Preclinical studies of human immunodeficiency virus/AIDS vaccines: inverse correlation between avidity of anti-Env antibodies and peak postchallenge viremia

Abstract

A major challenge for human immunodeficiency virus (HIV)/AIDS vaccines is the elicitation of anti-Env antibodies (Ab) capable of neutralizing the diversity of isolates in the pandemic. Here, we show that high-avidity, but nonneutralizing, Abs can have an inverse correlation with peak postchallenge viremia for a heterologous challenge. Vaccine studies were conducted in rhesus macaques using DNA priming followed by modified vaccinia Ankara boosting with HIV type 1 (HIV-1) immunogens that express virus-like particles displaying CCR5-tropic clade B (strain ADA) or clade C (IN98012) Envs. Rhesus granulocyte-macrophage colony-stimulating factor was used as an adjuvant for enhancing the avidity of anti-Env Ab responses. Challenge was with simian/human immunodeficiency virus (SHIV)-162P3, a CCR5-tropic clade B chimera of SIV and HIV-1. Within the groups receiving the clade B vaccine, a strong inverse correlation was found between the avidity of anti-Env Abs and peak postchallenge viremia. This correlation required the use of native but not gp120 or gp140 forms of Env for avidity assays. The high-avidity Ab elicited by the ADA Env had excellent breadth for the Envs of incident clade B but not clade C isolates, whereas the high-avidity Ab elicited by the IN98012 Env had excellent breadth for incident clade C but not clade B isolates. High-avidity Ab elicited by a SHIV vaccine with a dual-tropic clade B Env (89.6) had limited breadth for incident isolates. Our results suggest that certain Envs can elicit nonneutralizing but high-avidity Ab with broad potential for blunting incident infections of the same clade.

FIG. 1.

Postchallenge control of SHIV-162P3 viral RNA and protection of CD4 T cells. (A) Titers of viral RNA and frequencies of CD4 T cells at the indicated time points. Titers of viral RNA in blood are geometric means ± SDs. Frequencies of CD4 T cells are means ± SDs. Data for colorectal CD4 T cells are available for only some time points due to restrictions on the frequency of biopsies. (B) Levels of viremia in different groups. Box plots provide graphic summaries for peak viral RNA and areas under the curve (0 to 24 weeks). The white lines in the boxes are medians; the upper and lower limits of the boxes indicate the 75th and 25th percentiles, respectively. Minimum and maximum values are indicated by the upper and lower brackets. For numbers of animals per group, see Table 1.

FIG. 2.

Responding T cells measured by intracellular cytokine staining for IFN-γ. CD8 (A) and CD4 (B) responses over time measured as IFN-γ producing cells as a percentage of total CD4 or CD8 T cells are presented as geometric means ± SDs. The bar graphs on the left of figures show peak responses after each MVA boost following stimulation with consensus clade B peptide pools for Gag or Env. The line graphs show postchallenge responses to pools of consensus clade B (Con B) Env peptides, 162P3 Env peptides, consensus clade B Gag peptides, and SIV239 Gag peptides (indicated at the tops of panels). Symbols for groups are to the right of panels. The shaded blue area indicates responses below the limit for accurate quantification.

FIG. 3.

Anti-Env binding Ab and avidity for the indicated time points. (A) Titers of binding Ab relative to a standard curve for macaque IgG. (B) Avidities of binding Ab. Horizontal lines in panel A indicate geometric means; in panel B, horizontal lines indicate medians. All assays were conducted using Env captured from VLPs produced in transient transfections by the clade-matched DNA vaccine. Vaccine groups are indicated at the tops of panels. Prechallenge sera were not taken for the naïve group. These responses were likely similar to those in preimmune macaques in other groups. Pre imm, preimmunizations; 2nd MVA, 2 weeks after the second MVA boost; pre ch, prechallenge; and +3 wk, 3 weeks postchallenge.

FIG. 4.

Neutralizing Ab production over time. Neutralization titers are the reciprocal for serum dilutions at which relative luminescence units were reduced 50% compared to virus control wells. The threshold value of 20 was used for negative values. This threshold is shown in shaded blue. The isolates used in neutralization assays are indicated at the tops of panels. Symbols for groups are identified at right. MN, HIV-1 MN; SF162, HIV-1 SF162; ID₅₀, 50% infectious dose.

FIG. 5.

Inverse correlation between the avidity of anti-Env binding Ab and peak viremia. Inverse correlations between peak viremia and the avidity index for ADA Env captured from VLPs for Ab in serum harvested prechallenge (A) or at 3 weeks postchallenge (B). (C) Inverse correlation between peak viremia and the 162P3 challenge Env captured from pseudovirions for Ab harvested at 3 weeks postchallenge. There was no correlation between peak viremia and the avidity index of Ab harvested at 3 weeks postchallenge for ADA gp140 (D) or ADA gp120 (E) forms of Env. (F) Lack of a correlation between peak viremia and the relative titers of binding Ab at 3 weeks postchallenge. Correlations were conducted for the 15 animals in clade B-vaccinated groups using the Pearson method. The form of ADA Env used in assays and the sera being tested in the assay (+3 weeks, 3 weeks postchallenge) are given at the tops of the panels. See Materials and Methods for assays. ns, not significant.

FIG. 6.

Intraclade but not interclade activity of high-avidity Ab elicited by clades B and C HIV-1 vaccines. Avidity tests were conducted with pooled sera from GM-CSF-adjuvanted immunizations. Panels A, C, and E show avidities of vaccine-elicited sera for the Env in the vaccine and its challenge. Panels B and D show avidities of the clade B and C vaccine-elicited sera for clade B incident Envs (B) and for clade C incident Envs (D). Panel F shows avidities of clade B and SHIV-89.6 vaccine-elicited sera for clade B incident Envs. Color codes for the vaccines are identified at right; the Envs tested in avidity assays are identified at the tops of panels. PM, serum harvested after the second MVA boost; PC, serum harvested at 3 weeks postchallenge.