was pelleted through a sucrose cushion as described above. Following PAGE, the proteins were transferred to a polyvinylidene difluoride membrane (Immobilon-P; Millipore, Bedford, MA) and analyzed by immunoblotting using either anti-gp41 antibodies (murine monoclonal antibody produced from the Chessie 8 hybridoma provided by George Lewis; AIDS Research and Reference Reagent Program) or anti-p24 mouse monoclonal antibodies (183-H12-5C; provided by Bruce Chesebro and Kathy Wehrly, AIDS Research and Reference Reagent Program) and a chemiluminescence detection kit (Pierce, Rockford, IL). The levels of gp41 were indirectly measured by assessing the band intensity using the public domain NIH ImageJ software (available at http://rsb.info.nih.gov/ij/).
ELISAs. Released virus was quantitated by determining the amount of p24 antigen in the culture supernatant according to the manufacturer's instructions (HIV-1 p24 antigen enzyme-linked immunosorbent assays [ELISA] kit; PerkinElmer Life Sciences, Boston, MA). The concentrations of gp120 on the surface of virions and of gp120 shed into the medium were measured by a capture ELISA. To measure the relative amount of gp120 on virions, the cell supernatant was clarified, and virus was pelleted through a sucrose cushion as described above. The tubes were washed three times with 1 ml of PBS following collection of the supernatant to prevent contamination of virion gp120 with shed gp120; then viral pellets were resuspended in 500 µl of LBB. The anti-gp120 capture antibody, D7324 (Cliniqa, Fallbrook, CA), was coated overnight at 37°C onto 96-well plates at a concentration of 4 µg/ml in PBS. The plates were then blocked with blocking buffer (1.0% milk protein, 0.5% BSA, and 1.0% gelatin, in 50 mM carbonate/bicarbonate buffer pH 9.6) for 1 h at 37°C. After the plate was washed, the blocking step was repeated. Virions were diluted in Env sample diluent (1.0% milk protein, 0.5% BSA, 1.0% gelatin, and 0.5% Triton X-100 in 10 mM PBS) and incubated on D7324-coated plates for 2 h at 37°C. The amount of captured gp120 was determined using HIV-positive patient serum (designated DOCR) at a dilution of 1:6,400 in Env sample diluent and incubating for 1 h at 37°C. Detection was performed using a horseradish peroxidase-conjugated goat anti-human immunoglobulin G (heavy and light chain) (Pierce) at a dilution of 1:5,000 in Env sample diluent for 1 h at 37°C. Colorimetric analysis was performed using an Immunopure TMB Substrate Kit (Pierce), and absorbance was read at 450 nm with a reference at 620 nm. Recombinant BaLgp120, obtained from the Institute of Human Virology, University of Maryland, was used for the standard curve. The linear detection of this assay was sensitive to less than 20 pg of gp120.
Virus infectivity assay. 293T cells were transfected with 800 ng/well of NL4.3 WT and mutant proviral plasmid DNA in 48-well plates and incubated for 48 h. The levels of p24 antigen in the culture supernatants were determined by ELISA as per the manufacturer's instructions (Beckman Coulter, Fullerton, CA), and normalized amounts of virus, based on this assay, were used to infect TZM-BL cells. Quantification of luciferase activity 36 to 40 h postinfection (Luciferase Assay System; Promega) provided a measure of virus infection in the target cells.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
-helix (17). Cytoplasmic domain truncations in HIV-1 gp41 have been reported to enhance the fusogenic potential of the Env (16). We therefore truncated the cytoplasmic tail of mutant I-ITN (designated I-ITN-Tr) to determine whether this mutant showed enhanced cell-cell fusion and Env incorporation.
Cell-cell fusion, Env biosynthesis, and flow cytometry.
In order to investigate the fusogenic potential of the different Env proteins, we performed a cell-cell fusion assay in which 293T cells transfected with WT and mutant proviruses were mixed with TZM-BL cells. Cell fusion was scored by counting blue (β-galactosidase positive) cells and by quantitating luciferase enzyme activity. In these cell-cell fusion assays, the I-RR mutant was severely defective although it still exhibited approximately 7% of the fusion observed with WT Env (Fig. 2A). The CP-RR mutant, in which each of the prolines was replaced by an alanine residue, yielded near-background levels of fusion, suggesting that the substitution was detrimental to biological activity. Strikingly, replacing the two membrane-proximal arginines in I-RR with alanines (I-AA) increased fusion up to 35% of the WT level, and adding the ITN spacer peptide (I-ITN) further increased fusion up to 50% of that of the WT despite the presence of three proline residues in this new MPER region. In contrast, CP-ITN showed only marginally higher fusion (5%) than the I-RR variant. The I-ITN-Tr, which lacked the cytoplasmic domain, showed reduced fusion (18 to 21% of that of the WT) compared to its parental I-ITN Env, while the truncated version of the WT (WT-Tr) showed increased fusion (125% of WT).
|
To determine whether substitution of foreign amino acid sequences into the MPER region interfered with normal biosynthesis, processing, and transport of Env, a pulse-chase labeling experiment was performed. 293T cells were transfected with proviruses encoding the indolicidin mutants and after 48 h were radiolabeled with [35S]methionine as described in Materials and Methods. After lysis, immunoprecipitation, and SDS-PAGE, the Env protein bands were quantitated using a phosphorimager. The gp160/gp120 ratios of the mutants and the WT were similar (median ratio, 3.3) All of the mutant glycoproteins were expressed and processed at levels similar to those of the WT (Fig. 3).
|
Virus infectivity and Env incorporation. In order to determine whether the chimeric Env proteins could mediate virus entry into cells, WT and mutant proviruses were transfected into 293T cells, and at 48 h posttransfection culture supernatants were collected, quantitated for p24, and then used to infect TZM-BL cells. All of the mutants showed markedly decreased infectivity levels relative to the WT, with levels of expressed luciferase in some cases barely above background (Fig. 4). Mutant I-RR showed the least induction of luciferase, while CP-RR, I-ITN, CP-ITN, and I-AA all showed a similar reduction in infectivity compared to WT (by more than 150-fold). The relative infectivities of the mutants did not reflect the fusogenic capacity of the mutants observed in cell-cell fusion assays (Fig. 2A).
|
Despite the fact that similar levels of p24-containing virions were released from WT- and mutant-transfected cells (Fig. 5A), glycoprotein incorporation was undetectable in all of the chimeric Env virions. All of the mutants exhibited shedding of Env into the culture supernatant at a level similar to that of the WT (Fig. 5C).
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-helical structure on membrane contact (31, 35). The indole ring of tryptophan has a strong tendency to insert into the membrane-water interface and disrupt the phospholipid organization (27). Indeed, isolated peptides corresponding to the MPER have been shown to induce vesicle leakage (35) and thus behave as true fusogens.
A striking finding of the studies presented here is that, despite the nearly perfect conservation of the MPER region within the 1,400 Env sequences in the Los Alamos National Laboratory database, this region can be replaced by a foreign, membrane-active, tryptophan-rich sequence without significant loss of fusogenic activity. These results from the indolicidin chimeras suggest that it is the membrane-disruptive properties of the MPER rather than its conserved sequence that play a key role in fusion.
When the entire antibacterial peptide, indolicidin, replaced most of the MPER (mutant I-RR), fusion was acutely affected but not abrogated. The efficiencies were lowered to approximately 7% of WT. However, at least part of the defect could be attributed to the C-terminal arginine residues in mutant I-RR because when they were replaced by alanines in mutant I-AA, the fusion efficiency increased to approximately 35% of that of WT; the two alanines likely allowed the C-terminal region to associate more stably with the membrane. The spacing of the tryptophans in this region also appears to contribute to fusion activity. The I-ITN mutant was designed to better mimic the MPER, while retaining three proline residues, by inserting the spacer peptide ITN into the indolicidin peptide so that the original positions of all the MPER tryptophan residues are maintained. The insertion increased the fusion efficiency of I-AA from approximately 33% to up to 50% (I-ITN) of that of the WT.
It is noteworthy that even though the I-AA and I-ITN chimeric Envs have a high percentage of proline residues in the MPER region and a completely different sequence from that seen in HIV-1, they retained significant fusogenic potential. Prolines are known to be helix breakers in aqueous environments due to the size of their pyrrolidine rings and the lack of an amide proton that promotes intramolecular H-binding (21, 41). Thus, it is unlikely that either I-AA or I-ITN MPERs assume an -helical conformation, although this cannot be completely ruled out because the role of prolines might vary depending on differences in the molecular environment imposed by membrane lipids and proteins (22). Nevertheless, our conclusion that the MPER conformation is more complex than simply a stable -helix is supported by the phenotypes of the CP10A-based mutants (the alanine analog of the indolicidin mutants). These exhibited greater reductions in fusion efficiency than the indolicidin-based mutants even though the peptide CP10A has been shown to be more membrane disruptive than indolicidin itself and exhibits a strong -helical character (17) due to the presence of alanines, which are reported to be helix stabilizers (6). These results indicate that forming an inflexible -helical structure can be inhibitory to the process.
There have been a variety of predictions for the structure that the MPER assumes at various stages of fusion, but all of these are based on studies with peptides corresponding to the MPER and are limited by the lack of a structure for the ectodomain of the native Env trimer. In addition to the results of Schibli et al. (31), Biron et al. (5) proposed that residues 659 to 671 overlapping the N-terminal end of the MPER assume a 310-helical conformation with 3.2 amino acid residues per turn of the helix. This suggestion was supported by a report by Barbato et al. (3), which argued that the MPER initially assumes a closed β-turn conformation prior to gp120 binding and receptor activation. This would be followed by an extended conformation and, finally, an -helix. The recent structural analysis by Sun et al. (37) of the MPER peptide in a lipid bilayer allowed them to propose an L-shaped conformation, with an N-terminal region (residues 664 to 672) that contains a two-turn helix joined by a flexible 2-amino-acid linker region to a C-terminal region (675 to 683) that contains a one-turn -helix and a 310-helix. The authors conclude that the N terminus of the MPER is flexible and can move independently of the C terminus, which is rigidly associated with the plasma membrane. Antibody binding studies further suggest that the flexibility of the MPER domain conferred by the kink in the -helix might be important both for its insertion into the membrane and for the potency of neutralizing antibodies that target this region. This flexible kinked-helix model may explain why replacing the proline-rich indolicidin peptide in mutant I-AA or I-ITN continues to allow fusion while replacing the MPER with the strongly -helical CP-ITN peptide yielded a much less fusogenic Env. It remains possible that the MPER needs to adopt various intermediate conformations as it transitions from its function in the trimeric prefusion form of Env, for which we have no structural information, to that which plays a role in the fusion process.
The data presented here support the hypothesis that the ability of the MPER to destabilize the viral membrane is critical to the region's role in the HIV-1 fusion process. Investigators have speculated on the manner in which the region might disrupt the membrane and on the stage of fusion at which the MPER attaches to the bilayer. The tryptophans in the MPER might reorganize the lipids in the viral membrane, in a mechanism similar to that proposed by Robison and Whitt (28) for vesicular stomatitis virus G proteins, causing it to curve and aid subsequent fusion with the target membrane. Schibli et al. (31) have also suggested that after gp120-CD4 binding, the new conformation of gp41 allows the topologically adjacent fusion peptide in the target cell membrane and the MPER in the viral membrane to destabilize the apposed membranes and facilitate fusion. They have also theorized that by inserting itself, in its entirety, parallel to the viral membrane, the MPER facilitates the remainder of the gp41 ectodomain to engage the target cell membrane and bring it closer to the virus. In contrast, Sun et al. (37) have proposed that the flexible N terminus of the MPER that moves independently of the more rigid C terminus might be responsible for destabilization of the target membrane.
Decreased surface expression was not the basis for the significantly reduced fusion observed with the indolicidin-Env chimera, I-RR, or the CP10A chimeras, CP-RR and CP-ITN, since all of the Env chimeras were expressed at WT levels. This would indicate that the MPER region is actually quite accommodating in terms of the sequence changes it can tolerate with regard to folding, intracellular transport, and cell surface expression. Moreover, because the proline-rich I-ITN Env chimera retains approximately 50% of the fusion activity of the WT, it can be further argued that the high degree of amino acid conservation observed within the MPER region, which extends beyond the tryptophan residues themselves, is not being driven by constraints that are required for fusion alone.
Truncation of the cytoplasmic domain of HIV-1 has been reported to increase the fusogenicity of the protein (16). We therefore investigated the effects of truncation of the I-ITN mutant on its fusogenicity. While truncation of the cytoplasmic domain in WT pNL4-3 Env (mutant WT-Tr) did moderately increase fusogenicity, in the context of the I-ITN Env chimera, fusion was actually decreased by approximately 40 to 50% (mutant I-ITN-Tr). Cell surface expression as measured by binding of the CD4 binding site monoclonal antibody b12 was increased more than threefold for both truncated constructs, but this may be in part due to induced alterations in the ectodomain of gp41 and gp120, since enhanced exposure of conserved antibody-binding epitopes has been reported previously following truncation of the cytoplasmic domain (13). These experiments suggest, therefore, that the MPER modifications combined with a more highly exposed CD4 binding domain result in a protein with less fusion activity.
In single-cycle infection assays, all of the mutants showed very poor levels of infectivity. A major reason for this defect appears to be a lack of stable Env incorporation into virions, with all of the mutants exhibiting a greater than 90% reduction in glycoprotein incorporation relative to WT when gp120 was quantitated. From analysis of gp41 immunoblots, it was evident that some amount of Env does get incorporated into the mutant virions, but even for Env mutant I-ITN, which exhibits the highest level of cell-cell fusion, the amount of gp41 incorporated is less than one third that observed in WT. It is likely that the Env complex that is assembled at the viral membrane is less stable so that gp120 is lost from the virion surface. It should be noted, however, that this reduced stability is not manifest in the level of cell surface Env and results in only a small increase in soluble gp120 in the cell culture supernatant, suggesting that it might be most pronounced after Env is incorporated into virions. An understanding of how the MPER influences Env incorporation has been hampered by a lack of information on the native structure of Env on the surface of the virion. Recent cryo-electron tomography studies have unfortunately been conflicting in their conclusions in this regard. The work of Roux et al. (44) on a form of the simian immunodeficiency virus Env with a deletion of the cytoplasmic tail has suggested a splayed-legs model for this Env, similar to that proposed for the Env of Moloney murine leukemia virus (15), wherein the Env trimer positions itself in the viral membrane like a tripod. Their results suggest that the MPER regions of three gp41 molecules in the trimer do not self-associate but interact extensively with the plasma membrane as widely spaced parts of a single spike. This would be consistent with X-ray crystallography studies of the neutralizing monoclonal antibodies 2F5 and 4E10 bound to their peptide epitopes, which also indicated a close membrane association of the MPER region in its antibody-sensitive conformation (26). In contrast, recent reports by Zanetti et al. (43) and Bennet et al. (4), who studied the same simian immunodeficiency virus Env, suggest that the MPER regions in a trimer self-associate, leading to a compact stalk-like Env structure where there is no separation between the monomers.
Irrespective of which structural analysis is correct, it seems likely that two possible explanations could account for the reduction in Env incorporation mediated by the mutations described here, in addition to any instability of the Env complex that results in shedding of gp120 after incorporation into virions. First, the conformational changes in the MPER induced by the mutations could, in the native prefusion form of Env, prevent the necessary interactions between Env and the assembling Gag precursors. In the splayed-legs model for Env, mutations that altered the interaction of MPER with the lipid bilayer might be expected to alter the spacing of the legs, potentially preventing their insertion into holes in the Gag lattice. In contrast, a structure where the three MPER regions interact in a stalk might well be disrupted by the mutations we have introduced into this region, allowing the MPER to interact instead with the membrane through its tryptophan residues, which in turn could alter the ability of gp41 to interact with assembling Gag molecules (10). Alanine substitutions for hydrophobic residues in the MPER region did increase binding of 2F5 and 4E10 monoclonal antibodies to Env, consistent with an enhanced association of this mutated region with the lipid bilayer (45). A second explanation for the defect in chimeric Env incorporation into virions could be that this altered interaction of the MPER region of Env with the lipid bilayer causes the viral glycoproteins to be partitioned into a region of the membrane that is distinct from the site of Gag assembly and budding (8).
In conclusion, although the MPER can accommodate large substitutions and retain biological function (seen in cell-cell fusion assays), this region is very sensitive to modifications in its sequence as far as the prefusion, native form of Env that must be incorporated into virions is concerned. It is likely that both structural (for glycoprotein stability and incorporation) and functional (membrane disruption) constraints may contribute to the highly conserved nature of the MPER. Given the roles that it plays in ensuring glycoprotein incorporation and mediating membrane fusion, as well as the fact that this region has linear epitopes for key neutralizing monoclonal antibodies and fusion-inhibiting peptides, MPER remains an attractive target for antiviral drug and vaccine design.
|
ACKNOWLEDGMENTS |
|---|
We thank Ling Yue for help with some of the radiolabeling and pulse-chase experiments; Rama Amara and William Diehl for helpful technical discussions; Jacqueline Allen and Jeffrey Meisner for technical assistance; and Cynthia Derdeyn, Jerry Blackwell, Richard Haaland, and Paul Spearman for critical input on the manuscript. HIV-1 gp120 monoclonal antibody (F105) was obtained through the NIH AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, NIH.
|
FOOTNOTES |
|---|
Published ahead of print on 19 March 2008.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | Mol. Cell. Biol. | Microbiol. Mol. Biol. Rev. |
|---|
| Clin. Vaccine Immunol. | ALL ASM JOURNALS |
|---|
