Immunogenicity and Cross-Reactivity of Rhesus Adenoviral Vectors

Abstract

Adenovirus (Ad) vectors are being investigated as vaccine candidates, but baseline antivector immunity exists in human populations to both human Ad (HuAd) and chimpanzee Ad (ChAd) vectors. In this study, we investigated the immunogenicity and cross-reactivity of a panel of recently described rhesus adenoviral (RhAd) vectors. RhAd vectors elicited T cells with low exhaustion markers and robust anamnestic potential. Moreover, RhAd vector immunogenicity was unaffected by high levels of preexisting anti-HuAd immunity. Both HuAd/RhAd and RhAd/RhAd prime-boost vaccine regimens were highly immunogenic, despite a degree of cross-reactive neutralizing antibodies (NAbs) between phylogenetically related RhAd vectors. We observed extensive vector-specific cross-reactive CD4 T cell responses and more limited CD8 T cell responses between RhAd and HuAd vectors, but the impact of vector-specific cellular responses was far less than that of vector-specific NAbs. These data suggest the potential utility of RhAd vectors and define novel heterologous prime-boost strategies for vaccine development.IMPORTANCE To date, most adenoviral vectors developed for vaccination have been HuAds from species B, C, D, and E, and human populations display moderate to high levels of preexisting immunity. There is a clinical need for new adenoviral vectors that are not hindered by preexisting immunity. Moreover, the development of RhAd vector vaccines expands our ability to vaccinate against multiple pathogens in a population that may have received other HuAd or ChAd vectors. We evaluated the immunogenicity and cross-reactivity of RhAd vectors, which belong to the poorly described adenovirus species G. These vectors induced robust cellular and humoral immune responses and were not hampered by preexisting anti-HuAd vector immunity. Such properties make RhAd vectors attractive as potential vaccine vectors.

Keywords: HIV; adenovirus; rhesus; simian immunodeficiency virus; vector.

FIG 1

RhAd vector-induced cellular immunologic phenotype. Mice were immunized i.m. with 10⁹ vp of the indicated adenoviral vectors. (A) Phylogenetic trees showing the full-genome (left) and hexon (right) relationships among various HuAds, ChAd24, RhAd52, RhAd53, and RhAd56. (B) Longitudinal analysis of D^b/AL11 tetramer-positive, PD-1⁺, and KLRG1⁺ CD8⁺ T cells from PBMCs. (C) Frequency of D^b/AL11⁺, PD-1⁺, and KLRG1⁺ CD8⁺ T cells from splenocytes. (D) Frequency of IFN-γ⁺ CD8⁺ T cells from splenocytes. Blue bars indicate splenocytes stimulated with the SIV_mac239 Gag peptide pool, and red bars are unstimulated samples. Box-and-whisker plots indicate minimum and maximum values. For all experiments, data are for 8 to 12 mice per group. Lines above the graphs denote significance: solid bars, P < 0.0001; dotted lines, P < 0.01. Error bars indicate the standard error of the mean (SEM).

FIG 2

RhAd vectors in prime-boost regimens. C57BL/6 mice (n = 8 to 10/group) were primed with 10⁹ vp of the indicated adenoviral vector. After 8 weeks, mice were boosted with the vector shown in the key. (A) Longitudinal analysis of D^b/AL11⁺ CD8⁺ T cells in PBMCs of vaccinated mice. For Ad26 prime, P was <0.0001 RhAd53 versus Ad26 (****) and P was equal to 0.0464 for RhAd53 versus ChAd24 (*); for RhAd52 prime, P was equal to 0.0142 for RhAd53 versus Ad26 (*). (B) Ad-specific neutralization titers 4 weeks after prime, before boosting vaccinations were administered. (C) Ad-specific neutralization titers 4 weeks after the boosting vaccinations were administered. Dotted lines indicate the limit of detection. Error bars indicate the standard error of the mean (SEM). For tetramer analysis, 8 to 10 mice per group were used, and for neutralization data, 4 to 8 mice per group were used. IC90, maximum serum dilution that neutralizes 90% of virus.

FIG 3

RhAd vector-induced antibody binding titers. (A) C57BL/6 mice were immunized with 10⁹ vp of the indicated adenoviral vectors (n = 5/group). Antibody binding titers are shown for weeks 0, 2, 4, and 8 after vaccination. (B) C57BL/7 mice were primed with Ad26-Env and 8 weeks later were boosted with the indicated adenoviral vectors (n = 5/group). Antibody binding titers are shown for weeks 0, 2, and 8 postprime and weeks 10, 12, and 22 postboost. Dots represent individual animals. Dotted lines above the graphs denote significance (P < 0.01). Error bars indicate the standard error of the mean (SEM).

FIG 4

RhAd vector immunogenicity in mice with baseline Ad5 immunity. (A) Experimental schema. C57BL/6 mice (n = 50) were immunized at week −8 and week −4 with 10⁹ vp of Ad5-empty. At week 0, mice were injected with the indicated vector expressing either Gag or Env (n = 5/group). (B) Ad5-neutralizing antibody titers at weeks −8, −4, and 0. (C) Longitudinal analysis of D^b/AL11 tetramer binding responses following the priming immunization with the indicated Gag-encoding vector. (D) Antibody binding titers for weeks 1, 2, and 4 after priming immunization with the indicated Env-encoding vectors. Error bars indicate the standard error of the mean (SEM).

FIG 5

RhAd prime-boost regimens in mice with baseline Ad5 immunity. (A) Experimental study design (n = 4/group). C57BL/6 mice were primed with the indicated vector and then boosted 8 weeks later with the boosting vector denoted in the key. Both immunizations were done at 10⁹ vp. (B) Frequency of D^b/AL11⁺ CD8⁺ T cells. Priming responses were pooled and displayed as one line (brown) for Ad26-Gag and RhAd52-Gag. (C) Ad-specific neutralizing antibody titers 4 weeks after boosting immunization for Ad5, Ad26, RhAd52, and RhAd53. Dotted horizontal lines indicate the limit of detection. Error bars indicate the standard error of the mean (SEM).

FIG 6

Suppression of RhAd52 immunogenicity with baseline RhAd immunity. (A) Experimental study design (n = 5/group). C57BL/6 mice were injected with 10⁹ vp of various Ad-empty vectors either once or twice to induce low or high levels of preexisting immunity. Mice were then vaccinated with 10⁹ vp of RhAd52-Gag. (B) Frequency of D^b/AL11⁺ CD8⁺ T cells following RhAd52-Gag vaccination following one injection of Ad-empty (left) or two injections of Ad-empty (right). For two injections, P was equal to 0.0079 for PBS versus RhAd52 (**), P was equal to 0.015 for PBS versus RhAd56 (*), and P was equal to 0.055 for PBS versus RhAd53. (C) Neutralizing antibody titers 4 weeks after the first (top) or 4 weeks after the second (bottom) empty vector injection, but before RhAd52 vaccination. Dotted horizontal lines indicate the limit of detection. Error bars indicate the standard error of the mean (SEM).

FIG 7

Suppression of RhAd immunogenicity by adoptive transfer of RhAd52-specific IgG. (A) Experimental schema (n = 5/group). IgG was purified from pooled serum from mice injected with RhAd52-empty or naive mice, and 500 μg of IgG was transferred to naive recipient mice. One day after transfer, recipient mice were vaccinated with 10⁹ vp of RhAd52-Gag, RhAd53-Gag, RhAd56-Gag, or Ad26-Gag (n = 5/group). (B) Neutralizing antibody titers 1 day after adoptive transfer, but before vaccination. (C) Frequency of D^b/AL11⁺ CD8⁺ T cells following adoptive transfer for each vaccine group (*, P < 0.05; **, P < 0.01). Error bars indicate the standard error of the mean (SEM).

FIG 8

Partial suppression of RhAd immunogenicity by adoptive transfer of RhAd52-specific splenocytes. (A) Frequency of IFN-γ⁺ CD4⁺ and CD8⁺ T cells responding to peptide pools of 15-mers overlapping by 11 amino acids from the hexon regions of Ad5, Ad26, RhAd52, and RhAd53 from mice injected twice with the indicated Ad-empty vector or PBS control. (B) Experimental schema (n = 5/group). Splenocytes were pooled from mice that were injected twice with RhAd52-empty or from naive mice. Donor splenocytes were transferred into naive hosts, and 1 day later they were vaccinated with either RhAd52-Gag or RhAd56-Gag (n = 5/group). (C) Frequency of D^b/AL11⁺ CD8⁺ T cells following cell transfer and vaccination. Error bars indicate the standard error of the mean (SEM).