Recent studies evaluating the evolution and specificities of broadly-neutralizing antibodies during HIV-1 infection 
have provided important insights regarding the significance of CD4i antibodies and their potential role in vaccine against HIV-1. So far, recombinant monomeric gp120 or oligomeric/trimeric gp140 glycoproteins have failed to elicit broad and potent neutralizing antibodies in experimental animal models. Therefore, rationale structural alterations to soluble Env for effective presentation of conserved ‘neutralizing’ epitopes represent an important strategy towards HIV-1 vaccine development. Described herein is a practical approach of eliciting CD4i epitope-directed virus neutralizing antibodies using a stably cross-linked complex of recombinant oligomeric gp140 and miniCD4 (M64U1-SH) 
, to target the conserved co-receptor binding site of the HIV-1 Env.
In earlier studies, two CD4 mimetic miniproteins (miniCD4) were cross-linked to various forms of HIV-1 Env 
. Based on results from those studies, the M64U1-SH miniCD4 was selected here for generating the cross-linked gp140-miniCD4 (gp140-S-S-M64U1) complex. Importantly, we observed that formation of a stable covalent complex by cross-linking of miniCD4 and gp140 protein via disulfide bond formation 
resulted in significantly higher-titers of CD4i epitope-directed binding and virus neutralizing antibodies in rabbits immunized with the cross-linked (gp140-S-S-M64U1) complex, not seen in animals that received a ‘bench-side’ mixing of miniCD4 and gp140 (gp140+M64U1). Nevertheless, both miniCD4 groups [cross-linked or mixed] elicited antibodies of higher avidities () than the gp140 alone group, potentially contributing to the improved virus neutralization seen with 2wp3/2wp4 sera from the (cross-linked or mixed) complexes ().
Moreover, rabbits immunized with the cross-linked gp140-S-S-M64U1 complex elicited antibodies targeted to the CD4i epitope(s) while those immunized with gp140 alone or mixed (gp140+M64U1) complex did not. By ‘serum mapping’ experiments, it appeared that the majority of the animals (except the control group immunized with the miniCD4) elicited V3-specific antibodies. Only one animal in the gp140-immunized group elicited CD4BS antibodies () and all animals (5/5) immunized with cross-linked gp140-S-S-M64U1 complex elicited CD4i epitope-directed antibodies, whereas the mixed (gp140+M64U1) complex did not. This observation of selective elicitation of CD4i epitope-directed antibodies was confirmed by the HIV-2 neutralization assay ().
Subtype B immunogens, such as the SF162 gp140 used here, elicit substantial levels of anti-V3 antibodies that can be blocked by subtype-specific V3 peptide 
. In addition, Env undergoes a conformational change upon CD4 binding that results in the relocation of the inner domain of gp120 to the molecular surface and exposure of V3 loop from the core domain to the solvent-accessible surface of gp120 
. Therefore, the induction of anti-V3 antibody responses by the gp140-miniCD4 [gp140-S-S-M64U1 (cross-linked) and gp140+M64U1 (mixed)] complex immunogens was evaluated using a competition assay that employed V3-cyclic peptide (). While over 80% of the gp140-specific binding activity of sera from gp140-immunized group was inhibited by V3-peptide, only about 20% of the binding activity of sera from both gp140-miniCD4, cross-linked and mixed, complex immunogen immunized animals was inhibited by V3-peptide. This observation that the elicitation of anti-V3 antibodies was not enhanced by both gp140-miniCD4 complex immunogens was also supported by HxB2-neutralization results. Sera from both the gp140-miniCD4, cross-linked or mixed, complex immunized groups neutralized HxB2, a virus that is not sensitive to anti-V3 antibodies 
, more potently (although without statistical significance) than sera from the group that was immunized by gp140 alone (). These data suggested that the structural alterations upon Env-CD4 protein binding and alterations upon gp140-miniCD4 interaction may differ concerning V3 loop exposure and subsequent V3-specific antibody elicitation. In addition, the results here also confirm that the breadth and potency by sera from cross-linked (gp140-S-S-M64U1) complex group is likely due to the elicitation of antibodies directed to CD4i epitope(s) rather than to the V3 loop. Hence, the use of a stable, covalently cross-linked complex of gp140 and miniCD4 appears to represent a viable method for the elicitation of higher titer antibodies directed to conserved CD4i epitope(s) on HIV-1 Env.
It was reported previously that some CD4i full-length antibodies encounter steric hindrances in accessing the ‘induced’ epitope on viral envelope glycoprotein available post-CD4 attachment, due to constrained space between the cell and the viral membrane following (viral) Env and (cellular) CD4 interaction 
. Nevertheless, numerous studies have reported the elicitation of high titers of CD4i antibody during natural human infection against multiple HIV-1 subtypes and circulating recombinant forms (CRFs). Importantly, these antibodies bind and neutralize not only various HIV-1 isolates but also HIV-2 in CD4-induced manner 
, indicating the functional constraints on receptor binding that create opportunities for broad humoral immune recognition and neutralization of viral quasispecies. Although speculations could be made as to how these antibodies are so commonly generated and how they neutralize various HIV-1 isolates, the precise in vivo
mechanism, more importantly in neutralization, is unknown.
The elicitation of CD4i antibodies signifies the extraordinary degree of antigenic conservation linked to co-receptor binding exhibited by diverse HIV-1 and HIV-2 lineages, and at the same time, the ability of human humoral immune system to exploit these constraints 
. But since humoral response determined by analyzing circulating antibodies are likely to change significantly over time, due to the high mutability of HIV-1 envelope, this measure in sera is unlikely a reflection of long-term protection. Rather analyses of memory B-cell compartment may be more telling about protective immunity. Recently, Y Guan et al. reported that although “elite controllers” showed low level of circulating antibody response against CD4i epitope(s), CD4i epitope-specific B lymphocyte memory cells were present at high frequencies (greater than a third of the anti-Env antibodies) in all ‘elite controllers’ subjects tested 
. Besides, similar conclusion on broader diversity of anti-HIV-1 antibodies isolated from memory B-cells have been echoed by JF Scheid et al. 
, in which ≥15% of the neutralizing antibodies were directed to the CD4i epitope(s). The induction of CD4i Abs in patients infected by HIV-1 
highlights the exposure of the co-receptor binding site on the virus surface, which may occur after binding of gp120 to CD4 on the target cell despite the steric hindrance reported by Labrijn et al. 
, or because of CD4-independent variants exposing the co-receptor binding site in most of the infected patients 
. Presence of circulating soluble CD4 or of CD4-gp120 complexes at the surface of targeted cells, following shedding of gp120 from gp41 and consequently from the virus particle following HIV-1 binding to CD4, could also contribute to elicitation of CD4i antibodies, as most recently exemplified in primates 
. Overall, these and other studies highlight the role of CD4i antibodies during the course of HIV-1 infection.
In the past, several studies have been performed, based on gp120-CD4 (or CD4 mimic) complexes or constrained ‘core’ gp120 antigens as vaccine candidates, aiming at inducing CD4i antibodies 
. T Fouts et al. showed that gp120 cross-linked to CD4 D1D2 domains raised antibodies that neutralized primary viruses regardless of co-receptor usage and genetic subtype in nonhuman primates 
. These findings were extended in a challenge study by A DeVico et al. 
, where macaques immunized with a single chain complex containing gp120BaL-rhesus macaque CD4 D1D2 showed better CD4i antibody response that correlated with the control of infection when challenged with SHIV162P3. Although this correlation is not a proof of the neutralizing efficiency of CD4i antibodies, it shows that the presence of these MAbs is fully related to the CD4-bound conformation of HIV-1 envelope in vivo
. Besides, recent approach towards constraining ‘gp120 core’ protein by site-directed disulfide linkage(s) has been a novel step towards eliciting epitope-specific response against the co-receptor binding site 
. These vaccination studies indicate the potential importance of strategies directed to raising antibodies against the CD4i site.
Some success in protection against HIV-1 acquisition has been afforded by active prophylactic vaccination using either DNA prime-recombinant Env protein boost 
or viral vector prime-recombinant Env protein boost 
or adjuvanted recombinant Env proteins 
in studies performed in nonhuman primate models, and most recently, in humans using a Canarypox vector prime plus recombinant Env protein boost approach in the RV144 Phase III trial performed in Thailand 
. Recent correlates of risk assessments of the RV144 trial indicated that V1V2-directed IgG antibodies in sera from vaccinees were associated with vaccine-elicited protection 
. Despite this early “hypothesis generating” finding in this first promising HIV-1 vaccine trial, the importance of the diversity of antibodies elicited by other epitopes on Env such as the CD4i epitope(s) 
, during infection and vaccination, and their role in immune protection against HIV-1 infection warrant further investigation.
The present work provides a powerful next generation approach using a novel and stable cross-linked complex of oligomeric gp140 glycoprotein and a CD4 mimetic miniprotein, in order to elicit broadly reactive functional antibody responses targeted to the highly conserved co-receptor binding site of the HIV-1 Env. Moreover, it does so while avoiding the elicitation of undesirable anti-CD4 reactivity. This cross-linked gp140-miniCD4 (gp140-S-S-M64U1) complex is suitably stable for future vaccine studies in non-human primate models, and represents a viable strategy for further evaluations of the role of CD4i epitope-directed antibody responses in protection against HIV-1 infection during vaccination.