![[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}
|
|
|
|
|
||
Vol. 19, Issue 10, 4298-4309, October 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,,||
*Cellular Protein Chemistry, Bijvoet Center for Biomolecular Research, Faculty of Science, Utrecht University, 3584 CH Utrecht, The Netherlands; and Department of Human Retrovirology, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands
Submitted December 26, 2007;
Revised July 2, 2008;
Accepted July 10, 2008
Monitoring Editor: Reid Gilmore
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; van Anken and Braakman, 2005). Disulfide bonds are considered essential for folding and/or for structure and function of mature proteins, which explains why the disulfide-bonded structure of proteins usually is conserved across species and within protein families. An intriguing example is the Envelope glycoprotein (Env) of human immunodeficiency virus-1 (HIV-1): it has completely conserved disulfide bonds in the context of a hypervariable sequence. This and the fact that Env has easily testable and well-defined functions for the virus made Env an excellent model for studies on the role of individual, conserved disulfide bonds for a protein's folding and function.
Env is the sole viral protein present on the surface of HIV-1 virions (Allan et al., 1985). It is synthesized as a 160-kDa precursor type I membrane protein (gp160). The protein folds and trimerizes in the ER of the host cell, where it obtains 10 disulfide bonds and 
30 N-linked glycans (Leonard et al., 1990). In the Golgi complex, gp160 is cleaved by a cellular furin(-like) protease into its soluble subunit gp120 and transmembrane subunit gp41 (Stein and Engleman, 1990; Moulard and Decroly, 2000). These remain noncovalently associated on the surface of infected cells and virions. Together, the two Env subunits mediate viral entry: gp120 is responsible for binding to its receptor (CD4) and coreceptor (CCR5 or CXCR4) on the host cell, and gp41 is needed for subsequent fusion of the viral and cellular membranes (Wyatt and Sodroski, 1998; Doms and Moore, 2000; Eckert and Kim, 2001).
Sequence variability of Env among different HIV strains is notorious and in part explains why the humoral immune response against HIV-1 is inadequate in patients. Amid the countless changes in surrounding residues, the cysteine residues are strictly conserved among all isolates of HIV-1 (Louwagie et al., 1995; Los Alamos HIV sequence database: http://hiv-web.lanl.gov). This suggests that the antigenic shield of Env may vary, but that the disulfide-bonded structure of Env and hence the basic architecture of the molecule is constant.
Disulfide bonds may be important for the folding of Env in the ER. Lack of particular disulfide bonds often renders newly synthesized proteins in the ER maturation incompetent: they either aggregate and fail to fold or they partially oxidize into species similar to folding intermediates of the wild-type protein. Alternatively, disulfide bonds contribute to the structural integrity of the mature protein. In the case of Env, disulfide bonds seem to be important also for the conformational changes during receptor binding and subsequent fusion of the viral and host cell membranes (Barbouche et al., 2003; Matthias and Hogg, 2003; Markovic et al., 2004).
Cysteine mutants of ER client proteins are excellent tools to provide insight into the relevance of individual disulfide bonds, but so far no systematic cysteine mutation studies of highly disulfide-linked ER clients have been published. Some Env cysteine mutants have been analyzed in different strains (Hemming et al., 1989; Tschachler et al., 1990; Bolmstedt et al., 1991; Freed and Risser, 1991; Syu et al., 1991; Dedera et al., 1992; Lekutis et al., 1992), but together they do not give a consistent notion of the relative importance of the 10 disulfide bonds of Env for folding and function.
To systematically assess the role of each individual disulfide bond in Env folding and function, we analyzed a complete series of mutants, where all cysteines were replaced by alanines, both individually and pair-wise. In spite of the strict conservation of all 10 disulfide bonds, only five turned out to be essential for folding of Env in the ER, whereas two disulfide bonds were expendable not only for folding but even for function of Env.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) of the HIV-1LAI isolate. Although we studied Env of the LAI isolate, we followed the canonical HXB2 residue numbering, which relates to the LAI numbering as follows: because of an insertion of five residues in the V1 loop of LAI Env, all cysteine residues beyond this loop have a number 5 residues lower in HXB2 than in LAI: until Cys131, numbering is identical, but Cys162 in LAI becomes 157 in HXB2, etc.
Cells
HeLa cells were cultured in MEM (Life Technologies, Rockville, MD) supplemented with 10% FCS (Life Technologies), penicillin and streptomycin (100 U/ml), 2 mM glutamax (Life Technologies), and nonessential amino acids (Life Technologies). C33A cervix carcinoma cells were maintained in DMEM (Invitrogen), supplemented with 10% FCS, penicillin, and streptomycin. SupT1 T-cells were cultured in RPMI medium supplemented with 10% FCS, penicillin, and streptomycin. Peripheral blood mononuclear cells (PBMCs) were isolated from fresh buffy coats (Central Laboratory Blood Bank, Amsterdam) by standard Ficoll-Hypaque density centrifugation. PBMCs were frozen in multiple vials at a high concentration and, when required, were thawed and activated with 5 µg/ml phytohemagglutinin (Sigma, St. Louis, MO) and cultured in RPMI medium (Life Technologies) containing 10% FCS, penicillin and streptomycin (100 U/ml) with recombinant interleukin-2 (rIL-2; 100 U/ml). On day 4 of culture, PBMCs underwent CD4+ enrichment by incubating them with CD8 immunomagnetic beads (Dynal, Lake Success, NY) and separating out the CD8+ lymphocytes.
Folding Assays
Subconfluent HeLa cells were infected with recombinant vaccinia virus expressing T7 polymerase (Fuerst et al., 1986) to drive expression of Env mutants under control of the T7 promoter. Thirty minutes after infection, cells were transfected with a mixture of 4 µg of mutant or wild-type pcDNA3-gp120 or pcDNA3-gp160, and 10 µl lipofectin (Invitrogen), according to the manufacturer's instructions. Pulse-chase experiments were performed essentially as described (Braakman et al., 1991; Land et al., 2003). Five hours after infection, cells were depleted of cysteine and methionine for 15–30 min before they were pulse-labeled with 50 µCi of Redivue promix L-[35S] in vitro labeling mix (Amersham Biosciences, Piscataway, NJ). Cells were chased for various intervals in medium containing an excess of unlabeled cysteine and methionine. Chase samples were stopped by aspirating the medium and adding ice-cold HBSS (Invitrogen-BRL) containing 20 mM iodoacetamide (IAM) to block free sulfhydryl groups. Cells were lysed in ice-cold MNT (20 mM MES, 100 mM NaCl, 30 mM Tris-HCl, pH 7.5) containing 0.5% (vol/vol) Triton X-100, 20 mM IAM, and protease inhibitor cocktail (10 µg/ml each of chymostatin, leupeptin, antipain, and pepstatin, 1 mM PMSF, and 1 mM EDTA). Cell lysates were spun for 10 min at 15,000 x g to pellet nuclei and postnuclear lysates were immunoprecipitated with a polyclonal antibody that recognizes all forms of HIV-1 Env. In addition, secreted or shed gp120 molecules were immunoprecipitated from the culture media at later chase times. Washed immunoprecipitates were resuspended in 0.2% SDS in 100 mM sodium acetate, pH 5.4, and incubated at 95°C for 5 min. An equal volume of 100 mM sodium acetate, pH 5.4, was added, which contained 2% Triton X-100 in MNT, protease inhibitor cocktail, and 0.0025 U endoglycosidase H (endoH; Roche, Rotkreuz, Switzerland). Samples were incubated for 2 h at 37°C. After incubation, SDS-PAGE sample buffer was added, and samples were incubated at 95°C for 5 min. Samples were analyzed by reducing or nonreducing 7.5% SDS-PAGE. Gels were dried and signals were detected on Biomax MR films (Eastman Kodak, Rochester, NY).
Viruses and Infections
Virus stocks were produced as follows: C33A cells were transfected with 10 µg wild-type or mutant pLAI constructs by calcium phosphate precipitation (Das et al., 1999). Three days after transfection, virus-containing culture supernatants were harvested, filtered, and stored at –80°C. Virus concentrations were quantitated by capsid CA-p24 ELISA (Jeeninga et al., 2000). These values were used to normalize the amount of virus for infection experiments. SupT1 cells (50 x 103) were infected with the equivalent of 500 pg CA-p24 of C33A-produced HIV-1LAI per well in a 96-well plate, and virus spread was measured for 14 d using CA-p24 ELISA. The 50% tissue culture infectious dose (TCID50) in SupT1 cells was determined after 14 d by endpoint dilution.
Ultracentrifugation of Virions
C33A cells were transfected with 40 µg pLAI per T75 flask. Medium was refreshed at day 1 after transfection. At 3 d after transfection, culture media were harvested, centrifuged, and passed through a 0.45-µm filter to remove residual cells and debris. Virus particles were pelleted from filtered culture media by ultracentrifugation (100,000 x g for 45 min at 4°C) and resuspended in 0.5 ml lysis buffer (50 mM Tris-HCl, pH 7.4, 10 mM EDTA, 100 mM NaCl, 1% SDS).
Quantitation of gp120 by ELISA
Concentrations of gp120 in cell, virion, and supernatant fractions were measured as described before (Moore and Ho, 1993; Sanders et al., 2002), with some modifications. ELISA plates were coated overnight with sheep antibody D7324 (10 µg/ml; Aalto Bio Reagents, Ratharnham, Dublin, Ireland), directed to the gp120 C5 region, in 0.1 M NaHCO3. After blocking by 2% milk in Tris-buffered saline (TBS) for 30 min, gp120 was captured by incubation for 2h at room temperature. Recombinant HIV-1LAI gp120 (Progenics Pharmaceuticals, Tarrytown, NY) was used as a reference. Unbound gp120 was washed away with TBS and purified HIV-1+ serum Ig (HIVIg) was added for 1.5 h in 2% milk, 20% sheep serum (SS), 0.5% Tween. HIVIg binding was detected with alkaline phosphatase–conjugated goat anti-human Fc (1:10,000, Jackson ImmunoResearch, West Grove, PA) in 2% milk, 20% SS, and 0.5% Tween. Detection of alkaline phosphatase activity was performed using AMPAK reagents (DAKO, Carpinteria, CA). The measured gp120 contents in virus were normalized for CA-p24.
HIV-1 Neutralization Assay Using Primary CD4+ Lymphocytes
Viruses were tested for their relative inhibition sensitivity against increasing concentrations of either sCD4 (BD ImmunoDiagnostics, Orangeburg, NY) or the CXCR4 binding compound AMD3100 (a generous gift from Dr. D. Schols, Katholieke Universiteit, Leuven, Belgium). TCID50 values were determined on purified CD4+ lymphocytes isolated from an individual who did not carry the 
32CCR5 allele (CCR5+/+) screened for by standard PCR. CD4+-enriched lymphocytes were plated at 2 x 105 cells/well in 96-well plates with fivefold serial dilutions of the virus. Cells were fed on day 7 with fresh media and scored on day 14 for CA-p24 levels, with the number of positive wells being used to identify the TCID50 value for each virus. Each virus (100 TCID50 in 50 µl of the culture RPMI medium) was mixed with an equal volume of serially diluted compound for 1 h at 37°C in flat-bottomed 96-well plates, and all neutralization reactions were performed in triplicate. After 1 h of incubation 2.0 x 105 positively selected CD4+ lymphocytes were added to each well in 0.1 ml of culture media containing rIL-2 (100 U/ml). For each neutralization experiment a positive control, virus incubated with cells in the absence of inhibitory compound, and a negative control, virus in the absence of cells, was included. The negative control CA-p24 concentration was subtracted from all test results. Supernatant was taken on days 7, 10, and 14 and assayed for CA-p24 with fresh media added to the cells. Neutralization responses were determined for each virus at the day CA-p24 levels peaked. We calculated the CA-p24 production in the presence of inhibitory compound as a percentage of that in the cultures with virus only.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Chen et al., 2005; Figure 1B). Six disulfide bonds locate to the gp120 core. The C1 region contains a single disulfide bond, 54–74, whereas the C2 region contains two disulfide bonds: 218–247 and 228–239. The 378–445 disulfide bond covalently links the C3 and C4 regions. Two bonds are found at the base of variable loops: 296–331 (V3) and 385–418 (V4). The two most prominent loops, which encompass the V1 and V2 regions, have three disulfide bonds at their stem(s): 119–205, 126–196, and 131–157. The disulfide bond in gp41 closes off a very small loop: 598–604. We systematically replaced all cysteines in the ectodomain of Env by alanines and refer to mutants as follows: C54A and C74A are the single mutants, and C54/74A is the double mutant of disulfide bond 54–74.
|
). In HeLa cells, we expressed Env from plasmids under control of the T7 promoter, using a recombinant Vaccinia virus vector system (Fuerst et al., 1986). We radioactively pulse-labeled cells and chased them in the presence of excess unlabeled amino acids. Immediately after the pulse or after various chase periods, cells were cooled on ice, incubated with an alkylating agent to block free sulfhydryl groups, and lysed. We immunoprecipitated Env from the lysates and deglycosylated samples with endoH. At the end of longer chase periods, we also immunoprecipitated Env from culture media. We then analyzed folding kinetics of Env by reducing and nonreducing SDS-PAGE. We not only expressed the gp160 Env precursor but also the soluble subunit gp120. The pulse-chase provides three assays for maturation of Env, two of which are based on differences in mobility of Env conformers in SDS-PAGE (Figure 2).
|
). Immediately after the pulse, wild-type gp160 and gp120 appeared as a fuzzy band in nonreducing gels (Figure 2, 0 h, nonreducing) with mobility close to that of the corresponding reduced protein. The reduced samples appeared as a discrete band however (Figure 2, 0 h, reducing), implying that some disulfide bonds had formed already before the end of the pulse. With time, "smears" of faster running protein appeared in the nonreduced samples, corresponding to various species of partially oxidized folding intermediates (ITs). After 4 h of chase, most of the wild-type gp160 and gp120 proteins appeared as a sharp band in nonreducing gels (Figure 2, 4 h, nonreducing), indicating that they reached the fully oxidized native state (NT).
Proteins that fold in the ER are targeted to enter this compartment by a signal peptide (Walter and Johnson, 1994), which in soluble or type I proteins is cleaved off already during synthesis. A striking exception however is the signal peptide of HIV-1 Env, which is removed long after synthesis has been completed (Li et al., 1994). The signal peptide of Env can be removed only after the protein has gone through some initial folding steps, making this cleavage another indicator of Env maturation (Land et al., 2003). By analyzing the chase samples via reducing SDS-PAGE, we hence used signal peptide cleavage as a second assay for Env maturation. Immediately after the pulse only a single band was visible (Figure 2, 0 h, reducing), which corresponds to the reduced protein with its signal peptide still attached (Ru = reduced uncleaved). With time, a second band appeared below Ru, which corresponds to the reduced protein from which the signal peptide had been removed (Rc = reduced cleaved). After 4 h of chase, the signal peptide had been removed from the majority of wild-type gp160 and gp120 molecules. Except for some faint band at the top of the gel, which likely represents aggregated Env, we detected no other bands or smears in the reducing gel. This implies that intracellular pools of Env were fully deglycosylated by endoH, as we previously reported (Land et al., 2003). Env did undergo heterogeneous glycan modifications in the Golgi as is evident from the "smeary" character of the shed and secreted gp120.
This gp120 harvested from culture media was analyzed by reducing SDS-PAGE as a third assay for Env maturation. Correctly folded Env exits the ER and travels along the secretory pathway. Env expressed as soluble subunit gp120 is secreted, whereas gp160 undergoes proteolytic cleavage into gp41 and gp120 subunits. Because the gp120 subunits are no longer covalently attached to the remainder of the Env spikes, they quantitatively shed into the culture medium. As anticipated, substantial amounts of both shed and secreted (endoH resistant) wild-type gp120 were present in the culture medium after 4 h of chase (Figure 2, 4 h, secr/shed). The pool of endoH-resistant intracellular Env is too small to detect because the average time it takes Env to travel via the Golgi to the cell surface and be cleaved and shed or secreted is very short compared with the time the protein resides in the ER.
Altogether, we concluded that wild-type Env matured in the vaccinia virus T7 heterologous expression system similar as in SupT1 cells infected with HIV-1 (Das et al., 1999) and in HeLa or CHO cells infected with gp120 or gp160 expressing recombinant vaccinia viruses (Kieny et al., 1986; Land et al., 2003). As shown before (Land et al., 2003), oxidative folding of the gp120 subunit was similar to that of the gp160 precursor, except that gp120 folded slightly faster and that differences in mobility in SDS-PAGE of Env species are more evident for gp120 because of its smaller size. Maturation of Env is slow, but after 4 h the majority of wild-type Env reaches the native state (Earl et al., 1991; Otteken et al., 1996; Land et al., 2003). We therefore compared maturation of Env mutants after 4 h of chase (Figure 3A) and quantified the results.
Five Disulfide Bonds Are Dispensable for Folding of Env
The 0-h chase samples of all mutants migrated similarly to that of wild type in both reducing and nonreducing gels, indicating that the starting point of folding was indistinguishable for mutants and wild-type proteins (Figure 3A, 0-NR/0-R). After 4 h of chase, however, the Env mutants displayed many differences in maturation (Figure 3A, 4-NR/4-R/4-S).
|
The folding patterns of maturation competent mutants were not necessarily identical to those of wild type. In case of the 378–445 mutants, fewer molecules than wild type had reached the native state and had lost their signal peptide (Figure 3A, 4-NR/4-R). Shedding from the 598–604 mutants was poorer than from wild type (Figure 3A, gp160, 4-S), and mutants of the V1/V2 loops accumulated a folding intermediate that was almost undetectable for wild type (Figure 3A, 4-NR). Results were similar for gp160 and gp120. We therefore concluded that five disulfide bonds (119–205, 126–196, 131–157, 378–445, and 598–604) were not strictly essential for Env to reach a folded state similar to wild type, but rather improved folding yields and rates.
Mutants of the remaining five disulfide bonds displayed severe folding defects: of the single disulfide bond in the C1 region (54–74), of the two disulfide bonds in the C2 region (218–247 and 228–239), and of the single disulfide bonds at the stem of the V3 loop (296–331) and the V4 loop (385–418). Compared with wild type, signal peptide cleavage of these mutants was poor (Figure 3A, 4-R). Disulfide bond formation varied: the persistence of the reduced-like band that was already present at the end of the pulse, in mutants of the 296–331 and 385–418 disulfide bonds, indicated that few if any disulfide bonds were formed in the majority of these Env species (Figure 3A, 4-NR). In contrast, mutants of the 54–74 disulfide bond and, to a lesser extent, of the 228–239 and 218–247 disulfide bonds did reach a native-like band (Figure 3A, 4-NR). These mutants nevertheless failed to shed gp120 (Figure 3A, gp160, 4-S). Conversely, mutants of the 385–418 disulfide bond shed marginal amounts of gp120 into the chase medium (Figure 3A, gp160, S; Figure 3B, gp160, shedding).
None of the gp120 mutants formed aggregates, whether folding deficient or not (Figure 3A, gp120, 4-NR). In case of the gp160 mutants, high-molecular-weight species were present after 4 h of chase, which represent aggregates of trimeric and larger size (Figure 3A, gp160, 4-NR/R). Aggregation was modest (never exceeding 13% of the signal; Figure 3B), but was more apparent for mutants that were maturation incompetent (54–74, 218–247, 228–239, 296–331, and 385–418) or displayed slower maturation (378–445). The single mutants C598A and C604A produced more of these high-molecular-weight species than the corresponding double mutant because of their odd number of cysteines, which leaves one free for intermolecular bonding.
Aggregates were fully reduced in the reducing gels (Figure 3A, gp160, 4-R), revealing that Env from lysates was fully endoH sensitive after 4 h. On the one hand this implies that the travel of the folding competent mutants through the Golgi and subsequent surface expression and shedding were similarly as fast as for wild type, because endoH resistance develops only as a result of glycan modifications in the Golgi. On the other hand, the lack of shedding (Figure 3A, gp160, 4-S) and of endoH resistance of the mutants with folding problems (Figure 3A, gp160, 4-R) implied that they where strictly retained in the ER, as we anticipated. Still, for some of these mutants gp120 was released from cells more by secretion than by shedding from gp41, in particular for the gp120s from the 54–74 mutants, for C247A, and for C296/331A (Figure 3A, gp120, 4-S). Retention in the ER hence was less stringent for gp120 than for gp160, possibly because gp120 is a monomeric soluble protein whereas gp160 must trimerize and is a membrane protein (Singh et al., 1990). We concluded that elimination of any of the 54–74, 218–247, 228–239, 296–331, or 385–418 disulfide bonds led to severe folding deficiencies in the ER. We categorized them as class 1 disulfide bonds, as opposed to class 2 disulfide bonds, which were expendable for productive folding.
Folding patterns were almost identical for gp120 and gp160, but the resolution was better for gp120, because of its smaller size (Figures 2 and 3A). We therefore used gp120 constructs for all mutants except for the 598–604 mutants (because these mutations reside in the gp41 domain), to analyze the effects of disulfide deletions on folding kinetics of Env in a time-course pulse-chase assay from 0 to 4 h (Figure 4). For wild-type gp120, the half-time of folding (of reaching the native state) and of signal peptide cleavage was >30 min (Figure 4, gp120 wt), consistent with our previous results (Land et al., 2003). Gp160 was somewhat slower with a half-time of folding and signal peptide cleavage of >60 min. Note that the gain in signal between the pulse and the first chase sample reflects elongation and completion of ribosome-bound nascent chains during the first 15 min of chase.
|
The extent to which folding intermediates of a mutant obtained disulfide bonds, correlated in most cases with their recognition by signal peptidase: the more oxidized, the better signal peptide cleavage (Figure 4, nonreducing vs. reducing). One exception was C385/418A: Env obtained few if any disulfide bonds, whereas after 4 h of chase the signal peptide was removed from a substantial portion of the mutant molecules. Before the signal peptide can be cleaved, Env must undergo some disulfide bond formation (Land et al., 2003). The pool of Env molecules that had their signal peptides removed in the case of the 385–418 mutants therefore is likely to correspond with the pool that displays some oxidation. Another exception was C54/74A, which showed the opposite: it reached an oxidized "endpoint" that migrated in nonreducing gels only slightly slower than native Env, yet signal peptide cleavage was minimal even at later chase times, suggesting that formation of the 54–74 disulfide bond in particular is important for subsequent signal peptide removal.
In general, ER client proteins that fail to fold correctly eventually are disposed of, for instance by a process that is referred to as ER-associated degradation (ERAD; Meusser et al., 2005). Signals decreased only slightly with time for both wild type and most mutants irrespective of their class (Figure 4), which suggested that even the most defective mutants were poor substrates for degradation. The mutants that were secreted, like those of the 54–74 disulfide bond, of course were expected to be lost from the cell lysates (Figure 3A, gp120, 54–74, 4-S).
Two Disulfide Bonds of Env Are Expendable for Function in the Virus
Because all disulfide bonds of Env are strictly conserved, we expected that the disulfide bonds that were expendable for folding would be essential for viral functions of Env instead. To examine the importance of particular disulfide bonds for viral fitness, we analyzed the replication of HIV-1 bearing wild-type or mutant Env proteins in cell cultures. Virus stocks were generated in C33A cells and SupT1 cells were infected with equivalent amounts of virus, as determined by CA-p24 content. Virus spread was monitored for 14 d by CA-p24 ELISA (see Figure 5A). In an independent assay we established whether the double cysteine mutants could mediate viral infectivity: the 50% tissue culture infectious dose (TCID50) in SupT1 cells was calculated by endpoint dilution (see Figure 5B).
|
Mutants of the 126–196, 131–157, or 598–604 disulfide bonds were folding competent, but the corresponding mutant viruses did not replicate (Figure 5A). These disulfide bonds of Env indeed were essential for function rather than for folding. Contrary to our expectations, the remaining two disulfide bonds were required neither for folding nor for function: virus lacking the 119–205 or the 378–445 disulfide bonds in Env sustained viral infectivity (Figure 5B) and replicated efficiently in SupT1 T-cells, as did the C378A mutant (Figure 5A). Still, the corresponding single mutants of 119–205, C119A and C205A, did not replicate, whereas C445A replicated markedly poorer than C378/445A, suggesting that an odd number of cysteines in gp120 is disadvantageous for virus replication. Also in primary T-cells, natural host for HIV, the C119/205A and C378/445A double mutants replicated efficiently (data not shown). We therefore subdivided class 2 disulfide bonds into class 2a: expendable for folding, but essential for the virus (126–196, 131–157, and 598–604) and class 2b: expendable for both folding and function (119–205 and 378–445).
Lack of infectivity and replication of class 2a mutant viruses could result from dysfunctional Env on virions or from their absence in viral particles. We therefore measured incorporation of Env in virus particles by gp120 ELISA on purified virus fractions and found that the molar Gag-to-Env ratio in the wild-type virus sample was 100:1 (data not shown). This corresponds to an average of 4–8 trimeric Env spikes per virus particle, which agrees with previously published results (O'Brien et al., 1994; Chertova et al., 2002). As expected, considerable amounts of C119/205A Env spikes were incorporated onto virions (80% of wild-type levels; Figure 5C). Incorporation of C378/445A spikes was lower (45% of wild-type levels; Figure 5C), which correlated with the reduced folding rate (Figures 3 and 4), delayed replication (Figure 5A), and slightly reduced viral infectivity (Figure 5B) compared with wild type. We did not anticipate to detect any incorporation of those mutants that could not fold and exit the ER. The residual incorporation we found for the class 1a mutants (Figure 5C) therefore likely represents background. Incorporation of the class 2a mutants that did fold but failed to replicate, C126/196A, C131/157A, and C598/604A, was in that same range. We concluded that the lack of viral replication of the class 2a mutants was due to absence or low abundance of functional Env spikes on the mutant virions.
Functional Env Mutants Are Similar But Not Identical to Wild Type
To ascertain whether the class 2b mutant viruses had different Env properties than the parental strain, we tested usage of the receptor (CD4) and coreceptor (CXCR4 for the LAI isolate) for C119/205A and C378/445A. CD4+-enriched lymphocytes were inoculated with equivalent amounts of mutant and wild-type virus, but viral entry, and hence propagation, was inhibited with limiting dilutions of either soluble CD4 (sCD4) or the CXCR4-binding compound AMD3100. Compared with the LAI parental strain and the C378/445A virus, less sCD4 was required to inhibit viral replication of the C119/205A mutant virus (Figure 6, top), which indicates that receptor usage was different for this mutant. The C378/445A mutant on the other hand displayed alterations in CXCR4 coreceptor usage, with lower concentrations of AMD3100 required to inhibit virus replication than for either the wild-type or C119/205A viruses (Figure 6, bottom).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Classification of the 10 Disulfide Bonds of Env
Because all disulfide bonds in Env are strictly conserved, we anticipated that most if not all would be crucial for Env folding, let alone for its function. In the case of influenza virus hemagglutinin (HA), for instance, all disulfide bonds are necessary for the protein to reach the native state, be it that one displayed a conditional phenotype (Braakman et al., unpublished data). To our surprise, we found that disruption of several disulfide bonds was compatible with Env folding and/or function. We categorized the 10 disulfide bonds of Env, based on the consequences of their elimination for folding and function: class 1 comprises disulfide bonds whose elimination led to severe folding deficiencies: 54–74, 218–247, 228–239, and, as reported earlier, 296–331 (Tschachler et al., 1990; Freed and Risser, 1991; Travis et al., 1992) and 385–418 (Hemming et al., 1989; Tschachler et al., 1990; Bolmstedt et al., 1991; Lekutis et al., 1992). Because deletion of the latter disulfide bond was compatible with some residual viral function in spite of folding failure, we further distinguished class 1 into class 1b, which only contained 385–418, with residual viral function despite folding deficiency, and class 1a, which contained the other disulfide bonds in class 1. Likewise, we divided the disulfide bonds whose elimination still allowed proper folding into class 2a, containing the disulfide bonds that were essential for viral function despite their dispensability for folding, 126–196, 131–157, and 598–604, and class 2b, containing the disulfide bonds that were expendable for both folding and function, 119–205 and 378–445 (Figure 7).
|
). The smears therefore may represent heterogeneous but "genuine" folding intermediates. Altogether, we cannot pinpoint a single obligatory "initiation" disulfide bond that must form before any of the others can. It was therefore impossible to delineate a straightforward folding pathway whereby formation of one disulfide bond was essential for formation of the next. Instead, we found evidence that different modules of Env fold independently of each other.
For one, gp120 folding appeared largely independent of gp41, because mutations in 598–604 did not affect oxidative folding of the gp120 moiety in gp160 and because folding of gp120 alone was almost identical to gp160. Likewise, folding of the V1/V2 region seemed to be independent. All mutants of the V1/V2 region reached a native state with rates similar to wild type, except that a distinct folding intermediate was more prominently present. This suggests that the mutations did change the kinetics of folding or the folding path, but did not interfere with folding of Env as a whole. This may come expected, because Env can function even in the absence of the complete V1/V2 loops (Wyatt et al., 1995; Stamatatos et al., 1998).
Quality Control of Env
Current models of ER quality control surveillance (Ellgaard and Helenius, 2003) predict that only correctly folded proteins can exit the ER and that ER client proteins that fail to fold aggregate (Kopito and Sitia, 2000) and/or are cleared from the ER lumen via ERAD, by retrotranslocation into the cytosol and subsequent proteasomal degradation (Meusser et al., 2005). We found that the Env mutants did not always obey these "rules."
None of the mutants appeared prone to substantial aggregation. The gp160 mutants did display some aggregation, most prominently in the case of poorly folding mutants, but aggregates were not detected in the corresponding gp120 mutants. The class 1 mutant molecules, at the same time, seemed poor ERAD substrates, because their decrease was similar to that of the class 2 mutants. This could mean that the ER quality control mechanisms either failed to "brand" them as misfolded or that they are recognized as misfolded, but reside in large Env-chaperone complexes, too complex or bulky to be retrotranslocated.
ER quality control of the class 2a mutants failed from another perspective. The "approval" to exit from the ER hopefully would warrant a protein's functionality, but the class 2a mutants were of no use to the virus despite their folding success. In agreement, defective cell entry of C131G and C196V single mutants (Tschachler et al., 1990) and reduced gp160 precursor cleavage and replication deficiency of 598–604 mutants (Syu et al., 1991; Dedera et al., 1992) had been reported before. We found that these mutants were not incorporated into virions. This implies that for the class 2a mutants quality control is exerted at the level of virion assembly/budding and not at the level of exit from the ER, because shedding, and hence cell surface appearance were similar to wild type.
Dispensability of Disulfide Bonds for Function of Env
The fate of the class 1b mutant is perhaps the most puzzling: how can we reconcile misfolding with marginal viral function of Env? The minimal shedding of the 385–418 mutants suggests that a minority of these mutant Env molecules did manage to fold "correctly," exit the ER, appear on the cell surface and therefore did get incorporated into virus particles as functional spikes, which would account for the residual infectivity and replication. Altogether, the folding defects of the class 1b mutants appeared severe but not insurmountable. Accordingly, we found that in vitro evolution of the C385/418A mutant easily gave rise to Env revertants with increased replication capacity (Sanders et al., 2008a; Sanders et al., 2008b).
The dispensability of the class 2b disulfide bonds (119–205 and 378–445) raises the question why they have been conserved. Loss of disulfide bonds in the course of evolution would involve Env variants with odd cysteines, because simultaneous pairwise elimination of cysteines is unlikely to occur. We found that the single cysteine mutants of the 119–205 disulfide bond replicated with lower efficiency than the corresponding double mutants. Perhaps a single cysteine intermediate is too large an obstacle on the evolutionary path toward Env variants lacking this disulfide bond, which would rule out its occurrence. This argument does not hold for the 378–445 disulfide bond, however, because the C378A single mutant replicated with efficiency similar to the double mutant.
Alternatively, the importance of these two disulfide bonds becomes manifest only during natural infection, when Env and the virus particles at large are challenged by the host's immune system. The differences in (co-)receptor usage suggest that the architecture or flexibility of these mutant Env molecules was slightly altered compared with wild type. The class 2b disulfide bonds may thus contribute to the integrity of the outer antigenic shield that disguises the immunosusceptible core of Env.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
These authors contributed equally to this work. 
Present addresses: 
Department of Biochemistry and Biophysics, University of California, San Francisco, CA 94158-2517;
|| Crucell, PO Box 2048, 2301 CA Leiden, The Netherlands;
¶ Department of Immunology, Erasmus MC, Dr. Molewaterplein 50, 3015 GE Rotterdam, The Netherlands.
Address correspondence to: Ineke Braakman (I.Braakman{at}uu.nl)
Abbreviations used: endoH, endoglycosidase H; Env, envelope glycoprotein; ER, endoplasmic reticulum; HIV, human immunodeficiency virus; PBMC, peripheral blood mononuclear cell.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Barbouche, R., Miquelis, R., Jones, I. M., and Fenouillet, E. (2003). Protein-disulfide isomerase-mediated reduction of two disulfide bonds of HIV envelope glycoprotein 120 occurs post-CXCR4 binding and is required for fusion. J. Biol. Chem 278, 3131–3136.
Bolmstedt, A et al. (1991). Effects of mutations in glycosylation sites and disulphide bonds on processing, CD4-binding and fusion activity of human immunodeficiency virus envelope glycoproteins. J. Gen. Virol 72, 1269–1277.
Braakman, I., Hoover-Litty, H., Wagner, K. R., and Helenius, A. (1991). Folding of influenza hemagglutinin in the endoplasmic reticulum. J. Cell Biol 114, 401–411.
Chen, B., Vogan, E. M., Gong, H., Skehel, J. J., Wiley, D. C., and Harrison, S. C. (2005). Structure of an unliganded simian immunodeficiency virus gp120 core. Nature 433, 834–841.[CrossRef][Medline]
Chertova, E et al. (2002). Envelope glycoprotein incorporation, not shedding of surface envelope glycoprotein (gp120/SU), Is the primary determinant of SU content of purified human immunodeficiency virus type 1 and simian immunodeficiency virus. J. Virol 76, 5315–5325.
Das, A. T., Land, A., Braakman, I., Klaver, B., and Berkhout, B. (1999). HIV-1 evolves into a nonsyncytium-inducing virus upon prolonged culture in vitro. Virology 263, 55–69.[CrossRef][Medline]
Dedera, D., Gu, R. L., and Ratner, L. (1992). Conserved cysteine residues in the human immunodeficiency virus type 1 transmembrane envelope protein are essential for precursor envelope cleavage. J. Virol 66, 1207–1209.
Doms, R. W., and Moore, J. P. (2000). HIV-1 membrane fusion: targets of opportunity. J. Cell Biol 151, F9–F14.
Earl, P. L., Moss, B., and Doms, R. W. (1991). Folding, interaction with GRP78-BiP, assembly, and transport of the human immunodeficiency virus type 1 envelope protein. J. Virol 65, 2047–2055.
Eckert, D. M., and Kim, P. S. (2001). Mechanisms of viral membrane fusion and its inhibition. Annu. Rev. Biochem 70, 777–810.[CrossRef][Medline]
Ellgaard, L., and Helenius, A. (2003). Quality control in the endoplasmic reticulum. Nat. Rev. Mol. Cell Biol 4, 181–191.[CrossRef][Medline]
Freed, E. O., and Risser, R. (1991). Identification of conserved residues in the human immunodeficiency virus type 1 principal neutralizing determinant that are involved in fusion. AIDS Res. Hum. Retroviruses 7, 807–811.[Medline]
Fuerst, T. R., Niles, E. G., Studier, F. W., and Moss, B. (1986). Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc. Natl. Acad. Sci. USA 83, 8122–8126.
Hemming, A., Bolmstedt, A., Flodby, P., Lundberg, L., Gidlund, M., Wigzell, H., and Olofsson, S. (1989). Cystein 402 of HIV gp 120 is essential for CD4-binding and resistance of gp 120 to intracellular degradation. Arch. Virol 109, 269–276.[CrossRef][Medline]
Jeeninga, R. E., Hoogenkamp, M., Armand-Ugon, M., de Baar, M., Verhoef, K., and Berkhout, B. (2000). Functional differences between the long terminal repeat transcriptional promoters of human immunodeficiency virus type 1 subtypes A through G. J. Virol 74, 3740–3751.
Kieny, M. P. et al. (1986). AIDS virus ENV protein expressed from a recombinant Vaccinia virus. Bio/Technology 4, 790–795.[CrossRef]
Kopito, R. R., and Sitia, R. (2000). Aggresomes and Russell bodies. Symptoms of cellular indigestion? EMBO Rep 1, 225–231.
Kwong, P. D., Wyatt, R., Robinson, J., Sweet, R. W., Sodroski, J., and Hendrickson, W. A. (1998). Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393, 648–659.[CrossRef][Medline]
Land, A., Zonneveld, D., and Braakman, I. (2003). Folding of HIV-1 envelope glycoprotein involves extensive isomerization of disulfide bonds and conformation-dependent leader peptide cleavage. FASEB J 17, 1058–1067.
Lekutis, C., Olshevsky, U., Furman, C., Thali, M., and Sodroski, J. (1992). Contribution of disulfide bonds in the carboxyl terminus of the human immunodeficiency virus type I gp120 glycoprotein to CD4 binding. J. Acquir. Immune Defic. Syndr 5, 78–81.[Medline]
Leonard, C. K., Spellman, M. W., Riddle, L., Harris, R. J., Thomas, J. N., and Gregory, T. J. (1990). Assignment of intrachain disulfide bonds and characterization of potential glycosylation sites of the type 1 recombinant human immunodeficiency virus envelope glycoprotein (gp120) expressed in Chinese hamster ovary cells. J. Biol. Chem 265, 10373–10382.
Li, Y., Luo, L., Thomas, D. Y., and Kang, C. Y. (1994). Control of expression, glycosylation, and secretion of HIV-1 gp120 by homologous and heterologous signal sequences. Virology 204, 266–278.[CrossRef][Medline]
Louwagie, J., Janssens, W., Mascola, J., Heyndrickx, L., Hegerich, P., van der Groen, G., McCutchan, F. E., and Burke, D. S. (1995). Genetic diversity of the envelope glycoprotein from human immunodeficiency virus type 1 isolates of African origin. J. Virol 69, 263–271.[Abstract]
Markovic, I., Stantchev, T. S., Fields, K. H., Tiffany, L. J., Tomic, M., Weiss, C. D., Broder, C. C., Strebel, K., and Clouse, K. A. (2004). Thiol/disulfide exchange is a prerequisite for CXCR4-tropic HIV-1 envelope-mediated T-cell fusion during viral entry. Blood 103, 1586–1594.
Matthias, L. J., and Hogg, P. J. (2003). Redox control on the cell surface: implications for HIV-1 entry. Antioxid. Redox Signal 5, 133–138.[CrossRef][Medline]
Meusser, B., Hirsch, C., Jarosch, E., and Sommer, T. (2005). ERAD: the long road to destruction. Nat. Cell Biol 7, 766–772.[CrossRef][Medline]
Moore, J. P., and Ho, D. D. (1993). Antibodies to discontinuous or conformationally sensitive epitopes on the gp120 glycoprotein of human immunodeficiency virus type 1 are highly prevalent in sera of infected humans. J. Virol 67, 863–875.
Moulard, M., and Decroly, E. (2000). Maturation of HIV envelope glycoprotein precursors by cellular endoproteases. Biochim. Biophys. Acta 1469, 121–132.[Medline]
O'Brien, W. A., Mao, S. H., Cao, Y., and Moore, J. P. (1994). Macrophage-tropic and T-cell line-adapted chimeric strains of human immunodeficiency virus type 1 differ in their susceptibilities to neutralization by soluble CD4 at different temperatures. J. Virol 68, 5264–5269.
Otteken, A., Earl, P. L., and Moss, B. (1996). Folding, assembly, and intracellular trafficking of the human immunodeficiency virus type 1 envelope glycoprotein analyzed with monoclonal antibodies recognizing maturational intermediates. J. Virol 70, 3407–3415.[Abstract]
Peden, K., Emerman, M., and Montagnier, L. (1991). Changes in growth properties on passage in tissue culture of viruses derived from infectious molecular clones of HIV-1LAI, HIV-1MAL, and HIV-1ELI. Virology 185, 661–672.[CrossRef][Medline]
Sanders, R. W., Venturi, M., Schiffner, L., Kalyanaraman, R., Katinger, H., Lloyd, K. O., Kwong, P. D., and Moore, J. P. (2002). The mannose-dependent epitope for neutralizing antibody 2G12 on human immunodeficiency virus type 1 glycoprotein gp120. J. Virol 76, 7293–7305.
Sanders, R. W. et al. (2008a). The carbohydrate at asparagine 386 on HIV-1 gp120 is not essential for protein folding and function but is involved in immune evasion. Retrovirology 5, 10.[CrossRef][Medline]
Sanders et al. (2008b). Evolution rescues Folding of HIV-1 envelope glycoprotein gp120 lacking a conserved disulfide bond. Mol. Biol. cell in press.
Singh, I., Doms, R. W., Wagner, K. R., and Helenius, A. (1990). Intracellular transport of soluble and membrane-bound glycoproteins: folding, assembly and secretion of anchor-free influenza hemagglutinin. EMBO J 9, 631–639.[Medline]
Stamatatos, L., Wiskerchen, M., and Cheng-Mayer, C. (1998). Effect of major deletions in the V1 and V2 loops of a macrophage-tropic HIV type 1 isolate on viral envelope structure, cell entry, and replication. AIDS Res. Hum. Retroviruses 14, 1129–1139.[Medline]
Stein, B. S., and Engleman, E. G. (1990). Intracellular processing of the gp160 HIV-1 envelope precursor. Endoproteolytic cleavage occurs in a cis or medial compartment of the Golgi complex. J. Biol. Chem 265, 2640–2649.
Syu, W. J., Lee, W. R., Du, B., Yu, Q. C., Essex, M., and Lee, T. H. (1991). Role of conserved gp41 cysteine residues in the processing of human immunodeficiency virus envelope precursor and viral infectivity. J. Virol 65, 6349–6352.
Travis, B. M., Dykers, T. I., Hewgill, D., Ledbetter, J., Tsu, T. T., Hu, S. L., and Lewis, J. B. (1992). Functional roles of the V3 hypervariable region of HIV-1 gp160 in the processing of gp160 and in the formation of syncytia in CD4+ cells. Virology 186, 313–317.[CrossRef][Medline]
Tschachler, E., Buchow, H., Gallo, R. C., and Reitz, M.S.J. (1990). Functional contribution of cysteine residues to the human immunodeficiency virus type 1 envelope. J. Virol 64, 2250–2259.
van Anken, E., and Braakman, I. (2005). Versatility of the endoplasmic reticulum protein folding factory. Crit. Rev. Biochem. Mol. Biol 40, 191–228.[CrossRef][Medline]
Walter, P., and Johnson, A. E. (1994). Signal sequence recognition and protein targeting to the endoplasmic reticulum membrane. Annu. Rev. Cell Biol 10, 87–119.[CrossRef][Medline]
Wyatt, R., Moore, J., Accola, M., Desjardin, E., Robinson, J., and Sodroski, J. (1995). Involvement of the V1/V2 variable loop structure in the exposure of human immunodeficiency virus type 1 gp120 epitopes induced by receptor binding. J. Virol 69, 5723–5733.[Abstract]
Wyatt, R., and Sodroski, J. (1998). The HIV-1 envelope glycoproteins: fusogens, antigens, and immunogens. Science 280, 1884–1888.
This article has been cited by other articles:
|
|
 |
|
  R. W. Sanders, S.-T. D. Hsu, E. van Anken, I. M. Liscaljet, M. Dankers, I. Bontjer, A. Land, I. Braakman, A. M.J.J. Bonvin, and B. Berkhout Evolution Rescues Folding of Human Immunodeficiency Virus-1 Envelope Glycoprotein GP120 Lacking a Conserved Disulfide Bond Mol. Biol. Cell, November 1, 2008; 19(11): 4707 - 4716. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||
) and oligomannose/hybrid type (
) as well as the N-terminus (N) and C-terminus (C) are indicated. (B) Ribbon representation of the crystal structure of gp120 (Kwong et al., 1998
