Skip to main content
NCBI home page
Glycobiology logoLink to Glycobiology
. 2009 Nov 17;20(3):280–286. doi: 10.1093/glycob/cwp184

Antibodies against Manα1,2-Manα1,2-Man oligosaccharide structures recognize envelope glycoproteins from HIV-1 and SIV strains

Robert J Luallen 2, Caroline Agrawal-Gamse 3, Hu Fu 2, David F Smith 4, Robert W Doms 3, Yu Geng 2,1
PMCID: PMC2815653  PMID: 19920089

Abstract

Design of an envelope glycoprotein (Env)-based vaccine against human immunodeficiency virus type-1 (HIV-1) is complicated by the large number of N-linked glycans that coat the protein and serve as a barrier to antibody-mediated neutralization. Compared to normal mammalian glycoproteins, high-mannose-type glycans are disproportionately represented on the gp120 subunit of Env. These N-glycans serve as a target for a number of anti-HIV molecules that bind terminal α1,2-linked mannose residues, including lectins and the monoclonal antibody 2G12. We created a Saccharomyces cerevisiae glycosylation mutant, Δmnn1Δmnn4, to expose numerous terminal Manα1,2-Man residues on endogenous hypermannosylated glycoproteins in the yeast cell wall. Immunization of rabbits with whole cells from this mutant induced antibodies that bound to a broad range of Env proteins, including clade A, B, and C of HIV and simian immunodeficiency virus (SIV). The gp120 binding activity of these immune sera was due to mannose-specific immunoglobulin, as removal of high-mannose glycans and α1,2-linked mannoses from gp120 abrogated serum binding. Glycan array analysis with purified IgG demonstrated binding mainly to glycans with Manα1,2-Manα1,2-Man trisaccharides. Altogether, these data demonstrate the immunogenicity of exposed polyvalent Manα1,2-Manα1,2-Man structures on the yeast cell wall mannan and their ability to induce antibodies that bind to the HIV Env protein. The yeast strain and sera from this study will be useful tools for determining the type of mannose-specific response that is needed to develop neutralizing antibodies to the glycan shield of HIV.

Keywords: 2G12, high-mannose glycans, human immunodeficiency virus, Saccharomyces cerevisiae

Introduction

One of the main targets for a prophylactic HIV-1 vaccine is the gp120 subunit of Env, the major viral protein exposed on the surface of the virion. An effective mechanism that HIV uses to avoid antibody neutralization is to coat gp120 with a glycan shield composed of numerous N-linked oligosaccharides (Burton et al. 2004). These glycans are added by the host glycosylation machinery and are therefore considered ‘self’ glycans, allowing gp120 to avoid immunological surveillance and protect highly neutralizing peptide epitopes (Wei et al. 2003). However, the glycans that coat gp120 contain an abundance of partially processed high-mannose-type oligosaccharides, mainly Man8−9GlcNAc2, that potentially form clusters on the glycoprotein (Pashov et al. 2007). The presence of numerous high-mannose-type glycans on mammalian glycoproteins is unusual (Balzarini 2007), making the glycan shield a potential target for virus neutralization. Indeed, several mannose-binding molecules have been found to interact with gp120 and broadly neutralize HIV-1 infection in vitro, including the potent anti-HIV lectins cyanovirin N (CV-N) and griffithsin (GFRT) (Balzarini 2007). In addition, one of the few broadly neutralizing monoclonal antibodies (MAbs) isolated from HIV-infected patients, 2G12, binds to terminal α1,2-linked mannose residues, structures only found on high-mannose-type glycans (Sanders et al. 2002; Scanlan et al. 2002; Trkola et al. 1996). Multivalency is a key factor for these anti-HIV molecules, with CV-N and GFRT existing as dimers that can bind four α1,2-linked mannose residues and 2G12 having an unusual VH domain-exchanged antibody structure that is believed to create multiple, distinct mannose-binding sites (Balzarini 2007; Calarese et al. 2003). These multivalent proteins exhibit broad and potent neutralization capacity likely due to their interactions with clusters of high-mannose glycans on gp120 (Wei et al. 2003).

As a result, clusters of high-mannose glycans have become a viable and attractive target for HIV-1 vaccine design. Additionally, targeting these structures has potential for anti-HIV-1 therapy, as any resistance developed to mannose-binding molecules by the virus would likely require deletion of N-glycan sites, potentially exposing neutralizing epitopes (Balzarini 2005). Unfortunately, immunizations using different forms of HIV Env have failed to induce neutralizing antibodies to the glycan shield, despite the presence of clusters of high-mannose glycans on gp120 (Burton et al. 2004; Letvin 2006). This is likely due to the poor immunogenicity of high-mannose glycans, immunodominant epitopes on the gp120 polypeptide, and the presence of other carbohydrate types, like hybrid- and complex-type glycans. To overcome these problems, multivalent immunogens have been designed to array synthetic oligomannose glycans on various molecular scaffolds. A few of these immunogens use a tetramannose glycan that mimics the D1 arm of Man9GlcNAc2 (Astronomo et al. 2008; Wang et al. 2008), which contains an exposed Manα1,2-Manα1,2 structure, a core determinant on gp120 for binding by CV-N, GFRT, and 2G12 (Balzarini 2007; Calarese et al. 2003). Clustering of this core determinant on a scaffold has the potential to improve immunogenicity and elicit antibodies that crossreact with gp120. Yet, while studies have demonstrated that these synthetic tetramannose glycans can bind 2G12 (Calarese et al. 2005; Wang et al. 2008), they have yet to elicit mannose-specific antibodies that crossreact with gp120 (Astronomo et al. 2008).

Here, we describe an alternate immunogen with multivalent Manα1,2-Man1,2 structures. The yeast Saccharomyces cerevisiae expresses numerous cell wall glycoproteins with N-linked glycans that contain up to 200 mannose residues per glycan, called polymannose glycans. In wild-type (WT) S. cerevisiae, the extended side chain residues of polymannose glycans are tightly packed oligosaccharide arms attached to an α1,6-linked mannose backbone, resulting in a majority of mannose tetrasaccharides with the structure Manα1,3-Manα1,2-Manα1,2-Manα1,6 (Figure 1A, top) (Dean 1999). We created an S. cerevisiae double mutant to remove the terminal α1,3-Man and phosphomannose residues by deleting the MNN1 and MNN4 genes, respectively, resulting in polymannose glycans with numerous exposed Manα1,2-Manα1,2-Man structures (Figure 1A, bottom). While Δmnn1Δmnn4 yeast and glycoproteins from this strain failed to bind 2G12, immunizations with whole DM yeast induced Manα1,2-Manα1,2-Man-specific antibodies that crossreacted with a broad range of Env glycoproteins from HIV-1 and SIV.

Fig. 1.

Fig. 1

Open in a new tab

N-Glycans produced in S. cerevisiae strains. (A) Depiction of expected structures of polymannose-type N-glycans produced in wild-type and Δmnn1Δmnn4 (DM) yeast strains, with the indicated type and linkage of each monosaccharide. The shaded boxes represent repeated oligosaccharide units, with an average of ten units per N-glycan. Expected structures of N-glycans from WT and DM yeast are based on previous studies of WT yeast and the Δmnn1 mutant (Ballou 1990; Dean 1999). Man8GlcNAc2, an important glycan in the 2G12 epitope, is depicted with the D1 and D3 arms. (B) Whole yeast cell ELISA testing anti-α1,3-Man (left) and 2G12 (right) binding to wild type, DM, and TM yeast coated at 2.5 × 106 cells/well. (C) Western blots testing anti-Gp38, Ecm33, α1,3-Man, and 2G12 binding to 4.8 μg of total cell lysate from WT, DM, and TM yeast.

Results

Mutation of S. cerevisiae to expose Manα1,2-Manα1,2-Man trisaccharides

In a previous study, we created two S. cerevisiae strains through mating and sporulation of single open reading frame knockout strains, with verification of each gene knockout confirmed by PCR (Luallen et al. 2008). A triple mutant (TM) yeast strain deleted in the OCH1, MNN1, and MNN4 genes expressed N-glycans that were strictly Man8GlcNAc2, and a yeast double mutant (DM) strain deleted in MNN1 and MNN4 expressed polymannose glycans with numerous exposed Manα1,2-Manα1,2-Man trisaccharide structures (Figure 1A). The focus of this study is on the DM strain, Δmnn1Δmnn4. The expected N-glycan structure from DM yeast is based on previous studies of the Δmnn1 mutation that showed a lack of terminal α1,3-Man on the polymannose side chain and the core Man8GlcNAc2 (Ballou 1990; Dean 1999). Likewise, studies have shown that the Δmnn4 mutation helps prevent the variable addition of phosphomannose residues, which are strong immunological determinants (Ballou 1990). Aside from PCR verification of each gene knockout, we confirmed the phenotypic loss of terminal Manα1,3 residues in DM yeast by whole cell ELISA. A terminal α1,3-Man-specific antibody did not bind to whole DM yeast despite strong crossreaction with WT yeast (Figure 1B, left). Surprisingly, despite the exposure of numerous terminal Manα1,2-Man structures upon MNN1 deletion, whole DM yeast did not bind to the MAb 2G12, while whole TM yeast expressing strictly Man8GlcNAc2 did bind to 2G12 (Figure 1B, right).

To further confirm the phenotypic traits of polymannose glycans in DM yeast, we conducted Western blotting on two endogenous yeast glycoproteins, Gp38 and Ecm33, that are associated with the cell wall. Both Gp38 and Ecm33 are heavily glycosylated, with 15 and 13 putative N-linked sites, respectively (Luallen et al. 2008). Both proteins migrated in SDS–PAGE at greater than 200 kDa in WT and DM yeast, compared to 90–110 kDa when expressed with strictly Man8 in TM yeast (Figure 1C). This suggests that greater than 100 kDa of molecular mass was added as side chain polymannose when the proteins were expressed in WT and DM yeast. Additionally, Western blotting of total yeast cell lysate showed that only proteins from WT yeast, but not DM or TM yeast, could bind to an α1,3-Man-specific antibody (Figure 1C). This confirms the loss of α1,3-linked Man caps on the side chain polymannose and core high-mannose glycans. Finally, 2G12 did not significantly bind to any specific yeast glycoproteins expressed in DM yeast by Western blot, while 2G12 bound to several glycoproteins expressed in TM yeast, two of which were previously shown to be Gp38 and Ecm33 at 90 and 110 kDa, respectively (Figure 1C, bottom) (Luallen et al. 2008). The lack of 2G12 binding to whole DM yeast and DM-expressed glycoproteins may be a consequence of the structural specificity of the 2G12 epitope, a conformational epitope composed of up to three Man8−9GlcNAc2 glycans.

Immunogenicity of the modified yeast glycans and serum crossreactivity with HIV-1 Env glycoproteins

In yeast, numerous highly glycosylated proteins are localized to the exterior portion of the cell wall, called the mannan layer. Based on the exposed polyvalent Manα1,2-Manα1,2-Man structures expected on the DM yeast mannan layer, we immunized rabbits’ with heat-killed, whole DM cells to generate α1,2-Man-specific antibodies and tested the resulting sera for binding to HIV-1 gp120. Postimmune sera were initially screened by ELISA for binding to three gp120 proteins from clade B of HIV-1 (JR-FL, ADA, and HxB), and three of four rabbits showed moderate to strong antibody responses to gp120. The immune sera from these three rabbits, 1137, 1138, and 1139, were further tested against a panel of 13 Env glycoproteins from HIV-1 and SIV, all of which were produced in mammalian cells and purified under native conditions. All three rabbits showed time-dependent induction of gp120-binding antibodies over the course of the immunization schedule. Sera from rabbits 1137 and 1138 bound to the same 6 of 13 Env glycoproteins, when using an OD450 ≥ 0.5 as a cutoff (Figure 2A, top and middle). Serum from rabbit 1139 showed stronger and broader binding to all 13 tested Env glycoproteins, which included 8 from clade B, 2 from clade C, and 1 from clade A of HIV-1, and 2 from SIV (Figure 2A, bottom). The breadth of gp120 binding by serum from rabbit 1139 is stronger and broader than serum from a rabbit immunized with TM yeast, included as a point of reference (Figure 2B) (Luallen et al. 2008).

Fig. 2.

Fig. 2

Open in a new tab

Immune sera elicited with S. cerevisiae DM yeast crossreacted with gp120 glycoproteins from HIV-1 and SIV in ELISA. (A) Immune sera at weeks 0, 5, and 10 from rabbit 1137 (top), 1138 (middle), and 1139 (bottom) were tested by ELISA at 1:500 dilution against the indicated Env glycoproteins. (B) Comparison of ELISA binding by postimmune sera (week 10) from TM-immunized rabbit 1141 and DM-immunized rabbit 1139 against Env glycoproteins. All proteins were produced in mammalian cells. OD, optical density.

The titers of gp120 binding antibody in DM-induced sera at week 10 were further assessed against gp120 from HIV-1 strains JR-FL, R2, and R3A. Sera from 1137 and 1138 showed a similar binding curve with end points at approximately 1:2500 dilution for JR-FL and R3A and ≥ 1:12,500 for R2 (Figure 3). The titers of 1139 serum were approximately 5-fold higher than 1137 and 1138 against all three gp120s, except for 1138 serum tested against R2. Together, these results demonstrate that the genetically modified mannose oligosaccharides on DM yeast are immunogenic and that the elicited antibodies crossreact with Env glycoproteins from a broad range of HIV-1 and SIV strains.

Fig. 3.

Fig. 3

Open in a new tab

Titration of immune sera. Titers of gp120-specific IgG in rabbit serum samples were measured using purified JR-FL (top), R2 (middle), and R3A (bottom) in ELISA. Sera tested include DM-immunized rabbits 1137, 1138, and 1139 at week 10, and a positive control serum from a rabbit immunized with JR-FL gp120 (gp120 serum). The values are the averages ± standard errors of the means (error bars) of three independent experiments. All gp120 proteins were produced in 293-T cells. OD, optical density.

Antibody crossreactivity to HIV-1 gp120 is due to terminal α1,2-linked mannose residues on high-mannose glycans

To determine if the binding of DM-induced immune sera was via interaction with the carbohydrates on HIV-1 Env glycoproteins, we tested serum binding to PNGase F and mock-digested gp120 in parallel by Western blot. PNGase F cleaves any N-linked glycan between the asparagine residue of the polypeptide chain and the first GlcNAc residue of the glycan (Sanders et al. 2002). Serum from rabbit 1139 (week 10) showed binding to mock-digested gp120 from all four tested HIV-1 strains, R2, JR-FL, Du179, and R3A, but not to deglycosylated gp120 (Figure 4A). In contrast, a peptide-specific antibody against gp120 (anti-gp120) bound both glycosylated and deglycosylated Env at approximately 120 kDa and 60–70 kDa, respectively. We also tested sera from rabbits 1137 and 1138 (week 10) against mock-, PNGase F-, and Endo H-digested JR-FL and R2 gp120, and both immune sera showed binding to mock but not PNGase F- or Endo H-digested gp120 (data not shown). These results indicate that the DM-induced immune sera bind denatured gp120 glycoproteins and that the binding is dependent on N-linked glycans.

Fig. 4.

Fig. 4

Open in a new tab

Glycan binding analyses of DM-induced antibodies. (A) The indicated gp120 proteins were digested with PNGase F (PNF) or mock digested. Samples were run on SDS–PAGE gels for blotting with rabbit 1139 serum (top) and control anti-gp120 IgG (bottom). (B) The indicated gp120 proteins were digested with PNF, Endo H, and jack bean and A. saitoi mannosidases, with mock digestion using the same denatured conditions as PNGase F digestion but without the enzyme. Western blot was conducted using purified rabbit 1139 IgG (top) and control anti-gp120 IgG (bottom). (C) Glycan array analysis of DM-induced antibodies. Purified IgG from rabbit 1139 at week 0 and week 10, post-immune IgG from a WT yeast-immunized rabbit, and 2G12 were used to probe a microarray of 320 glycans. The glycan numbers (x-axis) correspond to the depicted high-mannose-type glycans. Group 1 contains synthetic glycans without terminal α1,2-linked mannose residues, group 2 with terminal Manα1,2-Man disaccharides, and group 3 with terminal Manα1,2-Manα1,2-Man trisaccharides. Group 4 contains natural high-mannose glycans with 2 GlcNAc residues at the base.

To further characterize the glycan specificity of the DM-induced sera in the context of gp120 binding, we tested purified IgG from rabbit 1139 serum (week 10) for binding to gp120 glycoproteins digested by a variety of glycosidases. Endo H cleaves high-mannose glycans at the β-1,4 linkage connecting the two GlcNAc residues; jack bean mannosidase cleaves terminal α1,6-, α1,3-, and α1,2-Man residues; and A. saitoi mannosidase cleaves only terminal α1,2-Man residues (Ichishima et al. 1981; Sanders et al. 2002). Digestion of gp120 resulted in migration at the expected relative masses as detected by anti-gp120 (Figure 4B, bottom). IgG from rabbit 1139 (week 10) detected only mock-digested gp120 from R2 and JR-FL (Figure 4B, top), though there was some residual binding to A. saitoi mannosidase-digested gp120; this may be a consequence of the free mannose produced by the enzyme inhibiting subsequent digestion (Ichishima et al. 1981). These results implicate the high-mannose N-glycans on gp120 as vital structures for the crossreactivity exhibited by the DM-elicited immune serum, and specifically implicate binding to terminal α1,2-linked mannose residues.

DM-induced antibodies preferentially bind to high-mannose glycans with exposed Manα1,2-Manα1,2-Man trisaccharides

To further characterize the glycan binding profile of DM-induced antibodies, purified IgG from rabbit 1139 (week 10) serum was analyzed for binding to 320 glycans on an array, and compared with the IgG from the corresponding prebleed (week 0), IgG from WT yeast-immunized rabbit sera, and MAb 2G12. The 16 mannose-containing glycans in the array were divided into four groups based on the linkage of terminal residues and on the source (natural versus synthetic) (Figure 4C). Purified antibodies from the 1139 prebleed and WT yeast post-immune sera showed less than 1000 relative fluorescence units (RFUs) for binding to any of the 16 glycans. By contrast, IgG from 1139 post-immune serum bound to six high-mannose glycans with RFUs ranging from 1549 to 31,709. The weakest binding was to synthetic glycan 190 containing two Manα1,2-Man disaccharide branches (group 2), while moderate to strong binding was seen against synthetic high-mannose glycans containing Manα1,2-Manα1,2-Man trisaccharides (group 3). Additionally, both 1139 IgG and 2G12 bound natural Man8GlcNAc2 with an intact D1 arm but not any other natural high-mannose glycan (group 4). Based on the higher RFUs to glycans containing terminal Manα1,2-Manα1,2-Man trisaccharides (group 3) compared to Manα1,2-Man disaccharides (group 2), the majority of binding by DM-induced sera to glycans on gp120 is likely due to crossreaction with the D1 arm of Man8−9GlcNAc2. This idea is further supported by the fact that 1139 IgG binding to glycan 189 (from the D1 arm) was almost 9-fold higher compared to glycan 191 (from the D2 arm) and greater than 2-fold higher compared to glycan 190 (the pentasaccharide from the D2 and D3 arm).

Discussion

Several lectins exhibit anti-HIV activity due to their ability to bind to Manα1,2-Manα1,2-Man structures present on the viral Env protein, and the broadly neutralizing MAb 2G12 likewise interacts with high-mannose structures on the gp120 subunit of Env. However, immunization with the HIV Env protein has consistently failed to elicit broadly neutralizing antibodies, suggesting that the glycan epitope recognized by these antiviral agents is not immunogenic. By manipulating the S. cerevisiae N-glycosylation pathway, we removed terminal α1,3-linked mannoses and exposed Manα1,2-Manα1,2-Man structures arrayed on the tightly packed side chain branches of polymannose-type glycans. Immunization with yeast bearing these structures induced production of Manα1,2-Manα1,2-Man-specific IgG against the DM yeast cell wall mannan that could crossreact with Env glycoproteins. Why in this context was it possible to elicit antibodies to polymannose glycans? Clustering of Manα1,2-Manα1,2-Man determinants may have improved immunogenicity, as numerous Manα1,2-Manα1,2-Man structures are absent in mammals (Balzarini 2007). Additionally, the foreign Manα1,2-Manα1,2-Manα1,6 trimannose found on the polymannose side chain branches was likely immunogenic, as seen in the 5- to 9-fold stronger binding by IgG to glycan 196 (containing a Manα1,2-Manα1,2-Manα1,6 structure) relative to glycans lacking similar polymannose side chain branches. The α1,6 linkage of this trimannose differs from that found on the D1 arm of high-mannose glycans, which may have made the whole structure foreign to mammals. Together, these factors may have improved the immunogenicity of the glycans resulting in the efficient induction of Manα1,2-Manα1,2-Man-specific IgG.

An important facet of DM yeast in this experiment was that 2G12 bound neither whole cells nor glycoproteins of the DM strain while 2G12 did bind to TM yeast expressing strictly Man8GlcNAc2. This introduces the idea that 2G12 binding is not a necessary facet for an immunogen to elicit mannose-specific antibodies that crossreact with HIV gp120. Rather, the clustering of Manα1,2-Manα1,2-Man structures may be enough, as this is the core determinant for a number of anti-HIV molecules, such as CV-N and GRFT. Yet, despite the creation of numerous scaffolds that can array high-mannose glycans to bind 2G12, only genetically modified yeast have been reported thus far to elicit mannose-specific antibodies that crossreact with gp120. The use of whole yeast to generate glycan-specific polyclonal sera is a well-established and known procedure, with different N-glycan mutants used to elicit α1,3-Man, α1,6-Man, and phosphomannose-specific antisera (Ballou 1990; Raschke et al. 1973). It remains to be determined why whole yeast cells can easily generate an immune response to mannose residues, which are generally poor immunogens; possibilities include the polyvalent nature of the yeast glycans or components of the yeast cell wall (mannan and β-glucan) acting as adjuvants.

Despite gp120 binding, the DM-induced sera failed to neutralize HIV-1 pseudovirus infection of TZM-bl cells, a result similar to our previous findings using TM yeast (Luallen et al. 2008). This may be due to a number of factors, including low affinity binding, low titer of mannose-specific antibodies (relative to total antibody titer) due to immunization with whole cells, and antibody binding to non-neutralizing epitopes on the glycan shield. Another possibility is that the high-mannose glycans present on gp120 to which antibodies are directed might be processed to more complex forms in the context of the native viral Env protein during the course of virus infection. Altogether, these results indicate that the epitope specificity of antibodies against high-mannose glycans can be quite complex and sensitive to fine structural differences. The specificity of an antibody could be influenced by a number of factors, including the number of sugar residues in a branch, linkages between sugar residues, and mannose clusters with different conformations. Taken together, it seems evident from this study and our previous work that eliciting antibodies that bind to high-mannose, carbohydrate-dependent epitopes on HIV-1 gp120 is not difficult, at least in the context of whole yeast immunogens. However, eliciting antibodies that recognize the subset of glycans that result in virus neutralization, such as those recognized by 2G12, is not trivial. The yeast strains we have developed, and the antibodies elicited by their use as immunogens, should be useful tools for understanding how antibodies can be elicited against high-mannose epitopes and may help us decipher the structural features needed to achieve a neutralizing antibody response against HIV.

Materials and methods

Materials

2G12 was purchased from Polymum Scientific (Vienna, Austria). Wild-type S. cerevisiae strain INVSc1 was obtained from Invitrogen (Carlsbad, CA). The AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH supplied the following reagents: SF162 gp120, 96ZM651 gp120, and SIVmac1A11 gp140. All other Env glycoproteins were produced as described previously (Luallen et al. 2009). The Δmnn1Δmnn4 double mutant strain (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 MNN1::kanMX4 MNN4::kanMX4) and polyclonal antibodies against the yeast glycoproteins Gp38 and Ecm33 were created previously (Luallen et al. 2008). Immune sera against terminal α1,3-linked mannoses were prepared as previously described with slight modification (Ballou 1990). Briefly, heat-killed INVSc1 yeast cells suspended in 0.9% NaCl at 3.0 × 107 cells/mL were used to immunize rabbits and the resulting sera were twice adsorbed to heat-killed Δmnn1 cells at 5 mL sera per 5 g (wet weight) cells in 50 mL of 0.9% NaCl.

Immunization

Four New Zealand white rabbits were immunized with whole, heat-killed Δmnn1Δmnn4 yeast using a modified protocol as previously described (Ballou 1990). Briefly, log-phase cells were suspended in sterile 0.9% NaCl at 1.0 × 109 cells/mL and heat-killed at 70°C or 1 h. Rabbits were injected in the marginal ear vein with 1 mL of yeast suspension three times a week for 9 weeks. Rabbits were bled at weeks 3, 5, and 8, three days after the last injection from the prior week, and at week 10, seven days after the final injection. ELISAs, whole-cell ELISA, and serum titration ELISAs were performed as previously described (Luallen et al. 2008).

Glycosidase digestion and Western blot

Deglycosylation of purified R2, JR-FL, Du179, and R3A gp120 was conducted using PNGase F, Endo H, jack bean mannosidase, and Aspergillus saitoi mannosidase as previously described (Luallen et al. 2008). Mock digestions were conducted under the same conditions as PNGase F digestion but without the enzyme. Samples were denatured by boiling for 5 min at 100°C in the Laemmli SDS sample buffer and loaded onto SDS–PAGE gels. For PNGase F Western blots, samples were loaded at 600 ng/lane for R2, JR-FL and Du179, and 1 μg/lane for R3A (for 1139 serum), and 50 ng gp120/lane for the control blot. For the glycosidase panel Western blot, samples were loaded at 300 ng gp120/lane for 1139 IgG (1 μg/mL) and 100 ng gp120/lane for anti-gp120 (100 ng/mL). 1139 IgG was purified from 1139 serum (week 10) using protein A-agarose (Invitrogen).

Glycan microarray

Antibody samples were tested for binding to a glycan array of synthetic and natural glycans printed on glass microscope slides as described previously (Luallen et al. 2008). This work was done by the Consortium for Functional Glycomics (CFG) Core H, Emory University. Printed array version 3.0 was tested with 70 μL of 20 μg/mL of protein A-agarose-purified IgG from rabbit 1139 at week 0 and week 10, and IgG from a control rabbit immunized with WT yeast. Printed array version 3.1 was tested with 70 μL of MAb 2G12 at 30 μg/mL.

Funding

The National Institutes of Health (AI058724 to Y.G., R01 40880 to R.W.D., T32 AI007632 to C.A.-G.) and the International AIDS Vaccine Initiative's Neutralizing Antibody Consortium.

Conflict of interest statement

Yu Geng, President and Chief Executive Officer of ProSci Incorporated, acknowledges a potential financial conflict of interest.

Glossary

Abbreviations

ELISA

enzyme-linked immunosorbent assay

Endo H

Endoglycosidase H

GlcNAc

N-acetyl-glucosamine

GNL

galanthus nivalis lectin

HRP

horseradish peroxidase

IgG

immunoglobulin G

Man

mannose

PNGase F

peptide-N-glycosidase F

SDS

sodium dodecyl sulfate

VH

heavy chain variable

References

  1. Astronomo RD, Lee HK, Scanlan CN, Pantophlet R, Huang CY, Wilson IA, Blixt O, Dwek RA, Wong CH, Burton DR. A glycoconjugate antigen based on the recognition motif of a broadly neutralizing human immunodeficiency virus antibody, 2G12, is immunogenic but elicits antibodies unable to bind to the self glycans of gp120. J Virol. 2008;82:6359–6368. doi: 10.1128/JVI.00293-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ballou CE. Isolation, characterization, and properties of Saccharomyces cerevisiae mnn mutants with nonconditional protein glycosylation defects. Methods Enzymol. 1990;185:440–470. doi: 10.1016/0076-6879(90)85038-p. [DOI] [PubMed] [Google Scholar]
  3. Balzarini J. Targeting the glycans of gp120: A novel approach aimed at the Achilles heel of HIV. Lancet Infect Dis. 2005;5:726–731. doi: 10.1016/S1473-3099(05)70271-1. [DOI] [PubMed] [Google Scholar]
  4. Balzarini J. Targeting the glycans of glycoproteins: A novel paradigm for antiviral therapy. Nat Rev Microbiol. 2007;5:583–597. doi: 10.1038/nrmicro1707. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Burton DR, Desrosiers RC, Doms RW, Koff WC, Kwong PD, Moore JP, Nabel GJ, Sodroski J, Wilson IA, Wyatt RT. HIV vaccine design and the neutralizing antibody problem. Nat Immunol. 2004;5:233–236. doi: 10.1038/ni0304-233. [DOI] [PubMed] [Google Scholar]
  6. Calarese DA, Lee HK, Huang CY, Best MD, Astronomo RD, Stanfield RL, Katinger H, Burton DR, Wong CH, Wilson IA. Dissection of the carbohydrate specificity of the broadly neutralizing anti-HIV-1 antibody 2G12. Proc Natl Acad Sci USA. 2005;102:13372–13377. doi: 10.1073/pnas.0505763102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Calarese DA, Scanlan CN, Zwick MB, Deechongkit S, Mimura Y, Kunert R, Zhu P, Wormald MR, Stanfield RL, Roux KH, et al. Antibody domain exchange is an immunological solution to carbohydrate cluster recognition. Science. 2003;300:2065–2071. doi: 10.1126/science.1083182. [DOI] [PubMed] [Google Scholar]
  8. Dean N. Asparagine-linked glycosylation in the yeast Golgi. Biochim Biophys Acta. 1999;1426:309–322. doi: 10.1016/s0304-4165(98)00132-9. [DOI] [PubMed] [Google Scholar]
  9. Ichishima E, Arai M, Shigematsu Y, Kumagai H, Sumida-Tanaka R. Purification of an acidic alpha-d-mannosidase from Aspergillus saitoi and specific cleavage of 1,2-alpha-d-mannosidic linkage in yeast mannan. Biochim Biophys Acta. 1981;658:45–53. doi: 10.1016/0005-2744(81)90248-5. [DOI] [PubMed] [Google Scholar]
  10. Letvin NL. Progress and obstacles in the development of an AIDS vaccine. Nat Rev Immunol. 2006;6:930–939. doi: 10.1038/nri1959. [DOI] [PubMed] [Google Scholar]
  11. Luallen RJ, Fu H, Agrawal-Gamse C, Mboudjeka I, Huang W, Lee FH, Wang LX, Doms RW, Geng Y. A yeast glycoprotein shows high-affinity binding to the broadly neutralizing human immunodeficiency virus antibody 2G12 and inhibits gp120 interactions with 2G12 and DC-SIGN. J Virol. 2009;83:4861–4870. doi: 10.1128/JVI.02537-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Luallen RJ, Lin J, Fu H, Cai KK, Agrawal C, Mboudjeka I, Lee FH, Montefiori D, Smith DF, Doms RW, et al. An engineered Saccharomyces cerevisiae strain binds the broadly neutralizing human immunodeficiency virus type 1 antibody 2G12 and elicits mannose-specific gp120-binding antibodies. J Virol. 2008;82:6447–6457. doi: 10.1128/JVI.00412-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Pashov A, Perry M, Dyar M, Chow M, Kieber-Emmons T. Defining carbohydrate antigens as HIV vaccine candidates. Curr Pharm Des. 2007;13:185–201. doi: 10.2174/138161207779313678. [DOI] [PubMed] [Google Scholar]
  14. Raschke WC, Kern KA, Antalis C, Ballou CE. Genetic control of yeast mannan structure. Isolation and characterization of mannan mutants. J Biol Chem. 1973;248:4660–4666. [PubMed] [Google Scholar]
  15. Sanders RW, Venturi M, Schiffner L, Kalyanaraman R, Katinger H, Lloyd KO, Kwong PD, Moore JP. The mannose-dependent epitope for neutralizing antibody 2G12 on human immunodeficiency virus type 1 glycoprotein gp120. J Virol. 2002;76:7293–7305. doi: 10.1128/JVI.76.14.7293-7305.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Scanlan CN, Pantophlet R, Wormald MR, Ollmann Saphire E, Stanfield R, Wilson IA, Katinger H, Dwek RA, Rudd PM, Burton DR. The broadly neutralizing anti-human immunodeficiency virus type 1 antibody 2G12 recognizes a cluster of alpha1→2 mannose residues on the outer face of gp120. J Virol. 2002;76:7306–7321. doi: 10.1128/JVI.76.14.7306-7321.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Trkola A, Purtscher M, Muster T, Ballaun C, Buchacher A, Sullivan N, Srinivasan K, Sodroski J, Moore JP, Katinger H. Human monoclonal antibody 2G12 defines a distinctive neutralization epitope on the gp120 glycoprotein of human immunodeficiency virus type 1. J Virol. 1996;70:1100–1108. doi: 10.1128/jvi.70.2.1100-1108.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Wang SK, Liang PH, Astronomo RD, Hsu TL, Hsieh SL, Burton DR, Wong CH. Targeting the carbohydrates on HIV-1: Interaction of oligomannose dendrons with human monoclonal antibody 2G12 and DC-SIGN. Proc Natl Acad Sci USA. 2008;105:3690–3695. doi: 10.1073/pnas.0712326105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Wei X, Decker JM, Wang S, Hui H, Kappes JC, Wu X, Salazar-Gonzalez JF, Salazar MG, Kilby JM, Saag MS, et al. Antibody neutralization and escape by HIV-1. Nature. 2003;422:307–312. doi: 10.1038/nature01470. [DOI] [PubMed] [Google Scholar]

Articles from Glycobiology are provided here courtesy of Oxford University Press

RESOURCES