Research ArticleHIV

Broadly neutralizing antibodies targeting the HIV-1 envelope V2 apex confer protection against a clade C SHIV challenge

See allHide authors and affiliations

Science Translational Medicine  06 Sep 2017:
Vol. 9, Issue 406, eaal1321
DOI: 10.1126/scitranslmed.aal1321

Potently protective antibodies

HIV is thought to be an elusive virus, but there are multiple epitopes on HIV that can be targeted by neutralizing antibodies, such as the V2 loop of the envelope protein. Julg et al. reasoned that potent anti-V2 HIV antibodies given prophylactically could prevent infection from taking place. They tested this in nonhuman primates with a novel clade C SHIV strain and observed protection at very low concentrations of circulating neutralizing antibody, suggesting that passive immunization with these types of antibodies might be protective in people.


Neutralizing antibodies to the V2 apex antigenic region of the HIV-1 envelope (Env) trimer are among the most prevalent cross-reactive antibodies elicited by natural infection. Two recently described V2-specific antibodies, PGDM1400 and CAP256-VRC26.25, have demonstrated exquisite potency and neutralization breadth against HIV-1. However, little data exist on the protective efficacy of V2-specific neutralizing antibodies. We created a novel SHIV-325c viral stock that included a clade C HIV-1 envelope and was susceptible to neutralization by both of these antibodies. Rhesus macaques received a single infusion of either antibody at three different concentrations (2, 0.4, and 0.08 mg/kg) before challenge with SHIV-325c. PGDM1400 was fully protective at the 0.4 mg/kg dose, whereas CAP256-VRC26.25-LS was fully protective even at the 0.08 mg/kg dose, which correlated with its greater in vitro neutralization potency against the challenge virus. Serum antibody concentrations required for protection were <0.75 μg/ml for CAP256-VRC26.25-LS. These data demonstrate unprecedented potency and protective efficacy of V2-specific neutralizing antibodies in nonhuman primates and validate V2 as a potential target for the prevention of HIV-1 infection in passive immunization strategies in humans.


Despite the efforts of the HIV-1 vaccine field, no HIV-1 envelope (Env) immunogen to date has been able to elicit antibodies with broadly neutralizing activity (1). In contrast, many HIV-1–infected individuals produce neutralizing antibodies with some degree of breadth during the course of infection (24). Over the past few years, several antibodies targeting distinct epitopes of the HIV-1 envelope (Env) trimer and with potent and broad activity against diverse clinical isolates have been identified (58). In particular, neutralizing antibodies directed toward the CD4 binding site and the V3 region have shown promise in preclinical studies, in which single intravenous doses of antibodies protected rhesus macaques against challenges with simian human immunodeficiency virus (SHIV) (912). In the absence of a vaccine that can elicit such broadly neutralizing antibody (bNAb) responses, passive immunization with bNAbs is being explored for HIV-1 prevention strategies.

Although antibodies against several regions of the Env trimer have been described (6), neutralizing antibodies to the V2 apex antigenic region of the HIV-1 Env trimer are among the most prevalent cross-reactive antibodies elicited during infection (1315). The V1V2 region, which harbors multiple glycans and is highly sequence-diverse, is located at the Env apex and plays a vital role in the Env function by stabilizing the trimeric spike on the virion surface. It also shields V3 and the co-receptor binding sites in the prefusion state and exposes them upon CD4 binding (16). Although these antibodies are common in HIV-1–infected individuals, we know very little about their ability to confer protection against infection. In the recent RV144 HIV-1 vaccine study, binding antibodies against the V1V2 region were associated with reduced risk of infection (17).

To date, V2-directed bNAbs have been isolated from several donors, including the International AIDS Vaccine Initiative (IAVI) Protocol G donor 24 (PG9 and PG16) (18), the Center for HIV/AIDS Vaccine Immunology (CHAVI) donor 0219 (CH01–CH04) (19), the Centre for the AIDS Programme of Research in South Africa (CAPRISA) 256 donor (CAP256-VRC26.01-33) (20, 21), and the IAVI Protocol G donor 84 (PGT141–145 and PGDM1400–1412) (5, 22). These antibodies bind to the intact trimer with a stoichiometry of one per trimer (20) and interact with glycans at N160 and, to a lesser extent, N156 (18). They also have a very long heavy-chain complementarity-determining region 3 (CDRH3), which enables them to effectively penetrate the glycan shield (21). For the present study, we selected two V2-specific monoclonal antibodies (mAbs), CAP256-VRC26.25 and PGDM1400, for their exquisite potency and neutralization breadth. CAP256-VRC26.25 neutralized 57% of global viral isolates and 70% of clade C isolates with a median inhibitory concentration (IC50) of 0.001 μg/ml against sensitive viruses (21, 23). Among the PGT145 antibody family, the somatic variant PGDM1400 had a particularly broad and exceptionally potent neutralization activity with 83% global coverage at a median IC50 of 0.003 μg/ml (22). These V2-specific antibodies have superior in vitro potency compared to the V3 glycan–dependent antibodies PGT121 and PGT128 (5), which are among the most potent bNAbs described to date. However, the protective efficacy of V2-specific bNAbs against pathogenic tier 2 SHIV challenges remains unexplored.

Here, we evaluated the protective efficacy of these V2-specific bNAbs against SHIV challenge in nonhuman primates. We created a novel SHIV-325c stock that included a clade C Env and against which PGDM1400 and CAP256-VRC26.25 showed potent neutralization activity. Animals received a single infusion of PGDM1400 or CAP256-VRC26.25-LS (engineered with the Fc-LS mutation to increase in vivo half-life) at three different concentrations and were challenged with SHIV-325c.


Generation and characterization of SHIV-325c

PGDM1400 and CAP256-VRC26.25 have been shown to neutralize HIV-1 broadly and with high potency (21, 22), but they did not neutralize several commonly used SHIVs (Table 1). We therefore generated a novel challenge stock SHIV-325c by cloning a clade C HIV-1 env sequence from an early HIV-1–infected individual from South Africa (24) into the SHIV KB9-AC backbone. Large-scale challenge stocks were generated by inoculation of the 293T transfection–derived supernatants into primary human peripheral blood mononuclear cells (PBMCs), resulting in virus infectivity titers of 3.49 × 105 TCID50 (median tissue culture infectious dose)/ml and viral loads ranging from 1.3 × 109 to 1.5 × 109 RNA copies/ml. We sequenced the env of the SHIV-325c challenge stock using single-genome amplification (SGA) and compared it to the original HIV-1 env patient sequence (fig. S1). The diversity of the viral stock was low (mean diversity, 0.06%; excluding APOBEC-mutated sequences). Among 25 SGA-derived sequences obtained from the SHIV-325c stock, 12 amplicons were 100% identical to the original HIV-1 env consensus sequences, and 11 amplicons showed one to three sporadic nucleotide differences (mean divergence, 0.03%; maximum divergence, 0.13%). Only one amplicon demonstrated several APOBEC-mutated sequences. These data show that the env sequences in the SHIV-325c stock were very similar to the original HIV-1 env sequence.

Table 1. Neutralization profiles of PGDM1400, CAP256-VRC26.25, and PG9 against SHIV challenge stocks.

IC50 and IC80 values are in micrograms per milliliter. Values between 0.001 and 0.01 μg/ml are highlighted in red; between 0.01 and 0.1 μg/ml, orange; between 0.1 and 1 μg/ml, yellow; between 1 and 10 μg/ml, light green; and between 10 and 50 μg/ml, dark green. Virus suffix indicates PBMC cell type used for production: .Rh, rhesus PBMCs; .Hu, human PBMCs.

Embedded Image
View this table:

We next evaluated the infectivity of the SHIV stock in vivo. Eight adult rhesus macaques were inoculated via the intrarectal route with 1 ml of the SHIV-325c challenge stock. All animals developed productive infection (Fig. 1A) and exhibited robust peak viral loads ranging from 4.7 to 6.7 log RNA copies/ml after inoculation. These viral loads were comparable to those after SHIV-SF162P3 or SHIV-AD8 infection (25, 26). Viral loads declined and reached chronic set point levels by about week 10. Set point viral loads varied among animals (2.2 to 5.2 log SIV RNA copies/ml), and 75% (six of eight) of infected animals exhibited chronic viremia at week 14, with mean set point viral loads of 3.9 log RNA copies/ml. During acute infection, mean CD4+ T cell numbers declined by 27% in SHIV-325c–inoculated animals (preinfection count, 1406 cells/mm3; postinfection nadir, 1027 cells/mm3; P = 0.016, paired t test) (Fig. 1B) but did not subsequently exhibit further decline.

Fig. 1. Plasma viral loads and CD4+ T cell counts in rhesus macaques challenged with SHIV-325c.

(A) Plasma viral RNA (log RNA copies/ml) are shown for animals that were challenged intrarectally with SHIV-325c. The red line represents the mean log RNA copies/ml. (B) CD4+ T cell numbers in the same animals before and after SHIV-325c infection (P = 0.016, paired t test).

When tested against a panel of 14 bNAbs, SHIV-325c exhibited sensitivity to the V2 antibodies PG9, CAP256-VRC26.25, and PGDM1400 but was resistant to several other bNAbs, including the CD4 binding site antibodies VRC01 and 3BNC117, the V3 glycan antibody PGT128, and several membrane-proximal external region (MPER) and CD4i antibodies (Table 2). SHIV-325c sensitivity across the tested bNAbs was comparable to the frequently used tier 2 SHIV-SF162P3, which demonstrated resistance against 7 of the 14 tested bNAbs. In contrast, SHIV-325c was more resistant than the tier 1 SHIV-BaL, which was neutralized by 11 of the 14 bNAbs, indicating that SHIV-325c had a tier 2 phenotype. SHIV-325c neutralization sensitivity to PGDM1400 and CAP256-VRC26.25 (IC80 of 0.104 and 0.006 μg/ml, respectively) was comparable to the median neutralization sensitivity of a multiclade panel of 208 HIV pseudoviruses (IC80 of 0.04 and 0.03 μg/ml, respectively) (Fig. 2A and table S1). PGDM1400 and CAP256-VRC26.25 also neutralized SHIV-325c at a similar potency to a panel of clade C pseudoviruses (IC80 of 0.02 and 0.01 μg/ml, respectively) (Fig. 2B). No neutralization curve plateaus below 100% neutralization were observed with these three V2 antibodies against SHIV-325c (Fig. 3), unlike those previously described for PG9 against SHIV-BaL (11).

Table 2. Neutralization properties of three different SHIV viruses in TZM-bl assays against a panel of bNAbs targeting distinct epitopes.

IC50 and IC80 values are in micrograms per milliliter. Values between 0.001 and 0.01 μg/ml are highlighted in red; between 0.01 and 0.1 μg/ml, orange; between 0.1 and 1 μg/ml, yellow; between 1 and 10 μg/ml, light green; and between 10 and 50 μg/ml, dark green.

Embedded Image
View this table:
Fig. 2. Neutralization profiles of PGDM1400 and CAP256-VRC26.25 against multiclade pseudoviruses.

(A) Each bar represents the IC80 (μg/ml) of PGDM1400 (top graph) or CAP256-VRC26.25 (bottom graph) against a single virus. Viruses are ranked according to increasing IC80 values. Bars reaching the dotted line represent IC80 values >50 μg/ml. The red bars highlight the IC80 values against SHIV-325c. (B) IC80 values for PGDM1400 and CAP256-VRC26.25 against SHIV-325c (red dot) and against the clade C pseudoviruses included in (A). The horizontal bars represent the mean of IC80 values of neutralizable pseudoviruses.

Fig. 3. Neutralization of SHIV-325c by CAP256-VRC26.25, PGDM1400, and PG9.

Neutralization was measured using replication-competent challenge stock SHIV-325c infection of TZM-bl cells.

These data indicate that SHIV-325c had a tier 2 neutralization phenotype and a representative neutralization profile for PGDM1400 and CAP256-VRC26.25 as compared with a large panel of clade C viruses, suggesting that it may be a useful challenge model for evaluating the protective activity of these antibodies.

In vivo protection against SHIV-325c challenge

To assess the protective efficacy of PGDM1400 and CAP256-VRC26.25-LS against SHIV-325c challenge, we performed an antibody titration challenge study in rhesus macaques. CAP256-VRC26.25-LS is a variant of CAP256-VRC26.25 that has been engineered to include the LS mutation in its Fc region that increases its in vivo half-life (27) from about 4.4 to 8.8 days in rhesus macaques (fig. S2).

Thirty-five rhesus macaques were randomized to the following groups: two groups of five animals each received 2 or 0.4 mg/kg of PGDM1400, and one group of four animals received 0.08 mg/kg of PGDM1400. Three additional groups of four animals each received 2, 0.4, or 0.08 mg/kg of CAP256-VRC26.25-LS. A control group of nine animals received saline. Antibodies were administered intravenously 24 hours before the animals were challenged via the intrarectal route with a single high dose of SHIV-325c (500 TCID50). All animals in the control group became infected with detectable viremia between days 7 and 28, and the median peak viremia was 5.8 log viral copies/ml (Fig. 4A), confirming the robust infectivity of the novel SHIV-325c challenge stock. In the high-dose group (2 mg/kg), one of five animals that received PGDM1400 became infected and showed plasma viremia on day 28 (Fig. 4B), whereas all of the CAP256-VRC26.25-LS pretreated animals at the same dose were protected (Fig. 4C). At 0.4 mg/kg, no infections occurred in either group (Fig. 4, D and E). In the low-dose group (0.08 mg/kg), three of four animals that received PGDM1400 were infected (Fig. 4F), whereas none of the animals that received CAP256-VRC26.25-LS were infected (Fig. 4G). These data demonstrate that the in vitro neutralization potency of these V2-specific antibodies translates into protective efficacy in vivo at doses of 0.08 to 0.4 mg/kg.

Fig. 4. Protective efficacy of PGDM1400 and CAP256-VRC26.25-LS against SHIV-325c in rhesus macaques.

Plasma viral RNA (vRNA) (log RNA copies/ml) are shown for animals that received saline control (A), PGDM1400 (2 mg/kg) (B), CAP256-VRC26.25-LS (2 mg/kg) (C), PGDM1400 (0.4 mg/kg) (D), CAP256-VRC26.25-LS (0.4 mg/kg) (E), PGDM1400 (0.08 mg/kg) (F), or CAP256-VRC26.25-LS (0.08 mg/kg) (G). The assay sensitivity limit was >50 RNA copies/ml.

Pharmacokinetics of PGDM1400 and CAP256-VRC26.25-LS in vivo

Serum samples were obtained throughout the study, and PGDM1400 and CAP256-VRC26.25-LS concentrations were determined by enzyme-linked immunosorbent assay (ELISA). The results show that the animals that received PGDM1400 at 2, 0.4, and 0.08 mg/kg had average serum antibody concentrations of 6.9, 2.5, and 0.22 μg/ml at the time of challenge (Fig. 5). In contrast, serum concentrations of CAP256-VRC26.25-LS were 19.8, 3.3, and 0.75 μg/ml for the animals that received 2, 0.4, and 0.08 mg/kg of CAP256-VRC26.25-LS, respectively. Serum concentrations of CAP256-VRC26.25-LS were thus slightly higher than PGDM1400 (P = 0.02 to 0.03 for all dose groups) at the time of challenge, likely reflecting differences in distribution or elimination of these two antibodies. The half-life of PGDM1400 based on the level of serum antibody decay (two-phase exponential decay) was calculated to be 7.7 days for the 2-mg/kg dose group, 6.7 for the 0.4-mg/kg dose group, and 6.3 for the 0.08-mg/kg group. The half-life of CAP256-VRC26.25-LS in this study was calculated to be 10.3 days for the 2 mg/kg-dose group, 9.9 days for the 0.4-mg/kg dose group, and 7.7 days for the 0.08-mg/kg dose group, consistent with the initial experiment (fig. S2). The higher serum concentrations and more potent neutralization activity of CAP256-VRC26.25-LS compared to PGDM1400 against SHIV-325c correlated with its higher protective efficacy.

Fig. 5. Serum concentrations of CAP256-VRC26.25-LS and PGDM1400 in antibody-treated animals.

Serum antibody concentrations of CAP256-VRC26.25-LS (A) and PGDM1400 (B) were determined by ELISA. The dotted vertical line marks the day of SHIV challenge.

Sequence analysis of breakthrough viruses

Although only 4 of 14 PGDM1400 pretreated animals developed plasma viremia, we were interested to examine potential escape/resistance pathways in SHIV-325c after PGDM1400 administration. We therefore generated single-genome env sequences from plasma viruses at the time of peak viremia after breakthrough infection. Overall, there was very little sequence variation (mean divergence, 0.08 to 0.19%) between the challenge strain and the plasma viruses. We then focused on the variable loop 1 and 2 region, and no specific pattern of sequence variation was noticeable when comparing PGDM1400 and saline control animals (Fig. 6). Position 167 in V2, a known resistance mutation for CAP256 lineage antibodies (18, 28, 29), showed D/N and rare D/E changes in several of the bNAb-pretreated animals but also in the saline control animals, and the D/N variant was also found in the challenge stock. Mutations at this position (D/A) also confer resistance to PGDM1400 (table S2), suggesting that the rare D/E variation seen in the PGDM1400-pretreated animals could be an escape pathway. Similarly, one PGDM1400-pretreated animal developed a D/E mutation at position 165, which has been associated with escape from CAP256 lineage antibodies (30). Although we do not see reduced neutralization susceptibility for PGDM1400 with an I/A mutation at this position (table S2), the effect of I/E needs to be evaluated. In animal M97, which received 2 mg/kg of PGDM1400 but nevertheless developed high plasma viremia at day 28, no variant stood out in the V1V2 region, but a fixed mutation (F210L) was observed in all 27 amplicons generated from this animal (Fig. 6, mutation F210L is marked with •). This variant was absent in the challenge stock and was not detected in any other animal. Whether this mutation contributed to viral escape from PGDM1400 remains to be determined.

Fig. 6. Analysis of SHIV-325c V1V2 envelope sequences in breakthrough infections.

Single-genome envelope amplification was performed with plasma viral RNA on day 28. For one saline control animal, H515, viral amplification failed and therefore is not shown here. The env sequence at the top presents the SHIV-325c molecular clone inoculum. The column on the right reports the frequency of distinct amplicons (based on unique sequence variations in V1V2) per total env amplicons obtained from a given animal.

Complementary coverage of HIV-1 clade C viruses by V2- and V3-specific antibodies

PGDM1400 and CAP256-VRC26.25 both have potent neutralization activity against HIV-1 clade C viruses and exhibited 61 to 65% coverage of a large panel of 200 clade C env pseudoviruses (median IC80 of 1.17 and 0.19 μg/ml for PGDM1400 and CAP256-VRC26.25, respectively) (Fig. 7A). A previous study reported that the combination of PGDM1400 with the V3 glycan–dependent antibody PGT121 provided 98% coverage of multiclade viruses with a median IC50 of 0.007 μg/ml (22). We therefore determined whether the coverage of clade C viruses would improve if PGDM1400 or CAP256-VRC26.25 was combined with PGT121. PGT121, which covers 65% of the clade C virus panel at a median IC80 of 1.85 μg/ml (23), targeted many of the “gaps” in viral coverage of PGDM1400 and CAP256-VRC26.25 (Fig. 7A). PGT121 plus CAP256-VRC26.25 as well as PGT121 plus PGDM1400 showed robust coverage of >90% of clade C viruses and additive potency as compared to single bNAbs (IC80 for combinations 0.017 and 0.07 μg/ml, respectively) (Fig. 7B).

Fig. 7. Potency and breadth profiles of single and combination bNAbs against 200 clade C HIV-1 Env pseudoviruses.

(A) Heat maps of IC80 values for single bNAbs and bNAb combinations. Rows represent Env pseudoviruses, and columns represent single and combination bNAbs. Darker hues of red indicate more potent neutralization, and gray cells indicate IC80 above threshold. (B) Potency-breadth curves for single bNAbs and bNAb combinations are shown. IC80 scores for combinations and single bNAbs were compared using Wilcoxon rank-sum test. Antibody a, PGDM1400; antibody b, CAP256-VRC26.25; antibody c, PGT121.


The V2 region of the HIV-1 Env trimer is a common target of bNAbs, and some bNAbs directed against this epitope have demonstrated exquisite neutralization potency and breadth. Our findings demonstrate the capacity of the V2-specific antibodies PGDM1400 and CAP256-VRC26.25-LS to protect rhesus macaques at extremely low titers against high-dose challenge with SHIV-325c. Protection was achieved at low serum antibody concentrations of <0.75 μg/ml for CAP256-VRC26.25-LS, consistent with the extraordinary neutralization potency of these bNAbs.

Protection at such low antibody concentrations has not previously been reported (31). Previous studies have tested the protective capacity of several bNAbs, including the CD4 binding site mAbs VRC01 and 3BNC117 as well as the V3 glycan–dependant bNAbs PGT121 and 10–1074 against various SHIV strains (for example, SHIV-AD8EO, SHIV-SF162P3, and SHIV-BaL) (911). Average serum concentrations at the time of challenge necessary for protection were variable but higher than that reported here (31). The lowest serum antibody concentrations that conferred complete protection were with PGT121 at 15 and 22 μg/ml against SHIV-SF162P3 and SHIV-AD8EO challenges, respectively, and partial or no protection was seen at serum level of 1.8 μg/ml (9, 10). VRC01 only protected 4 of 10 animals against the neutralization-sensitive tier 1 virus SHIV-BaL at serum concentration of 1.3 μg/ml, whereas 10E8 only protected 3 of 6 animals against this virus at serum concentration of 1.8 μg/ml (11). The low doses of PGDM1400 and CAP256-VRC26.25-LS required for protection suggest that they could be feasibly developed for subcutaneous administration (32).

No previous studies have assessed the protective efficacy of V2-specific bNAbs against tier 2 SHIV challenges, and thus, the present studies help validate V2 as a protective target for bNAbs. PG9 is the only other V2-specific bNAb that has been evaluated for protective efficacy to date in nonhuman primates. Partial protection was achieved at 5 mg/kg, and no protection was observed at 0.3 mg/kg (serum antibody levels at day of challenge were 3.7 and 0.28 μg/ml, respectively) against the neutralization-sensitive tier 1 virus SHIV-BaL, despite a mean IC50 of 0.06 μg/ml against SHIV-BaL (11). It remains to be determined whether the inability of PG9 to achieve 100% neutralization of the challenge stock by TZM-bl assays in vitro (33) may have affected its protective efficacy in vivo.

SHIV-325c was neutralized by PGDM1400 and CAP256-VRC26.25 at IC80 values that were comparable to the median IC80 values against large global virus panels. In addition to their exquisite potency, both PGDM1400 and CAP256-VRC26.25 have 60 to 65% (IC80) of 200 tested clade C viruses. The pattern of resistant viruses was relatively similar for these two V2-specific bNAbs, suggesting that these antibodies could be combined with other bNAbs that target different epitopes to improve coverage. PGT121 has demonstrated potent antiviral activity in both therapeutic and protection studies in rhesus macaques (9, 34) and shows striking complementary coverage with both PGDM1400 and CAP256-VRC26.25-LS here. This complementarity increases viral coverage of these antibody combinations to >90% of clade C viruses (23), suggesting that PGT121 may be a useful antibody to coadminister with either CAP256-VRC26.25-LS or PGDM1400 to prevent clade C HIV-1 infection in South Africa.

This study has several limitations: We did not measure anti-PGDM1400 or anti–CAP256-VRC26.25-LS antibodies in this study to determine whether bNAb levels were affected. Although xenogeneic responses against human antibodies are commonly induced in macaques (27), no anti-antibody responses have been observed to date in clinical studies with VRC01 and 3BNC117 (35, 36). Both bNAbs protected at low serum antibody concentrations against our clade C SHIV, and although PGDM1400 and CAP256-VRC26.25 are active against 57 to 83% of global viral isolates (2022), we did not assess in vivo protection against non–clade C SHIVs. Furthermore, the addition of PGT121 to either PGDM1400 or CAP256-VRC26.25 increased viral coverage of clade C virus strains in vitro, but a small percentage of viruses remained resistant to both antibodies. A fully protective bNAb cocktail might therefore require three bNAbs to fully cover global viral diversity.

V1V2 and V3 are in close spatial proximity (37, 38), and thus, it has been suggested that coadministration of antibodies targeting these regions (for example, PGT121 together with either PGDM1400 or CAP256-VRC26.25-LS) might affect each antibody’s binding kinetics. Previous data have suggested a certain level of competition between PGT121 and the V2-specific antibody PG9 (5), but this was not observed for PGDM1400, which showed very little to no competition with PGT121 (22). Similarly, binding of CAP256-VRC26.08, which is from the same lineage as CAP256-VRC26.25, was only minimally reduced by PGT121 in a standard binding competition assay (20). These data suggest that, overall, no significant reduction in antibody binding should be expected when these V2- and V3-specific bNAbs are coadministered.

In summary, our data demonstrate that potent protective efficacy can be achieved with bNAbs that target the HIV-1 env trimer apex at very low antibody doses. PGDM1400 and CAP256-VRC26.25-LS, particularly when combined with a V3 glycan–dependent antibody such as PGT121, should therefore be explored in clinical trials for HIV-1 prevention.


Animals and study design

The overall objective of this study was to investigate the protective efficacy of two HIV-1 envelope V2 apex–specific antibodies against a clade C SHIV challenge in rhesus macaques. Two additional substudies were performed (i) to characterize infectivity and pathogenicity of the SHIV-325c challenge stock in rhesus macaques and (ii) to determine the pharmacokinetics and half-life of CAP256-VRC26.25 with the LS mutation in rhesus macaques. No randomization or blinding was performed for any study. Saline controls were used for the main protection study to measure the protective efficacy of a single infusion of PGDM1400 or CAP256-VRC26.25-LS at different doses. For this study design, a minimum of four animals per group was used similar to previous studies (10, 11, 39). The number of animals analyzed is stated in the figure legends. Primary data for the substudies can be found in table S3.

SHIV-325c infection study. Eight Indian-origin, outbred, young adult, male and female, experimentally naïve rhesus macaques (Macaca mulatta) that did not express the class I alleles Mamu-A*01, Mamu-B*08, and Mamu-B*17 associated with spontaneous virological control were housed at Bioqual Inc. Animals were atraumatically challenged via the intrarectal route with 500 TCID50 of SHIV-325c. All animals were monitored for viral loads and CD4+ T cell counts. The animal studies were approved by the appropriate Institutional Animal Care and Use Committee (IACUC). The study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (NIH).

CAP256-VRC26.25 pharmacokinetic study. All animal experiments were reviewed and approved by the Animal Care and Use Committee of the Vaccine Research Center, National Institute of Allergy and Infectious Diseases (NIAID), NIH, and all animals were housed and cared for in accordance with local, state, federal, and institute policies in an American Association for Accreditation of Laboratory Animal Care–accredited facility at the NIH. Eight Indian-origin rhesus macaques were administered low-endotoxin antibody preparations [<1 EU (endotoxin unit)/mg] intravenously at 10 mg/kg of body weight. Whole-blood samples were collected before injection and at multiple time points until week 4 after administration.

SHIV-325c protection study. Thirty-five Indian-origin, outbred, young adult, male and female, experimentally naïve rhesus macaques that did not express the class I alleles Mamu-A*01, Mamu-B*08, and Mamu-B*17 associated with spontaneous virological control were housed at Bioqual Inc. and Alphagenesis Inc. Animals were randomly allocated to the different antibody dose and saline control groups. The antibody or saline was administered intravenously 24 hours before the animals were atraumatically challenged via the intrarectal route with 500 TCID50 of SHIV-325c. Serum samples for antibody detection and viral load determination were obtained at week −1 and days 0, 1, 3, 7, 14, 28, 42, and 56. The animal studies were approved by the appropriate IACUC. The study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the NIH.

Construction of SHIV-325c molecular clone

SHIV-325c was generated from an env sequence derived from an early HIV-1–infected individual, CA325 (24, 40). The env sequence was synthesized by GeneArt (Invitrogen) and inserted into the KB9-AC plasmid as described (40, 41). The plasmid was sequenced and transfected into 293T cells using LipoD293 (SignaGen Laboratories). Cell culture supernatants were collected after 48 hours and clarified through a 0.45-μm filter.

Generation of large-scale SHIV stocks

For the generation of the SHIV-325c virus stock, PBMCs were isolated from 120 ml of deidentified human buffy coats that were purchased. Cell culture supernatants harvested from transiently transfected 293T cells were used to inoculate to ConA-stimulated PBMCs in the presence of human interleukin-2 (20 U/ml; AIDS Research and Reference Reagent Program). The medium was replaced and collected every 3 days. Virus was quantified by SIV p27 ELISA (Zeptometrix), and the TCID50 was determined in TZM-bl cells (40). Virus was aliquoted and stored at −80°C.

Antibody production

PGDM1400 was generated as previously described (5) and purified by Protein A affinity matrix (GE Healthcare). CAP256-VRC26.25-LS was purified after transient transfection of Expi293 cells (Invitrogen) with expression vectors encoding for its heavy and light chains. Phosphate-buffered saline (PBS) was used as control in this study. The CAP256-VRC26.25-LS antibody included the Fc region mutation (LS) (27) that increased circulating half-life.

Synthesis and purification of the CNE58-strandC-CAP256.SU SOSIP trimer for ELISA

HIV-1 gp140 SOSIP-type molecule based on the clade C strain CNE58 (4244), including the “SOS” mutations (A>501C and T605C), the isoleucine-to-proline mutation at residue 559 (I559P), the mutation of the cleavage site to 6R (REKR to RRRRRR), and the truncation of the C terminus to residue 664 (all HIV-1 Env numbering according to the HX nomenclature), was used in this study. The CNE58 SOSIP construct also used a chimera strategy by using the gp41 and N and C termini of BG>505 and a part of the V1V2 C strand (residues 166 to 173) from the CAP256.SU strain (20, 28) and a C-terminal six–amino acid glycine-serine linker, HRV-3c cleavage site, His8 purification tag (GGSGGSGLEVLFQGPGHHHHHHHH). The HIV-1 SOSIP molecule was expressed as previously described (45) by cotransfecting with furin in human embryonic kidney–293F cells using HIV-1 SOSIP DNA and furin plasmid DNA. Transfection supernatants were harvested after 6 days, and the CNE58-strandC-CAP256.SU HIV-1 trimer supernatant was passed over a nickel–nitrilotriacetic acid affinity column. The eluate was concentrated, and the peak corresponding to trimeric HIV-1 Env was identified using size exclusion chromatography, pooled, and concentrated.

Enzyme-linked immunosorbent assay

PGDM1400 and CAP256-VRC26.25-LS antibody concentration in serum was determined by ELISA as described (46) using the BG505 SOSIP trimer or the novel CNE58-strandC-CAP256.SU SOSIP trimer, respectively. Briefly, microtiter plates were coated with SOSIP trimer (1 μg/ml) and incubated overnight at 4°C. The plates were washed with PBS/0.05% Tween 20 and blocked with PBS/1% casein (Pierce). After blocking, serial dilution of serum samples was added to the plate and incubated for 2 hours at 37°C. Binding was detected with a horseradish peroxidase–conjugated goat anti-human immunoglobulin G (IgG) secondary antibody (Fisher Scientific) and visualized with SureBlue tetramethylbenzidine microwell peroxidase (KPL Research Products).

Neutralization assays

Neutralization of diverse, replication-competent SHIV challenge stocks (including SHIV-325c) or Env pseudoviruses was evaluated in vitro by using TZM-bl target cells and a luciferase reporter assay as described (11, 47, 48). Briefly, HIV-1 Env pseudoviruses were generated by transfection in 293T cells, and SHIV challenge stocks were produced by transfection of 293T cells with infectious molecular clone plasmids, followed by propagation in human or rhesus PBMCs, as described above and as indicated in Table 1. The rhesus PBMC–derived stock of SHIV-AD8 was provided by M. Martin (12), and SHIV-1157ipd3N4 was obtained through the AIDS Reagent Program, Division of AIDS, NIAID, NIH from R. Ruprecht. SHIV stocks or HIV-1 pseudoviruses were incubated with the antibody for 30 min at 37°C before TZM-bl cells were added. The protease inhibitor indinavir was added to assays with SHIV stocks to a final concentration of 1 μM to limit infection of target cells to a single round of viral replication. Luciferase expression was quantified 48 hours after infection upon cell lysis and the addition of luciferin substrate (Promega).

SGA and analysis

SGA followed by direct sequencing of the Env gene was used to eliminate Taq-induced errors and in vitro recombination as described (49). Briefly, viral RNA was isolated from plasma using a QIAamp viral RNA kit (Qiagen). Reverse transcription of RNA to single-stranded complementary DNA (cDNA) was performed using SuperScript III Reverse Transcriptase according to the manufacturer’s recommendations (Invitrogen). In brief, a cDNA reaction of 1× RT buffer, 0.5 mM of each deoxynucleotide triphosphate (dNTP), 5 mM dithiothreitol, RNase OUT [RNase (recombinant ribonuclease) inhibitor] (2 U/ml), SuperScript III Reverse Transcriptase (10 U/ml), and 0.25 mM antisense primer EnvB3out 5′-TTGCTACTTGTGATTGCTCCATGT-3′ was performed. RNA, primers, and dNTPs were heated at 65°C for 5 min and then chilled on ice for 1 min; then, the entire reaction was incubated at >50°C for 60 min, followed by 55°C for an additional 60 min. Finally, the reaction was heat-inactivated at 70°C for 15 min and then treated with 1 μl of RNase H each at 37°C for 20 min. Then, cDNA templates were serially diluted until only a fraction (~25%) of amplicons are polymerase chain reaction (PCR)–positive under the following PCR conditions: PCR amplification was carried out using the Platinum Taq (Invitrogen) with 1× buffer, 2 mM MgCl2, 0.2 mM each dNTP, 0.2 μM each primer, and 0.025 U/μl Platinum Taq polymerase. The primers for the first-round PCR were EnvB5out 5′-TAGAGCCCTGGAAGCATCCAGGAAG-3′ and EnvB3out 5′-TTGCTACTTGTGATTGCTCCATGT-3′. The primers for the second-round PCR were EnvB5in 5′-CACCTTAGGCATCTCCTATGGCAGGAAGAAG-3′ and EnvB3in 5′-GTCTCGAGATACTGCTCCCACCC-3′. The cycler parameters were 94°C for 2 min, followed by 35 cycles of 94°C for 15 s, 55°C for 30 s, and 68°C for 4 min, followed by a final extension of 68°C for 10 min. The product of the first-round PCR (1 μl) was subsequently used as a template in the second-round PCR under the same conditions but with a total of 45 cycles. All PCR-positive amplicons were directly sequenced using BigDye Terminator chemistry (Applied Biosystems). Any sequence with evidence of double peaks was excluded from further analysis.

Statistical analyses

Analyses of independent data were performed by two-tailed Mann-Whitney U tests, Wilcoxon rank-sum test, or paired t test. P values less than 0.05 were considered significant. Combination IC80 titers were calculated using the Bliss-Hill model, which has been shown to provide accurate predictions of combination neutralization properties using those of individual bNAbs (23). Statistical analyses were performed using GraphPad Prism or the Stats module in Scipy (


Fig. S1. Highlighter amino acid sequence alignment of env derived from the SHIV-325c stock and the parental HIV-1 env.

Fig. S2. Serum concentration of CAP256-VRC26.25-LS and CAP256-VRC26.25 in naïve, uninfected rhesus macaques.

Table S1. IC80 (μg/ml) values of PGDM1400 and CAP256-VRC26.25 against SHIV-325c and a multiclade panel of 208 HIV-1 pseudoviruses.

Table S2. Neutralization of JR-CSF alanine variants by PGDM1400 or PG9.

Table S3. Primary data for Fig. 1 and fig. S2.


  1. Acknowledgments: We thank M. Lewis and W. Rinaldi for clinical conduct of the animal studies and J. Baalwa, D. Ellenberger, F. Gao, B. Hahn, K. Hong, J. Kim, F. McCutchan, D. Montefiori, L. Morris, J. Overbaugh, E. Sanders-Buell, G. Shaw, R. Swanstrom, M. Thomson, S. Tovanabutra, C. Williamson, and L. Zhang for the HIV-1 envelope plasmids. Funding: This work was supported by the NIH (AI106408, AI096040, AI100663, AI124377, AI126603, and HHSN261200800001E), amfAR (109219), the Ragon Institute of Massachusetts General Hospital, Massachusetts Institute of Technology, and Harvard University, and the intramural research program of the Vaccine Research Center, NIAID, NIH. The content of this publication does not necessarily reflect the views or policies of the U.S. Department of Health and Human Services, nor does the mention of trade names, commercial products, or organizations imply endorsement by the U.S. government. Author contributions: Project planning was performed by D.H.B., D.R.B., J.R.M., R.A.K., L.M., and S.S.A.K; the SHIV-325c was developed by L.J.T. and P.A.; antibodies were generated by A.P., P.L.M., D.S., and X.C.; viral neutralization assays were performed and analyzed by M.K.L., R.T.B., A.P., D.S., S.D.S., N.A.D-R., and K.L.; env sequences were generated and analyzed by B.F.K.; plasma viral loads were measured by P.A.; pharmacokinetic ELISA was performed by A.P., D.S., and K. Wang; synthesis and purification of CNE58-strandC-CAP256.SU SOSIP trimer for ELISA were performed by X.C., M.G.J., I.S.G., M.C., and P.D.K.; PGDM1400 alanine scanning mutants were produced by D.S.; bNAb coverage/breadth analysis was performed by K. Wagh, M.S.S., and B.K.; and the manuscript was written by B.J., D.R.B., L.M., J.R.M., and D.H.B. Competing interests: The authors declare that they have no competing interests. Data and materials availability: The data presented in this paper are tabulated in the main paper and in the Supplementary Materials. Materials are available with an appropriate material transfer agreement.

Stay Connected to Science Translational Medicine

Navigate This Article