Despite the development of highly effective antiretroviral therapy, HIV-1 caused 2.7 million new infections in 2007—roughly 7000 new infections per day.1 The continued global spread of HIV-1 outpaces our ability to treat each infected individual and underscores the urgency for new tools to effectively prevent HIV-1 infection. The extraordinary diversity of the virus, its capacity to evade adaptive immune responses, and our inability to induce broadly neutralizing antibodies against HIV represent unprecedented challenges for vaccine development. The recent demonstration of effective preexposure prophylaxis (PrEP) with tenofovir vaginal gel2 and combination oral tenofovir/emtricitabine3 has renewed hope for novel biomedical prevention strategies. However, even if effective, PrEP based on daily or near daily use of antiretroviral drugs faces difficulties in implementation because of poor adherence to the prescribed regimen, exemplified by the early closures of the FEM-PrEP trial and the tenofovir gel arm of the Vaginal and Oral Interventions to Control the Epidemic trial, both due to lack of efficacy or due to the concern over long-term adverse consequences of medications. Therefore, an alternative PrEP strategy might use antiretroviral agents that could be administered infrequently and are not associated with any serious side effects. Antibodies fit this target profile with their long half-life and favorable safety profile.4 Until recently, even the best HIV-neutralizing human monoclonal antibodies (mAbs), such as b12, 2G12, 2F5, and 4E10, lack sufficient potency or breadth. However, the feasibility of this strategy has improved substantially with the recent isolation and characterization of human mAbs PG9/16 and VRC01, both of which have shown remarkable potency and breadth in vitro.5–7
Ibalizumab, a humanized IgG4 monoclonal antibody that binds to human CD4 and blocks HIV-1 entry, may be another promising candidate for HIV prevention. This antibody has completed phase 1a, 1b, 2a, and 2b treatment trials in treatment-experienced subjects and completed dosing in a phase 1 prevention trial in HIV-1–uninfected volunteers.8–11 Epitope mapping12,13 and structural studies14 of the cocrystal of ibalizumab with CD4 demonstrated that this mAb is largely directed to domain 2 of human CD4, on the surface opposite both the major histocompatibility complex-class II binding site and the gp120 binding site. In vivo studies demonstrated significant antiviral activity, reducing SIV viremia in chronically infected rhesus macaques.15 In HIV-1–infected human subjects in need of salvage therapy, separate clinical trials testing single and multiple doses of ibalizumab showed a one log reduction in plasma viremia and increase in CD4+ T-cell count without evidence of serious adverse effects.9,10 Ibalizumab has been well tolerated and shown no serious adverse effects in more than 250 patients, some for up to 4 years of treatment.
Previous work using a limited number of laboratory-adapted and clinical HIV-1 isolates has shown that ibalizumab potently inhibits infection in vitro, with 100% inhibitory concentrations ranging from 0.08 to 1.25 μg/mL.12 With the advent of large clinically relevant Env reference panels, comprised of isolates from acute/early infection and primary transmitted/founder isolates,16 we sought to better characterize in vitro breadth and potency of ibalizumab against a diverse panel of 116 HIV-1 Envs and to identify any potential determinants of natural resistance. Our results indicate that ibalizumab in vitro breadth and potency compare well to the broadly neutralizing antibodies (bNAbs) PG9 and VRC01 and suggest that ibalizumab may be an appropriate candidate for long-acting PrEP strategies.
MATERIALS AND METHODS
Ibalizumab was provided by TaiMed Biologics, Inc. (Irvine, CA). 2G12, 4E10, 2F5, b12, VRC01, maraviroc, and T-20 (DAIDS) were obtained through the National Institutes of Health (NIH) AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, NIH as follows: 2G12, 4E10 and 2F5 from Dr Herman Katinger; b12 from Dr Dennis Burton and Dr Carlos Barbas; and VRC01 from Dr John Mascola. sCD4 was purchased from Progenics (Tarrytown, NY).
Viral Neutralization Assays
Neutralization was measured using single-round infection by HIV-1 Env-pseudoviruses and TZM-bl target cells, as described previously.16 Neutralization assays using wild-type pseudotypes were conducted at Mike Seaman's laboratory at the Collaboration for AIDS Vaccine Discovery (CAVD) Vaccine Immune Monitoring Consortium. For site-directed mutants and the corresponding wild-type Envs, neutralization assays were performed in-house using a similar assay. Briefly, 104 cells per well were plated in black 96-well culture plates overnight. Twelve to 16 hours later, cells were inoculated with 200 TCID50 per well in the presence of 15 μg/mL DEAE-Dextran (Sigma-Aldrich, Saint Louis, MO) and 48 hours later, infection was quantified by detection of β-galactosidase activity as per the manufacturer's instructions (Galacto-Star, Applied Biosystems).
Mutant HIV-1 Envs were created using the QuikChange II Site-Directed Mutagenesis kit as per the manufacturer's instructions (Stratagene).
The normalized V5 potential N-linked glycosylation site (PNGS) position was calculated as the primary amino acid position of the N of a PNGS from G459, normalized by the V5 length. This value ranged from 0 to 1, with 0 to <0.5 representing an N-terminal V5 location and >0.5 to <1 representing a C-terminal V5 location; viruses lacking a V5 PNGS were assigned a normalized V5 (nV5) PNGS position value of 1. Multivariate analysis suggested that the normalized (P = 0.002), rather than the nonnormalized (P = 0.269) V5 PNGS position more accurately explained the data.
Ibalizumab Potently Neutralizes a Diverse Panel of HIV-1
To assess ibalizumab HIV-1 neutralization breadth and potency, we analyzed its activity against a panel of 116 Tier-2 Env-pseudotyped viruses, specifically selected to represent envelope diversity by geography, clade, tropism, and stage of infection, including 30 transmitted/founder viruses.16 Because ibalizumab acts as a noncompetitive inhibitor of HIV-1 infection,13,14,17 reduced susceptibility to ibalizumab is primarily manifested as a plateaued neutralization curve, resulting in a maximum percent inhibition (MPI) of infection lesser than 100%. This is in contrast to HIV-1 inhibitors that act competitively, in which resistance manifests as an increased IC50 because of a rightward shift of the neutralization curve (Fig. 1A). The minor increases in IC50 seen in ibalizumab resistance seem to be a consequence of the reduced slope of the neutralization curve rather than a true rightward shift of the curve. This effect is well established for several other noncompetitive HIV-1 entry inhibitors, such as the CCR5 inhibitors maraviroc, vicriviroc, and aplaviroc, for which a reduction in MPI reflects the viruses' ability to use inhibitor-bound receptor for entry.18–21 When tested against the panel of 116 Env-pseudoviruses, ibalizumab exhibited favorable in vitro breadth and potency compared with well-characterized bNAbs b12, 2G12, 2F5, and 4E10, and the more potent and broader neutralizing antibodies PG9 and VRC01 (Fig. 1B). Ibalizumab neutralized 92% of viruses tested, as defined by 50% inhibition of infection, greater than that achieved by VRC01 (86%) and PG9 (79%), but slightly less than that observed by 4E10 (97%) (see Table S1, Supplemental Digital Content, http://links.lww.com/QAI/A359). However, ibalizumab loses substantial breadth when neutralization is defined as achieving 80% inhibition of infection, inhibiting only 65% of viruses, whereas VRC01 and PG9 maintain similar breadth when defined by 50% inhibition or 80% inhibition (Fig. 1B).
Ibalizumab potency is remarkable; median IC50 of ibalizumab (0.03 μg/mL) is an order of magnitude lower than the exceptionally potent mAbs PG9 (0.11 μg/mL) and VRC01 (0.22 μg/mL) and 2 orders of magnitude lower than mAbs 4E10, 2F5, 2G12 and b12 (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A359). Indeed, the remarkable potency of ibalizumab means that it has the greatest breadth at submicrogram concentrations (Fig. 1B), irrespective of clade (see Table S3, Supplemental Digital Content, http://links.lww.com/QAI/A359). Susceptibility to ibalizumab did not differ according to stage of infection or source of virus (data not shown), but clade AE viruses were more likely to be susceptible to ibalizumab than all other clades (P = 0.011; see Figure S1, Supplemental Digital Content, http://links.lww.com/QAI/A359).
Lack of V5 N-Linked Glycosylation Is the Dominant Determinant of Ibalizumab Resistance
Analyses of gp160 sequence polymorphism, including PNGSs and variable loop length, of the 116 HIV-1 Envs indicated that the dominant determinant of resistance involved glycosylation of V5 (Fig. 2A). We observed a highly significant association between ibalizumab MPI and the number of V5 PNGSs (ANOVA, P < 0.001; Fig. 2B). The complete absence of V5 PNGSs, which was relatively rare in this virus panel (frequency, 3.4%), was associated with marked resistance to ibalizumab, with median MPI lesser than 50% (n = 4, MPI 37.2 ± 16%, P < 0.001). Furthermore, reduction in the mean MPI was also evident among viruses with only one V5 PNGS compared with those with 2 V5 PNGS (n = 59, MPI 77.8 ± 18.3% vs n = 51, 89.2 ± 12.5%, P = 0.001).
We noticed, however, that viruses with 2 V5 PNGSs always had one of the PNGSs located in the N-terminal of V5 and, therefore, sought to determine the relative contributions of the number versus the location of the V5 PNGS to ibalizumab susceptibility. Due to the very low sequence homology and highly variable length of V5 (range, 4–15 amino acids), annotating the locations of the V5 PNGS using the Hxb2 numbering system was impractical. We thus defined the location of the V5 PNGSs in terms of its primary amino acid distance from the G459 residue of the conserved C4 CD4 binding site motif 455TRDGG459, adjusted for the V5 length, which we termed the normalized V5 (nV5) PNG position. Using this approach, we observed a strong correlation between ibalizumab MPI and the nV5 PNGS position (Figs. 2C, D; Pearson r = −0.555, P < 0.001), even when limited to viruses with only one V5 PNGS (Pearson r = −0.349, P = 0.007).
Among viruses with only one V5 PNGS, those with an N-terminal V5 PNGS had higher MPIs compared with viruses with a central or C-terminal V5 PNGS (n = 29, 83.4% ± 18.3% vs n = 30, 72.4% ± 16.8%, P = 0.02). Indeed, the MPI of viruses with one V5 PNGS, located in the N-terminal, reflected the MPIs of viruses with 2 V5 PNGSs (n = 51, 89.2 ± 12.5%). Furthermore, among viruses with 2 V5 PNGSs, the position of the second V5 PNGS had no detectable effect on the virus's susceptibility to ibalizumab. The MPI of the few viruses with 2 N-terminal PNGSs was similar to that of viruses with both an N- and C-terminal V5 PNGSs (n = 5, 83.6% ± 9.2% vs n = 46, 89.9% ± 12.7%, P = 0.290) and linear regression analysis of MPI and the nV5 position of the 2nd PNGS was nonsignificant (P = 0.914). Although multiple regression analyses (including all viruses) indicated that the nV5 PNGS position had the strongest influence on MPI (P < 0.001) and that the number of V5 PNGS is likely to have had an independent effect (P = 0.053), the effect of the number of V5 PNGSs was lost once viruses that lack V5 PNGSs were excluded from the analysis (P = 0.102), indicating that the statistical significance of the number of V5 PNGS to ibalizumab MPI is contributed predominantly by the viruses that lack V5 PNGS.
To determine whether the associated mutations were directly influencing the virus's susceptibility to ibalizumab, we created various V5 PNGS mutants of 2 clade-B primary isolates and assessed their susceptibility to ibalizumab (Fig. 3). Consistent with the sequence analysis, the nV5 PNGS position, rather than the number of V5 PNGS, predominantly influenced the virus' susceptibility to ibalizumab. For an ibalizumab-sensitive isolate AC10.0.28 (MPI 100%; Fig. 3A), mutation of the C-terminal V5 PNGS had a negligible effect on ibalizumab MPI (MPI 98%), unlike mutation of the N-terminal V5 PNGS, which yielded a highly resistant virus (MPI 50%). Conversely, introducing a PNGS to the V5 N-terminal of a highly ibalizumab-resistant virus RHPA4259.7 (MPI 42%) rendered the virus entirely susceptible to neutralization by ibalizumab (MPI 100%; Fig. 3B). Thus, the presence or the absence of an N-terminal V5 PNGS was sufficient to mediate ibalizumab susceptibility or resistance, respectively.
Taken together, these data suggest that the absence of V5 glycosylation confers resistance to ibalizumab, and that when a V5 PNGS is present, the distance of the V5 PNGS (in terms of primary amino acid sequence) from the C4 CD4 binding site motif TRDGG459 influences the degree of susceptibility.
Ibalizumab Susceptibility Is Associated With Absence of the 386 PNGS, Bulk of Residue 375 and a Shortened V2 Length
In addition to V5 glycosylation, significant univariate associations with ibalizumab MPI were observed for the 386 PNG site (P = 0.003), residue 375 (P = 0.007), clade AE viruses (P = 0.01), position of the C3 PNG site (inverse of distance from C3, β-sheet 15, CD4bs motif, P = 0.022), and V1 length (P = 0.024). However, multivariate analyses indicate that only the 386 PNG site (P = 0.016) and residue 375 (P = 0.008) were independently associated with susceptibility to ibalizumab.
The absence of the 386 PNGS (Fig. 4A; n = 17, 91.2 ± 9.6 vs n = 99, 80.2 ± 19.4, P = 0.010), located at the base of the V4 loop and proximal to the C4 CD4bs motif 365SGGD368, or the presence of an amino acid with a long side chain (H/R/M) at position S375 (Fig. 4B; n = 16, MPI 93.9 ± 5.4 vs n = 100, 79.9 ± 19.3, P = 0.005) was associated with enhanced susceptibility to ibalizumab. Even in the absence of a V5 N-terminal PNGS, viruses that lack the 386 PNG or contain 375H/R/M maintain an ibalizumab MPI (88%) similar to that observed in viruses with a V5 N-terminal PNG (87%) (Fig. 4C).
The length of the V2 loop was also associated with susceptibility to ibalizumab, although its effect was only evident in viruses that lack a V5 N-terminal PNG (Fig. 4D; P = 0.001); in viruses that contain a V5 N-terminal PNG, no association existed (P = 0.673). Interestingly, although the V2 loop association was not restricted to clade AE viruses, clade AE viruses in this panel had significantly shorter V2 loop length than other clades (see Table S1, Supplemental Digital Content, http://links.lww.com/QAI/A359; P = 0.037). V2 length was highly associated with number of V2 PNGS (Spearman ρ = −0.651, P < 0.001), with the increased V2 length associated with insertion of a PNGS at various positions between the 186 PNGS and the highly conserved 197 PNGS located at the C-terminal base of the V2 loop, consistent with previous observations.22,23 However, we could not identify a direct relationship between ibalizumab MPI and V2 potential N-linked glycosylation.
Ibalizumab Resistance Is Associated With Enhanced Sensitivity to VRC01 and sCD4 Neutralization
As passive immunization as a prevention measure against HIV-1 infection4,24 is currently being reenergized and to gain a better understanding of resistance to ibalizumab, we assessed the relationship between resistance to ibalizumab and to other bNAbs. Among the 116 viruses tested, no associations existed with ibalizumab and PG9 (P = 0.625), PG16 (P = 0.216), b12 (P = 0.670), 2G12 (P = 0.157), 4E10 (P = 0.098), or HIVIG (P = 0.519) (see Figures S2a–h, Supplemental Digital Content, http://links.lww.com/QAI/A359). However, susceptibility to ibalizumab was associated with resistance to sCD4 (MPI vs IC50, Spearman ρ = 0.420, P < 0.001; Fig. 5A) and sensitivity to the b12/2F5/2G12 cocktail TriMab (see Figure S2g, Sup-plemental Digital Content, http://links.lww.com/QAI/A359; ρ = 0.196, P = 0.035) and tended to be associated with resistance to 2F5 (see Figure S2f, Supplemental Digital Content, http://links.lww.com/QAI/A359; Spearman ρ = −0.174, P = 0.062).
Because susceptibility to ibalizumab was predominantly mediated by V5 glycosylation and resistance to the broad and potent neutralizing mAb VRC01 involves bulky V5 residues,7 we hypothesized that there might be complementary resistance between ibalizumab and VRC01. For 78 of the 116 viruses tested, the corresponding VRC01 IC50 and IC80 values were determined in the same CAVD reference laboratory.7 Among these viruses, we discovered an inverse association between resistance to ibalizumab and VRC01 (Fig. 5B, Spearman ρ = 0.308, P = 0.006) that seemed, at least in part, mediated by V5 N-terminal glycosylation (Fig. 5C, Mann–Whitney U test, P = 0.021).
Using the same site-directed mutants used to investigate ibalizumab resistance, we found that indeed, N-linked glycosylation of V5, in particular the N-terminal of V5, substantially reduced sensitivity to VRC01 (Figs. 5D, E). For AC10.0.29 and RHPA4259.7, isolates for which alteration of V5 glycosylation had remarkable effects on susceptibility to ibalizumab, sensitivity to VRC01 was enhanced 22-fold and 17-fold by the absence of an N-terminal V5 PNGS, respectively. These data indicate that ibalizumab and VRC01 exhibit complementary resistance, in part, mediated by mutually exclusive V5 N-terminal glycosylation.
We observed a minor strain-specific effect of V5 glycosylation on sCD4 and b12 sensitivity, consistent with a report by McCaffrey et al25 using mutants of SF162. For the ibalizumab-resistant virus, RHPA4259.7, the N-terminal V5 PNGS modulated sCD4 and b12 sensitivity 3- and 2-fold, respectively. Sensitivity to entry inhibitors maraviroc and T-20 (DAIDS), and infectivity, were unaffected (see Table S5, Supplemental Digital Content, http://links.lww.com/QAI/A359).
The development of bNAbs for long-acting PrEP against HIV-1 infection requires careful attention to the parameters of potency, breadth, and resistance. In this study, we show that ibalizumab exhibits broad and potent HIV-1 neutralizing activity in vitro against a large panel of diverse clinically relevant pseudoviruses.16 When assessing neutralization as commonly defined by 50% inhibition of infection, ibalizumab breadth was superior to other bNAbs, including PG9 and VRC01. However, when defined by 80% inhibition, PG9 and VRC01 appeared broader (Fig. 1B). On the other hand, ibalizumab is approximately one log more potent than PG9 and VRC01 (see Table S1, Supplemental Digital Content, http://links.lww.com/QAI/A359) equating to superior breadth at submicrogram concentrations (Fig. 1B; see Table S3, Supplemental Digital Content, http://links.lww.com/QAI/A359), which may prove to be an important advantage as dosing and frequency are optimized and the practicality of scale-up is assessed, for this novel approach in the field of HIV prevention. The degree of inhibition of infection in in vitro assays required for effective PrEP and the serum and tissue concentrations of passively administered or gene-transferred delivered antibodies that can be practically maintained must be determined as a mAb PrEP strategy is pursued.
The natural preexisting resistance to ibalizumab described here seems very similar to resistance that develops in response to ibalizumab therapy. In a recent study of ibalizumab resistance during a phase 1b trial, loss of V5 glycosylation, in particular at the N-terminus, was observed in ibalizumab-resistant variants isolated at the time of virological failure.26 Furthermore, the ibalizumab-resistant variants had enhanced sensitivity to sCD4 compared with pretherapy isolates that was not mediated by V5 glycosylation.26 This similarity is consistent with the model proposed by Toma et al26 that clinical ibalizumab resistance likely developed because of an outgrowth of preexisting resistant variants, rather than de novo mutations. Therefore, there seems to be at least 2 pathways that mediate resistance to ibalizumab; one involving abrogation of V5 N-terminal glycosylation and another related to sCD4 sensitivity, and both may provide a selective advantage to the virus independent of ibalizumab.
The mechanism by which V5 N-terminal PNG mediates susceptibility to ibalizumab is unclear. Although ibalizumab does not inhibit binding of sCD4 to monomeric gp120, consistent with epitope mapping data12,13 and the co-crystal structure of ibalizumab Fab complexed with 2 domain CD4,14 it does inhibit binding of virions and HIV-1–infected cells to sCD4, although not sufficient enough to account for its antiviral activity.17 Thus, ibalizumab is thought to inhibit post–gp120-CD4 binding events required for fusion. Indeed, ibalizumab has been shown to inhibit certain gp160 conformational changes that occur during fusion.17 It is evident from the proximity of the V5 loop and the ibalizumab epitope highlighted on the cryo-electron microscopy model of the gp120-CD4 complex before and after CD4 binding–induced conformational change27 that the Fab of ibalizumab, which is sufficient for inhibiting HIV-1 entry,14 may sterically hinder the conformational change of the complex (Fig. 6). Indeed, based on the available crystal structures of gp120-CD4 complexes and the ibalizumab-sCD4 complex, the distance between the V5 loop and the ibalizumab Fab ranges from 5.5 to 10 Å, which approximates that of a glycan.14 Furthermore, the N-terminus of the V5 is closer to the ibalizumab Fab than the V5 C-terminus, which may explain why the N-terminal glycan but not the C-terminal glycan is associated with ibalizumab resistance.
In addition to steric hindrance, the lack of the V5 N-terminal PNGS may mediate escape by facilitating altered gp120-CD4 interactions required for fusion in the presence of ibalizumab. The V5 glycan flanks the 455TRDGG459 CD4 binding motif and deletion of the V5 glycan has been shown to enhance exposure of CD4bs epitopes.28 Escape from the murine progenitor of ibalizumab (5A8) during in vitro passaging experiments suggest escape is due to altered CD4 interactions yielding increased sCD4 sensitivity, cross-resistance to other anti-CD4 mAbs, and enhanced ability to undergo conformational change.29 Interestingly, escape from CADA, a small molecule inhibitor of CD4 biosynthesis, has been observed because of a V5 S463P mutation that abrogates the glycosylation motif and enhanced the virus's sensitivity to sCD4 and allowed the viruses to use low CD4 density.30 We observed moderate strain-specific effects of V5 glycosylation on sCD4 sensitivity using V5 PNGS site-directed mutants (see Table S5, Supplemental Digital Content, http://links.lww.com/QAI/A359), and sCD4 sensitivity in this panel was not associated with V5 glycosylation. Therefore, while the absence of the V5 N-terminal glycan may moderately influence sCD4 sensitivity, V5 glycosylation is unlikely to be mediating escape solely by enhancing sensitivity to CD4.
The association of short V2 loop lengths (Fig. 4D), long side chains at position 375 (Figs. 4B, C), the absence of the 386 PNGS (Fig. 4A), and clade AE (see Figure S1, Supplemental Digital Content, http://links.lww.com/QAI/A359) with enhanced susceptibility to ibalizumab further suggest that alterations of the CD4bs may contribute to ibalizumab resistance. The envelope of clade AE viruses are known to be immunologically distinct from other subtypes,31–33 most notably the V2 loop and CD4bs.34 The V2 loop length and V2 glycosylation modulates the accessibility of the CD4bs.35–37 V2 loops of each protomer interact at the apex of the trimer27,38 and longer V2 loops could therefore provide greater flexibility to the trimer, possibly facilitating viruses' ability to use ibalizumab-bound CD4 for entry. Escape from ibalizumab during in vitro passaging experiments occurs because of shortening V2 loop lengths (data not shown), but the mechanism remains undetermined. Furthermore, amino acids with long side chains at position 375, which are conserved among clade AE envelopes, are selected during escape for multiple virus strains during in vitro passaging experiments with CD4 mimetic BMS806,39 fills the Phe43 cavity, stabilizes the association of the gp120 inner and outer domains in the CD4-bound conformation,40 and enhances gp120 binding to sCD4 and sCD4 sensitivity. Clade AE viruses in this panel did have significantly shorter V2 loop lengths, were entirely 375H, and were significantly more resistant to sCD4 compared with other subtypes. However, we note that the V2 loop length and sCD4 associations with ibalizumab susceptibility were independent of clade AE.
The association of the absence of the 386 PNGS with enhanced ibalizumab susceptibility is intriguing. Removal of the 386 PNG influences antibody and sCD4 accessibility to the CD4bs41 and is associated with enhanced macrophage tropism and HIV-associated dementia.37 Macrophages express low levels of CD4 compared with CD4+ T cells, and it has been shown that enhanced macrophage tropism is associated with the ability of the virus to use low CD4 density. However, the absence of the 386 PNGS was associated with enhanced susceptibility to ibalizumab, without affecting sCD4 sensitivity. This highlights the complexity of ibalizumab resistance and gp120 conformation and suggests that an altered, rather than an open, CD4bs per se influences susceptibility to ibalizumab.
The significant inverse relationship between susceptibility to ibalizumab and sensitivity to VRC01 (Fig. 5B), which was mediated at least in part, by mutually exclusive V5 glycosylation, suggests they would complement each other well as part of a passive immunization cocktail and for HIV-1 treatment and prevention and would be expected to provide a greater barrier to resistance than other mAb combinations. Indeed, of the antibodies examined, only ibalizumab exhibited an inverse association with sensitivity to other bNAbs.
The plethora of mAbs that have been isolated and characterized in recent years have substantially improved breadth or potency compared with the prototype bNAbs. However, improvements may still be required for an effective prevention strategy. The in vitro antiviral activity of ibalizumab presented here indicates that ibalizumab rivals the best anti-HIV mAbs, such as PG9, VRC01, and the most recently discovered PGT mAbs,42for use as PrEP. Ibalizumab exerts a broader anti-HIV effect, particularly at low concentrations, whereas it failed to inhibit some viruses beyond 80%, highlighting the importance of assessing the entire antiviral profile of mAbs rather than simply their IC50 or IC80. Although the plethora of mAbs isolated in recent years have the best antiviral properties demonstrated to date, like ibalizumab, they fail to provide complete coverage against circulating strains and warrant the continued search and characterization of new bNAbs. We are currently optimizing ibalizumab's potency, breadth, and pharmacokinetic profile while investigating ibalizumab in a phase 1 trial in uninfected subjects.
The authors are grateful to Dennis Burton and Wayne Koff for the PG9 and PG16 neutralization data, TaiMed Biologics, Inc for ibalizumab, the Bill and Melinda Gates Foundation Collaboration for AIDS Vaccine Discovery for financial support and Yaoxing Huang, Neal Padte, Sandhya Vasan, Moriya Tsuji, Faye Yu, Jian Yu, and Sun Ming for helpful discussions.
2. Abdool Karim Q, Abdool Karim SS, Frohlich JA, et al.. Effectiveness and safety of tenofovir gel, an antiretroviral microbicide, for the prevention of HIV infection in women. Science. 2010;329:1168–1174.
3. Grant RM, Lama JR, Anderson PL, et al.. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med. 2010;363:2587–2599.
4. Abela IA, Reynell L, Trkola A. Therapeutic antibodies in HIV treatment—classical approaches to novel advances. Curr Pharm Des. 2010;16:3754–3766.
5. Walker LM, Phogat SK, Chan-Hui PY, et al.. Broad and potent neutralizing antibodies from an African donor reveal a new HIV-1
vaccine target. Science. 2009;326:285–289.
6. Wu X, Yang ZY, Li Y, et al.. Rational design of envelope identifies broadly neutralizing human monoclonal antibodies to HIV-1
. Science. 2010;329:856–861.
7. Zhou T, Georgiev I, Wu X, et al.. Structural basis for broad and potent neutralization
VRC01. Science. 2010;329:811–817.
8. Dimitrov A. Ibalizumab
, a CD4-specific mAb to inhibit HIV-1
infection. Curr Opin Investig Drugs. 2007;8:653–661.
9. Jacobson JM, Kuritzkes DR, Godofsky E, et al.. Safety, pharmacokinetics, and antiretroviral activity of multiple doses of ibalizumab
(formerly TNX-355), an anti-CD4 monoclonal antibody
, in human immunodeficiency virus type 1-infected adults. Antimicrob Agents Chemother. 2009;53:450–457.
10. Kuritzkes DR, Jacobson J, Powderly WG, et al.. Antiretroviral activity of the anti-CD4 monoclonal antibody
TNX-355 in patients infected with HIV type 1. J Infect Dis. 2004;189:286–291.
11. Zhang XQ, Sorensen M, Fung M, et al.. Synergistic in vitro antiretroviral activity of a humanized monoclonal anti-CD4 antibody
(TNX-355) and enfuvirtide (T-20). Antimicrob Agents Chemother. 2006;50:2231–2233.
12. Burkly LC, Olson D, Shapiro R, et al.. Inhibition of HIV infection by a novel CD4 domain 2-specific monoclonal antibody
. Dissecting the basis for its inhibitory effect on HIV-induced cell fusion. J Immunol. 1992;149:1779–1787.
13. Song R, Franco D, Kao CY, et al.. Epitope mapping of ibalizumab
, a humanized anti-CD4 monoclonal antibody
activity in infected patients. J Virol. 2010;84:6935–6942.
14. Freeman MM, Seaman MS, Rits-Volloch S, et al.. Crystal structure of HIV-1
primary receptor CD4 in complex with a potent antiviral antibody
. Structure. 2010;18:1632–1641.
15. Reimann KA, Cate RL, Wu Y, et al.. In vivo administration of CD4-specific monoclonal antibody
: effect on provirus load in rhesus monkeys chronically infected with the simian immunodeficiency virus of macaques. AIDS Res Hum Retroviruses. 1995;11:517–525.
16. Seaman MS, Janes H, Hawkins N, et al.. Tiered categorization of a diverse panel of HIV-1
Env pseudoviruses for assessment of neutralizing antibodies. J Virol. 2009;84:1439–1452.
17. Moore JP, Sattentau QJ, Klasse PJ, et al.. A monoclonal antibody
to CD4 domain 2 blocks soluble CD4-induced conformational changes in the envelope glycoproteins of human immunodeficiency virus type 1 (HIV-1
) and HIV-1
infection of CD4+ cells. J Virol. 1992;66:4784–4793.
18. Westby M, Smith-Burchnell C, Mori J, et al.. Reduced maximal inhibition in phenotypic susceptibility assays indicates that viral strains resistant to the CCR5 antagonist maraviroc utilize inhibitor-bound receptor for entry. J Virol. 2007;81:2359–2371.
19. McNicholas P, Wei Y, Whitcomb J, et al.. Characterization of emergent HIV resistance
in treatment-naive subjects enrolled in a vicriviroc phase 2 trial. J Infect Dis. 2010;201:1470–1480.
20. Tsibris AM, Sagar M, Gulick RM, et al.. In vivo emergence of vicriviroc resistance
in a human immunodeficiency virus type 1 subtype C-infected subject. J Virol. 2008;82:8210–8214.
21. Tilton JC, Amrine-Madsen H, Miamidian JL, et al.. HIV type 1 from a patient with baseline resistance
to CCR5 antagonists uses drug-bound receptor for entry. AIDS Res Hum Retroviruses. 2010;26:13–24.
22. Masciotra S, Owen SM, Rudolph D, et al.. Temporal relationship between V1V2 variation, macrophage replication, and coreceptor adaptation during HIV-1
disease progression. AIDS. 2002;16:1887–1898.
23. Sagar M, Wu X, Lee S, et al.. Human immunodeficiency virus type 1 V1-V2 envelope loop sequences expand and add glycosylation sites over the course of infection, and these modifications affect antibody neutralization
sensitivity. J Virol. 2006;80:9586–9598.
24. Zolla-Pazner S, Gorny MK. Passive immunization for the prevention and treatment of HIV infection. AIDS. 1992;6:1235–1247.
25. McCaffrey RA, Saunders C, Hensel M, et al.. N-linked glycosylation of the V3 loop and the immunologically silent face of gp120 protects human immunodeficiency virus type 1 SF162 from neutralization
by anti-gp120 and anti-gp41 antibodies. J Virol. 2004;78:3279–3295.
26. Toma J, Weinheimer SP, Stawiski E, et al.. Loss of asparagine-linked glycosylation sites in variable region 5 of human immunodeficiency virus type 1 envelope is associated with resistance
to CD4 antibody ibalizumab
. J Virol. 2011;85:3872–3880.
27. Liu J, Bartesaghi A, Borgnia MJ, et al.. Molecular architecture of native HIV-1
gp120 trimers. Nature. 2008;455:109–113.
28. Rong R, Li B, Lynch RM, et al.. Escape from autologous neutralizing antibodies in acute/early subtype C HIV-1
infection requires multiple pathways. PLoS Pathog. 2009;5:e1000594.
29. Klasse PJ, Sattentau QJ. Altered CD4 interactions of HIV type 1 LAI variants selected for the capacity to induce membrane fusion in the presence of a monoclonal antibody
to domain 2 of CD4. AIDS Res Hum Retroviruses. 1996;12:1015–1021.
30. Vermeire K, Van Laethem K, Janssens W, et al.. Human immunodeficiency virus type 1 escape from cyclotriazadisulfonamide-induced CD4-targeted entry inhibition is associated with increased neutralizing antibody
susceptibility. J Virol. 2009;83:9577–9583.
31. Quinnan GV Jr, Zhang PF, Fu DW, et al.. Expression and characterization of HIV type 1 envelope protein associated with a broadly reactive neutralizing antibody
response. AIDS Res Hum Retroviruses. 1999;15:561–570.
32. Mascola JR, Louder MK, Surman SR, et al.. Human immunodeficiency virus type 1 neutralizing antibody
serotyping using serum pools and an infectivity reduction assay. AIDS Res Hum Retroviruses. 1996;12:1319–1328.
33. Gorny MK, Williams C, Volsky B, et al.. Human monoclonal antibodies specific for conformation-sensitive epitopes of V3 neutralize human immunodeficiency virus type 1 primary isolates from various clades. J Virol. 2002;76:9035–9045.
34. Moore JP, McCutchan FE, Poon SW, et al.. Exploration of antigenic variation in gp120 from clades A through F of human immunodeficiency virus type 1 by using monoclonal antibodies. J Virol. 1994;68:8350–8364.
35. Stamatatos L, Cheng-Mayer C. An envelope modification that renders a primary, neutralization
-resistant clade B human immunodeficiency virus type 1 isolate highly susceptible to neutralization
by sera from other clades. J Virol. 1998;72:7840–7845.
36. Wyatt R, Moore J, Accola M, et al.. 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. 1995;69:5723–5733.
37. Fox DG, Balfe P, Palmer CP, et al.. Length polymorphism within the second variable region of the human immunodeficiency virus type 1 envelope glycoprotein affects accessibility of the receptor binding site. J Virol. 1997;71:759–765.
38. Wu SR, Loving R, Lindqvist B, et al.. Single-particle cryoelectron microscopy analysis reveals the HIV-1
spike as a tripod structure. Proc Natl Acad Sci U S A. 2010;107:18844–18849.
39. Madani N, Perdigoto AL, Srinivasan K, et al.. Localized changes in the gp120 envelope glycoprotein confer resistance
to human immunodeficiency virus entry inhibitors BMS-806 and #155. J Virol. 2004;78:3742–3752.
40. Pan Y, Ma B, Keskin O, et al.. Characterization of the conformational state and flexibility of HIV-1
glycoprotein gp120 core domain. J Biol Chem. 2004;279:30523–30530.
41. Koch M, Pancera M, Kwong PD, et al.. Structure-based, targeted deglycosylation of HIV-1
gp120 and effects on neutralization
sensitivity and antibody
recognition. Virology. 2003;313:387–400.
42. Walker LM, Huber M, Doores KJ, et al.. Broad neutralization
coverage of HIV by multiple highly potent antibodies. Nature. 2011;477:466–470.