Characterization of HIV-1 gp120 antibody specificities induced in anogenital secretions of RV144 vaccine recipients after late boost immunizations

Abstract

Sexual transmission is the principal driver of the human immunodeficiency virus (HIV) pandemic. Understanding HIV vaccine-induced immune responses at mucosal surfaces can generate hypotheses regarding mechanisms of protection, and may influence vaccine development. The RV144 (ClinicalTrials.gov NCT00223080) efficacy trial showed protection against HIV infections but mucosal samples were not collected, therefore, the contribution of mucosal antibodies to preventing HIV-1 acquisition is unknown. Here, we report the generation, magnitude and persistence of antibody responses to recombinant gp120 envelope and antigens including variable one and two loop scaffold antigens (gp70V1V2) previously shown to correlate with risk in RV144. We evaluated antibody responses to gp120 A244gD and gp70V1V2 92TH023 (both CRF01_AE) and Case A2 (subtype B) in cervico-vaginal mucus (CVM), seminal plasma (SP) and rectal secretions (RS) from HIV-uninfected RV144 vaccine recipients, who were randomized to receive two late boosts of ALVAC-HIV/AIDSVAX®B/E, AIDSVAX®B/E, or ALVAC-HIV alone at 0 and 6 months. Late vaccine boosting increased IgG geometric mean titers (GMT) to gp120 A244gD in AIDSVAX®B/E and ALVAC-HIV/AIDSVAX®B/E CVM (28 and 17 fold, respectively), followed by SP and RS. IgG to gp70V1V2 92TH023 increased in AIDSVAX®B/E and ALVAC-HIV/AIDSVAX®B/E CVM (11-17 fold) and SP (2 fold) two weeks post first boost. IgG to Case A2 was only detected in AIDSVAX®B/E and ALVAC-HIV/AIDSVAX®B/E CVM. Mucosal IgG to gp120 A244gD (CVM, SP, RS), gp70V1V2 92TH023 (CVM, SP), and Case A2 (CVM) correlated with plasma IgG levels (p<0.001). Although the magnitude of IgG responses declined after boosting, anti-gp120 A244gD IgG responses in CVM persisted for 12 months post final vaccination. Further studies in localization, persistence and magnitude of envelope specific antibodies (IgG and dimeric IgA) in anogenital secretions will help determine their role in preventing mucosal HIV acquisition.

Conflict of interest statement

Competing Interests: I have read the journal's policy and the authors of this manuscript have the following competing interests: Faruk Sinangil is employees of Global Solutions for Infectious Diseases. James Tartaglia and Sanjay Phogat are employees of Sanofi Pasteur. This does not alter our adherence to PLOS ONE policies on sharing data and materials. All other authors declare no potential conflicts of interest.

Figures

Fig 1. Specimen collection algorithm.

Blood specimens were collected from each study participant to obtain plasma. Cervico-vaginal mucus (CVM) was collected from consenting female participants. Seminal plasma (SP) and rectal secretions (RS) were collected from consenting male participants. Blood contamination was tested on both CVM and RS using Hemoccult® SENSA® test kit (Beckmann Coulter, Brea, CA). All blood contaminated specimens were excluded from the analysis. Blood contamination was not tested on SP.

Fig 2. Percent of positive IgG responders to all HIV antigen tested in anogenital secretions.

Percent of positive IgG responders to gp120 A244gD (CRF01_AE), gp70V1V2 92TH023 (CRF01_AE) and gp70V1V2 Case A2 (subtype B) in CVM (A-C), SP (D-F), and RS (G-I) are shown. Each time point is color coded; red, week 0; orange, week 2; green, week 26; blue, week 48; magenta, week 72. CVM and RS were not collected from placebo recipient of groups AIDSVAX®B/E and ALVAC-HIV, respectively, at any time point. VAC = vaccine recipients; PLB = placebo recipients; ALVAC/AIDSVAX = ALVAC-HIV/AIDSVAX®B/E group; AIDSVAX = AIDSVAX®B/E group; ALVAC = ALVAC-HIV group.

Fig 3. IgG binding antibody responses to gp120 A244gD and gp70V1V2 scaffolds in cervico-vaginal mucus (CVM).

Reciprocal titers of IgG binding antibody responses to (A) gp120 A244gD (CRF01_AE), (B) gp70V1V2 92TH023 (CRF01_AE) and (C) gp70V1V2 Case A2 (subtype B) in CVM are shown along with numeric depiction of geometric mean titers above panels. Each group is color coded; red, ALVAC-HIV/AIDSVAX®B/E (ALVAC/AIDSVAX); green, AIDSVAX®B/E (AIDSVAX); blue, ALVAC-HIV (ALVAC). Error bars depict 95% confidence intervals. The cut-off level of responses (0.5-fold the reciprocal titers of initial dilution of specimens) is shown by the dotted line. RV305 vaccine administration time points are indicated by black arrows (weeks 0 and 24). The non-parametric Mann-Whitney U Test was used to assess within-group comparison of IgG responses between time points indicated by horizontal black bars. Comparisons reaching statistical significance at the level of p<0.05 are shown. *p<0.05 to 0.001, &p<0.001.

Fig 4. IgG binding antibody responses to gp120 A244gD and gp70V1V2 scaffolds in seminal plasma (SP).

Reciprocal titers of IgG binding antibody responses to (A) gp120 A244gD (CRF01_AE), (B) gp70V1V2 92TH023 (CRF01_AE) and (C) gp70V1V2 Case A2 (subtype B) in SP are shown along with numeric depiction of geometric mean titers above panels. Each group is color coded; red, ALVAC-HIV/AIDSVAX®B/E (ALVAC/AIDSVAX); green, AIDSVAX®B/E (AIDSVAX); blue, ALVAC-HIV (ALVAC). Error bars depict 95% confidence intervals. The cut-off level of responses (0.5-fold the reciprocal titers of the initial dilution of specimens) is shown by the dotted line. RV305 vaccine administration time points are indicated by black arrows (weeks 0 and 24). The non-parametric Mann-Whitney U Test was used to assess within-group comparison of IgG responses between time points indicated by black bars. Comparisons reaching statistical significance at the level of p<0.05 are shown. *p<0.05 to 0.001, &p<0.001.

Fig 5. IgG binding antibody responses to gp120 A244gD and gp70V1V2 scaffolds in rectal secretions (RS).

Reciprocal titers of IgG binding antibody responses to (A) gp120 A244gD (CRF01_AE), (B) gp70V1V2 92TH023 (CRF01_AE) and (C) gp70V1V2 Case A2 (subtype B) in RS are shown along with numeric depiction of geometric mean titers above panels. Each group is color coded; red, ALVAC-HIV/AIDSVAX®B/E (ALVAC/AIDSVAX); green, AIDSVAX®B/E (AIDSVAX); blue, ALVAC-HIV (ALVAC). Error bars depict 95% confidence intervals. The cut-off level of responses (0.5-fold the reciprocal titers of the initial dilution of specimens) is shown by the dotted line. RV305 vaccine administration time points are indicated by black arrows (weeks 0 and 24). The non-parametric Mann-Whitney U Test was used to assess within-group comparison of IgG responses between time points indicated by black bars. Comparisons reaching statistical significance at the level of p<0.05 are shown. *p<0.05 to 0.001, &p<0.001.

Fig 6. Correlation of IgG responses in anogenital secretions and plasma.

Spearman’s rank correlations of IgG responses for gp120 A244gD (CRF01_AE), gp70V1V2 92TH023 (CRF01_AE) and gp70V1 V2 Case A2 (subtype B) at weeks 2, 26, 48 and 72 in matched CVM (A-C), SP (D-E) and RS (F), and plasma of RV305 vaccine recipients are shown. Each group is color coded; red, ALVAC-HIV/AIDSVAX®B/E; green, AIDSVAX®B/E; blue, ALVAC-HIV. Numeric values above each plot depict r- and p-values. Significant p-value <0.05, GMT = Geometric Mean Titer.