Long trimer-immunization interval and appropriate adjuvant reduce immune responses to the soluble HIV-1-envelope trimer base

Abstract

Soluble 'SOSIP'-stabilized HIV-1 envelope glycoprotein (Env) trimers elicit dominant antibody responses targeting their glycan-free base regions, potentially diminishing neutralizing responses. Previously, using a nonhuman primate model, we demonstrated that priming with fusion peptide (FP)-carrier conjugate immunogens followed by boosting with Env trimers reduced the anti-base response. Further, we demonstrated that longer immunization intervals further reduced anti-base responses and increased neutralization breadth. Here, we demonstrate that long trimer-boosting intervals, but not long FP immunization intervals, reduce the anti-base response. Additionally, we identify that FP priming before trimer immunization enhances antibody avidity to the Env trimer. We also establish that adjuvants Matrix M and Adjuplex further reduce anti-base responses and increase neutralizing titers. FP priming, long trimer-immunization interval, and an appropriate adjuvant can thus reduce anti-base antibody responses and improve Env-directed vaccine outcomes.

Keywords: Immune response; Immunology; Virology.

Graphical abstract

Figure 1

In FP primed/trimer boosted NHPs, long trimer interval, but not long FP prime interval, significantly impacts plasma anti-base responses For all regimens, NHP plasma at three key time points of two weeks after the fifth FP immunization, the first trimer immunization and the second trimer immunization were analyzed (indicated with a red arrow). Two groups of five animals each were first primed three times with FP8-KLH followed by FP7-KLH, FP6-KLH, and two BG505 DS-SOSIP immunizations, with Adjuplex and I.M. Green and black circles and lines represent NHPs from short interval groups and long-interval groups, respectively. (A) One group of five animals were immunized with short intervals at weeks 0, 4, 8, 12, 20, 24, and 28, and one group of five animals were immunized with long intervals at weeks 0, 4, 20, 32, 44, 56, and 64. (B) Plasma binding to BG505 DS-SOSIP by ELISA as measured by OD450nm plotted against plasma dilution factors for short and long interval groups at the three key time points of two weeks post fifth FP (left panel), two weeks post first trimer (second panel from left) and two weeks post second trimer (right panel). (C) Total trimer response as measured by ELISA BG505 DS-SOSIP AUC from the values in B. (D) Comparison of the total ELISA OD450nm values for binding to BG505 DS-SOSIP with (hollow circles) and without (solid circles) competition with RM19R Fab between the short and long interval groups at dilutions of 1:20, 1:50, and 1:500 for the three time points, respectively. (E) Percentage of response targeting the base region calculated from the values in D. The percentage was calculated by taking the difference between OD450 values without and with RM19R Fab, divided by the OD450 values without RM19R Fab blocking, and multiplied by 100. (F) Plasma binding by ELISA to glycan-base BG505 DS-SOSIP trimer, which has glycans added on the base region. (G) Trimer response as measured by AUC from the values in F. Two-tailed Mann-Whitney nonparametric tests were used for statistical analysis to assess p values for mean ± SEM. ∗: p < 0.05; ∗∗: p < 0.01; ∗∗∗: p < 0.001; ∗∗∗∗: p < 0.0001; ns, not significant. The data in C represent two independent experiments with similar results. See also Figure S1.

Figure 2

In FP primed/trimer boosted NHPs, long trimer interval after FP prime enhances FP responses and neutralizing breadth Green circles represent NHPs from short-interval group, and black circles represent NHPs from long-interval group. (A – C) Correlation analysis of the percentage of anti-base response based on RM19R Fab competition assay and non-base trimer response detected with glycan-base BG505 DS-SOSIP at the three key time points of two weeks post fifth FP immunization (A), two weeks post first trimer immunization (B) and two weeks post second trimer immunization (C). (D) Correlation of the AUC values of non-base trimer response detected with glycan-base BG505 DS-SOSIP and neutralizing breadth at two weeks post second trimer immunization. Neutralizing breadth is the number of neutralized wild-type viruses within a 10-virus panel in which all viruses contain the same FP8 sequence and are sensitive to FP-specific antibodies. (E) Anti-FP plasma response at the three key time points detected with ELISA assay as shown with mean ± SEM. Two-tailed paired t test was performed for post first and second trimer immunization for short-interval group. Two-tailed Mann-Whitney nonparametric tests were used for statistical analysis to assess p values for short and long interval groups at each time points. (F) Correlation of anti-FP8 ELISA endpoint titers obtained two weeks post second trimer immunization and the percentage of base response on BG505 DS-SOSIP at two weeks post second trimer immunization by RM19R Fab competition. ET stands for endpoint titer. (G) Correlation of anti-FP8 ELISA endpoint titers obtained two weeks post second trimer immunization and the AUC values of non-base trimer response detected with glycan-base BG505 DS-SOSIP at two weeks post second trimer immunization. ET stands for endpoint titer. (H and I) Correlation between Anti-FP8 ELISA endpoint titers and the percentage of base response as determined by BG505 DS-SOSIP competition ELISA with RM19R Fab two weeks post fifth FP immunization (H) and two weeks post first trimer immunization (I). (J and K) Correlation between anti-FP8 ELISA endpoint titers two weeks post fifth FP immunization (J), two weeks post first trimer immunization (K) and the percentage of base response two weeks post second trimer immunization, as determined by BG505 DS-SOSIP competition ELISA with RM19R Fab. Two-tailed Pearson correlation coefficient test was performed for correlation analysis. ∗: p < 0.05; ∗∗: p < 0.01.

Figure 3

Long trimer interval reduces plasma anti-base responses in trimer-only immunized NHPs but does not enhance neutralizing activity For all regimens, NHP plasma at three key time points of two weeks after the first, second, and third trimer immunizations were analyzed. Red circles and lines represent NHPs from short-interval group, and black circles and lines represent NHPs from long-interval group. (A) Immunization regimen of short-interval and long-interval NHPs in trimer only immunization. Six NHPs were immunized with BG505 DS-SOSIP at weeks 0, 8, and 24 with S.Q (long interval), while five NHPs were immunized with BG505 DS-SOSIP at weeks 0, 4, and 16 with I.M. (short-interval). Both groups used Alum as adjuvant. (B) Plasma response to BG505 DS-SOSIP trimer at three key time points. (C) Total trimer response as measured by ELISA BG505 DS-SOSIP AUC values. (D) Comparison of the total ELISA OD450nm values for binding to BG505 DS-SOSIP with (hollow circles) and without (solid circles) competition with RM19R Fab between the short and long interval groups at dilutions of 1:20, 1:500, and 1:500 for the three time points, respectively. Two-tailed paired-t test were used for statistical analysis to assess p values with and without Fab competition, with mean ± SEM. (E) Percentage of trimer response targeting the base region calculated from the values in D. The percentage was calculated by taking the difference between OD450 values without and with RM19R Fab, divided by the OD450 values without RM19R Fab blocking, and multiplied by 100. (F) Plasma binding to glycan-base BG505 DS-SOSIP trimer, which has glycans added on the base region. (G) Plasma responses as measured by glycan-base BG505 DS-SOSIP trimer AUC from F. (H) Neutralizing ID50 on BG505 and MW965 at two weeks post third trimer immunization. (I) Anti-V3 peptide response at two weeks post third trimer immunization. Two-tailed Mann-Whitney nonparametric tests were used for statistical analysis for D-I, to assess p values for mean ± SEM. ∗: p < 0.05; ∗∗: p < 0.01; ∗∗∗: p < 0.001; ∗∗∗∗: p < 0.0001; ns, not significant. The data in C represent two independent experiments with similar results. See also Figures S1–S3.

Figure 4

Differential binding to BG505 DS-SOSIP with and without NaSCN shows that NHPs immunized with FP prime followed by trimer boost elicit antibodies with higher avidity compared to trimer-only immunized NHPs (A and B) Plasma binding to BG505 DS-SOSIP without (solid line) or with NaSCN (dashed line) by ELISA as measured by OD450nm for each group are plotted against plasma dilution factors for FP prime+trimer boost group (A) and trimer only immunized group (B) at the three key time points of two weeks post fifth FP immunization (left panel), two weeks post first trimer (second panel from left) and two weeks post second trimer (right panel) for (A); two weeks post first trimer (left panel), two weeks post second trimer (second panel from left) and two weeks post third trimer (right panel) for (B). (C) Percentage of plasma binding reduction with NaSCN, calculated with AUC values from A and B for FP primer/trimer boost NHPs (left) and trimer only NHPs (right). (D) Percentage of plasma binding reduction in FP prime/trimer boost NHPs (solid triangle) and trimer only NHPs (empty triangle). FP prime/trimer boost with short or long intervals are combined as “FP prime/trimer boost” group, and trimer only groups with short or long intervals are combined as “trimer only” group. The reduction was calculated by taking the difference between AUC values without and with NaSCN divided by the AUC values without NaSCN and multiplied by 100. Two-tailed Mann-Whitney nonparametric tests were used for statistical analysis to assess p values for mean ± SEM. ∗: p < 0.05; ∗∗: p < 0.01; ∗∗∗: p < 0.001; ∗∗∗∗: p < 0.0001; ns, not significant.

Figure 5

Frequency of non-base binding trimer-specific IgG+ B cells inversely correlates with the prevalence of anti-base responses detected in plasma (A) JG2 trimer showed similar binding to broadly neutralizing antibody panels and V3-panel, while significantly reduced binding to the anti-base monoclonal antibodies, detected with Octet. (B) Immunization schema of NHP study G1 (black), with CH505 DS-SOSIP trimer primed at weeks 0 and 4, and blood drawn at week 6, and Immunization schema of NHP study G2 (green), with BG505 DS-SOSIP trimer primed at weeks 0, 8, and 24, and blood drawn at week 26. G1 and G2 showed 93% and 8% anti-base response within total trimer response, respectively, detected with ELISA RM19R Fab competition assay. FACS analysis was performed using BG505 DS-SOSIP and JG2 trimer probes for G1 and G2 PBMC. (C) Frequency of IgG+ B cells that bind both BG505 DS-SOSIP trimer and JG2 glycan-base trimer in IgG+ B cells. (D) Frequency of BG505+JG2+ B cells within BG505+ IgG + B cells. (E) Frequencies of BG505+JG2+ B cells correlate inversely with percentage of trimer-base response detected with ELISA RM19R Fab competition assay. Two-tailed Mann-Whitney nonparametric tests were used for statistical analysis to assess p values for mean ± SEM. ∗: p < 0.05; ∗∗: p < 0.01; ∗∗∗: p < 0.001; ∗∗∗∗: p < 0.0001; ns, not significant. Correlation analysis was performed using a two-tailed Pearson correlation coefficient test. See also Figure S4.

Figure 6

In long-interval trimer immunization, adjuvant impacts neutralizing titers and anti-base response For all regimens, NHP plasma at two key time points of two weeks after the 2nd and 3rd trimer immunizations were analyzed. Red, green, blue and purple circles represent NHPs with Alum, Matrix M, GLA-LSQ and TQL-1055, respectively. (A) Immunization regimen. Six NHPs per group were immunized with BG505 DS-SOSIP at weeks 0, 8, and 24, S.Q. with different adjuvants. (B) Plasma response to BG505 DS-SOSIP trimer at two weeks post 2nd trimer and 3rd trimer immunizations. (C) Anti-base plasma responses as measured by BG505 DS-SOSIP ELISA with RM19R Fab competition (1:500 dilution, OD450nm). (D) Total trimer response as measured by glycan-base BG505 DS-SOSIP trimer AUC values. (E) Percentages of IgG+ B cell that bind both BG505 DS-SOSIP trimer and JG2 trimer probes in PBMC after 3rd trimer immunization. (F) Percentages of BG505+JG2+ B cells correlate with plasma response to glycan-base BG505 trimer. (G and H) Plasma response against V3 peptide (G) and FP8 peptide (H) at 2 weeks post 2nd and 3rd trimer immunization. (I) ID50 titers on BG505 post 2nd and 3rd trimer immunization. (J) Plasma anti-FP (left) and V3 (right) responses correlate with BG505 ID50, with all data from different groups combined. Data are shown with mean ± SEM for multiple comparison analysis, with Kruskal-Wallis test followed with two-stage linear step-up procedure of Benjamini, Krieger and Yekutieli tests to assess p values between different adjuvant groups. ∗: p < 0.05; ∗∗: p < 0.01; ∗∗∗: p < 0.001; ∗∗∗∗: p < 0.0001; ns, not significant. The Data in B–F represent 2 independent experiments with similar results. Correlation analysis was performed using a two-tailed Pearson correlation coefficient test. See also and Figures S1, S4, S5, and S7.

Figure 7

In short-interval trimer immunization, adjuvant impacts total trimer response and anti-base response For all regimens, NHP plasma at two key timepoints of two weeks after the second and third trimer immunization were analyzed. Red, green and blue circles represent Alum, Adjuplex and Iscomatrix groups, respectively. (A) Immunization regimen. Five NHPs per group were immunized with BG505 DS-SOSIP trimer at weeks 0, 4, and 16, I.M. with different adjuvants. (B) Total trimer response as measured by ELISA BG505 DS-SOSIP AUC values. (C) Percentage of anti-base response measured via competition with RM19R Fab. (D) Trimer response as measured by glycan-base BG505 DS-SODIP. (E) Correlation between total trimer response and anti-base response, calculated using ELISA assay with RM19R Fab competition assay (percentages, left) or glycan-base trimer (right) in ELISA assay. (F) Plasma response to BG505 V3 peptide (left) and FP peptide (right) as measured by ELISA two weeks post third trimer immunization. (G) Correlation between total trimer response vs. anti-V3 (left) and anti-FP (right) response measured by ELISA two weeks post third trimer. (H) ID50 neutralization titers against BG505 (left) and MW965 (right) two weeks post third trimer immunization. Dotted lines represent the starting dilution at 1:10. (I) Correlation between MW965 ID50 neutralization titers and anti-V3 peptide response (left) and glycan-base BG505 DS-SOSIP trimer response (right) measured via ELISA. Data are shown with mean ± SEM for multiple comparison analysis, with Kruskal-Wallis test followed with two-stage linear step-up procedure of Benjamini, Krieger and Yekutieli tests to assess p values between different adjuvant groups. ∗: p < 0.05; ∗∗: p < 0.01; ∗∗∗: p < 0.001; ∗∗∗∗: p < 0.0001; ns, not significant. The data in B–F represent 2 independent experiments with similar results. Correlation analysis was performed using a two-tailed Spearman correlation coefficient test. See also Figures S1, S6, and S7.