Research Highlights
HIV: A new starting point for HIV vaccine design
Nature Reviews Drug Discovery 8, 11 (01 November 2009) | doi:10.1038/nrd3035
Identifying broadly neutralizing antibodies (BNAbs) against HIV, which recognize most manifestations of this highly mutable virus, is important in achieving the elusive goal of developing a viable HIV vaccine. However, although the BNAbs that have been identified are effective in primate models of HIV, they show limited potency against non-clade B viruses, which are responsible for most HIV infections outside Europe and North America. By systematically examining the neutralization effects of antibodies from 1,800 individuals infected with HIV-1, Dennis Burton and colleagues have identified two new potent BNAbs that shed light on a novel vaccine target.
The study began with an African donor who showed broad and potent neutralizing serum activity against HIV-1. Using antibody-containing cultured supernatants from 30,000 activated memory B cells from this donor, the authors performed high-throughput screening for binding to monomeric forms of the viral envelope proteins glycoprotein 120 (gp120) or gp41, and for neutralizing activity against two relevant HIV-1 primary isolates.
Surprisingly, ∼98% of the B cells that neutralized the HIV-1 primary isolate JR-CSF did not bind to monomeric gp120 from this isolate or gp41 from the HxB2 isolate, and ∼47% of the B cells that neutralized the other isolate, SF162, also did not bind to these glycoproteins. In addition, only 2% of cells with neutralizing activity were able to neutralize both isolates.
Examination of the antibody genes from five B cell cultures that had differing functional profiles showed that the B cells that neutralized gp120 (3 of the 5) did not show substantial neutralization breadth or potency against a multi-clade panel of 16 pseudoviruses. By contrast, the two B cell cultures that did not bind to monomeric gp120 or gp41 — PG9 and PG16 — neutralized a large proportion of the viruses in the panel at concentrations below 1 μg per ml, indicating that a previously unrecognized binding site might be crucial to their broadly neutralizing activity.
When tested against a larger panel of 162 viruses, PG9 and PG16 neutralized 127 and 116 viruses, respectively. Although PG9 and PG16 exhibit different neutralization breadth and potency against particular viruses, they both showed greater average neutralization efficacy than the four previously identified BNAbs. Furthermore, both PG9 and PG16 were able to neutralize IAVI-C18, a virus isolate that shows resistance to all BNAbs identified so far.
Next, the authors investigated the epitopes that are targeted by these antibodies. The fact that the antibodies competed for binding to the JR-CSF isolate suggested that the epitopes might be the same or at least overlap. Antibody binding was abolished by the interaction of the HIV-1 envelope protein with a soluble version of the protein receptor CD4, but not by interaction with antibodies that target the CD4 binding site. This suggests that CD4 binding causes conformational changes that affect the targeted epitope. Competitive enzyme-linked immunosorbent assays and alanine scanning experiments revealed that residues in the conserved regions of variable loops 2 and 3 of gp120 are key components of the epitope recognized by PG9 and PG16. In particular, N156 and N160 — known glycosylation sites on variable loop 2 — are crucial in forming the targeted epitope, as amino-acid substitutions that prevent glycosylation conferred resistance to the antibodies. Both PG9 and PG16 preferentially bound to the trimeric HIV-1 envelope protein, and the authors showed that this is due to gp120 presentation in the context of the trimeric spike rather than to gp120 cross-linking. Together, these findings raise hope that we may be closer to discovering a weak spot of HIV for vaccine design.
