Scientists report on the structure of the HIV-1 gp120 glycoprotein’s V1/V2 domain and how a broadly neutralizing antibody binds to the structure and penetrates through the glycan coat surrounding the virus. Data generated by the NIH-led consortium of researchers indicate that antibody recognition of the epitope involves just two glycans and one of the four strands that form the β-sheet V1/V2 domain.
The investigators, led by Peter D. Kwong, Ph.D., who heads the structural biology section of the NIAID’s Vaccine Research Center, report their findings in Nature in a paper titled “Structure of HIV-1 gp120 V1/V2 domain with broadly neutralizing antibody PG9.”
THe HIV-1 viral spike is the only target of neutralizing antibodies but has also evolved to escape antibody recognition and neutralization, the authors explain. The V1/V2 region of the gp120 component of the viral spike is critical to antibody evasion. The 50–90 residue domain that comprises V1/V2 contains two of the most variable portions of the virus. Roughly 1 in 10 residues of V1/V2 are N-glycosylated.
Despite the diversity and glycosylation of V1/V2, broadly neutralizing human antibodies targeting this region have been isolated from HIV-1 patients. Interestingly, these antibodies all share specificity for an N-linked glycan at residue 160 in V1/V2, bind preferentially to the assembled viral spike rather than monomeric gp120, and are sensitive to changes in V1/V2 and some V3 residues. Nevertheless, the authors note, V1/V2 has to date resisted atomic-level categorization.
The team has now succeeded in crystallizing two scaffolds of V1/V2 proteins complexed with a broadly neutralizing antibody known as PG9. One scaffold comprised the gp120 V1/V2 region from the CAP45 strain of HIV-1, which contains five sites of potential N-linked glycosylation. The other scaffold comprised the gp120 V1/V2 region from the ZM109 strain of HIV-1, with N-linked glycans at positions 160 and 173 and asparagine to alanine substitutions at four other potential N-linked sites.
Resulting crystallographic data indicated that V1/V2 folds as four anti-parallel β-strands (A-D), in which key structural features such as the hydrophobic core, connecting loops, and disulphide bonds cross between each of the four strands. The strands that make up the overall sheet structure contain primarily conserved residues and are fixed in place by inter-strand disulphide bonds and hydrogen bonding.
The two faces of the sheet, have different features, meanwhile. The concave face is glycan-free and hydrophobic, while the convex face is cationic, and houses the key, N-linked glycan 160. The structure also features two strand-connecting loops that stick out from the same end of the sheet, which are highly glycosylated and variable in sequence.
As a result of their studies, the researchers say that they can now redefine the V1 loop as those residues connecting strands A and B and the V2 loop as those residues between strands C and D. The V1 loops appears to be the most variable and ranges in length from about 10–30 residues, while the V2 loop contains at its beginning the tripeptide motif recognized by integrin α4β7, the gut homing receptor for HIV-1.
PG9 and the somatically related antibody PG16 typically recognize the assembled viral spike with higher affinity than monomeric gp120.
Electron microscopy images of PG9 complexed with monomeric gp120 and with V1/V2 indicated that the position of V1/V2 varies in monomeric gp120, whereas it is fixed in the V1/V2 envelope trimer spike.The key interaction between antibody PG9 and V1/V2 occurs with N-linked glycan, the authors continue. The antibody grasps the entire 160 glycan and reaches through the glycan shield to contact the protein-proximal N-acetylglucosamine. Its protruding third complementarity-determining region of the heavy chain (CDR H3) reaches through the glycan shield to contact the protein-proximal N-acetylglucosamine.
A total of 11 hydrogen bonds are buried within the PG9–glycan-160 interface, with G9 contacting 5 of the 7 saccharide moieties of the Man5GlcNAc2 glycan. Interactions also occur between PG9 and the N-linked glycan at other residues on the CAP45 and ZM109 scaffolds. Effecting substitutions that resulted in changes in V1/V2 glycans confirmed that the glycan at 160 is critical for Pg9 recognition, the team continues.
As well as glycan recognition, a strand in the CDR H3 of PG9 forms intermolecular parallel β-sheet-like hydrogen bonds to strand C of V1/V2. This strand is the most variable of the V1/V2 strands, “and this sequence-independent means of recognition probably allows for increased recognition breadth,” the team notes. Electrostatic interactions also occur between cationic residues on strand C and acidic residues on PG9.
The results up to this point indicated that the V1/V2-PG9 interaction encompasses much of the PG9/PG16 epitope and that the structural integrity of the epitope is sensitive to how the viral spike is assembled. In terms of the structures of the broadly neutralizing antibodies, analyses of unbound Fab structures from antibodies derived from different individuals indicated that they all showed anionic protruding CDR H3s, all displayed β-hairpins , and all appeared capable of penetrating an N-linked glycan shield to reach a cationic protein surface.
The overall data generated thus indicated that for both the CAP43 and ZM109 strain of gp120, the V1/V2 site recognized by PG9 consists primarily of two glycans and a strand. There are minor additional interactions with strand B and the B-C connecting loop, and the entire PG9-recognized surface of V1/V2 is contained within the B-C hairpin.
“The minimal nature of this epitope suggests that it might be easier to engineer and to present to the immune system than other, more complex, epitopes,” they write. “It seems that N-linked glycosylation at particular residues is selected as a means of immune evasion but that these same glycans ... can be recognized by very broadly neutralizing antibodies.”
Interestingly, the published findings tie in nicely with a recently published paper describing how a different broadly neutralizing HIV antibody also binds to the virus via two sugars and a string of amino acid residues. They are also relevant in the context of a clinical trial of the first HIV vaccine candidate, RV144, data from which suggests that the presence of V1/V2-reactive antibodies is associated with reduced infection risk.