Effect of preexisting immunity to adenovirus on transgene product-specific genital T cell responses on vaccination of mice with a homologous vector

Abstract

We evaluated changes in global and human immunodeficiency virus (HIV)-specific genital T cells after vaccination of female mice with a replication-defective adenovirus vector of human serotype 5 (AdHu5) expressing Gag protein of HIV-1, in the presence or absence of preexisting immunity to the vaccine carrier. Our data show that preexisting immunity causes a rapid and transient decrease of genital CD4(+) T cells without increasing the expression of chemokine (C-C motif) receptor 5. Furthermore, preexposure to AdHu5 affects long-term alterations in the magnitude and quality of vaccine-induced Gag-specific CD8(+) T cell responses. AdHu5-specific antibodies interfere with the induction of transgene product-specific CD8(+) T cell responses in systemic compartments, whereas some mechanism other than antibodies also seems to affect those that home to the genital tract.

Figure 1.

Frequencies and numbers of genital CD4⁺ and CD8⁺ T cells. Graphs show percentages of CD4⁺ (A) and CD8⁺ (B) T cells, as well as absolute numbers of CD4⁺ (C) and CD8⁺ (D) T cells within the genital tract at different intervals after vaccination with AdHu5gag. Graphs show average frequencies or cell counts ± standard deviations (SD) for animals preexposed to adenovirus vector of human serotype 5 (AdHu5) and vaccinated with AdHu5gag 3 weeks later (i.m./i.m.; circles), mice vaccinated with the AdHu5gag vector without preexposure (-/i.m.; closed squares), and naive mice (open squares). Cells were isolated from the female genital tract and stained with a live cell stain and antibodies to CD4 and CD8. Statistical analyses compared 2 individual experiments, using 2-way analysis of variance followed by Bonferroni test. A, Both vaccinated groups had significantly lower frequencies of CD4⁺ cells over time than did the naive control group (P < .01). Frequencies of CD4⁺ cells for the 2 vaccinated groups differed only at week 8 after vaccination (P < .05). B, Frequencies of CD8⁺ cells were reduced in the 2 vaccinated groups early after vaccination (P < .01). Frequencies of CD8⁺ T cells in the nonpreexposed group were increased between weeks 1 and 8, compared with the naive group (P < .01). C, CD4⁺ cell counts were significantly different on days 3 and 7 between either of the vaccinated groups and the control group (P < .05 by t test). D, Cell counts were significantly different for CD8⁺ T cells only at week 1 when the 2 vaccinated groups were compared with the naive group (P < .05).

Figure 2.

Phenotypes of genital CD4⁺ and CD8⁺ T cells. A, Phenotypes of CD4⁺ cells. B, Phenotypes of CD8⁺ cells. Cells from 5 mice were isolated at different time points, pooled, and stained with a live cell stain and antibodies to CD4, CD8a, and the markers indicated in the graphs. Gray lines represent data from preexposed and vaccinated mice (i.m./i.m.); black lines, data from vaccinated mice that were not preexposed (-/i.m.); and gray areas at bottom of graphs, data for cells isolated from naive animals. Student's t test was performed to compare percentages of difference in median fluorescence intensity for samples derived from 2 individual experiments, of which 1 is displayed here. *P < .05 for comparison between immunized and naive groups; †P < .05 for comparison between immunized groups. CCR5, chemokine (C-C motif) receptor 5.

Figure 3.

Effect of preexposure on frequencies of Gag-specific CD8⁺ T cells. Graph shows the kinetics of Gag-specific CD8⁺ T cell responses in different compartments after vaccination. Mice were preexposed to adenovirus vector of human serotype 5 (AdHu5) and vaccinated with AdHu5gag 3 weeks later (i.m./i.m.) (closed bars); a control group received the vector intramuscularly without preexposure (-/i.m.) (open bars). Frequencies represent ratios of tetramer (Tet)⁺CD8⁺ cells to total CD8⁺ cells (means ± standard deviations from 2 individual experiments). Values obtained from naive animals were subtracted before plotting frequencies for the 2 test groups. Statistical analysis (2-way analysis of variance followed by Bonferroni correction) shows reduction in Gag-specific CD8⁺ T cell frequencies in mice on preexposure to the vaccine vector in iliac lymph nodes (ILN) (P < .05), spleen, and the genital tract (GT) (P < .01).

Figure 4.

Effect of immune serum transfer on Gag-specific CD8⁺ T cell frequencies. Mice were injected intramuscularly twice with AdHu5rab.gp, and serum samples were collected 2 weeks after the boost dose. Pooled immune serum was transferred intraperitoneally into naive mice, followed 1 week later by intramuscular vaccination with AdHu5gag (serum/i.m.). Control groups consisted of animals that received serum from sham-vaccinated mice (-/i.m.) or were preexposed to adenovirus vector of human serotype 5 (AdHu5) before vaccination with AdHu5gag (i.m./i.m.). Frequencies represent ratios of tetramer (Tet)⁺CD8⁺ cells to total CD8⁺ cells (means frequencies ± standard deviations from 2 individual experiments). Values obtained from naive mice tested in parallel were subtracted from frequencies of vaccinated mice before graphs were plotted. Statistical analysis (2-way analysis of variance followed by Bonferroni posttest) shows that frequencies of tetramer⁺CD8⁺ cells were significantly lower in the preexposed group than in the serum-transfer group for all compartments analyzed (P < .05). The serum-transfer group had significantly lower frequencies of Gag-specific CD8⁺ T cells in blood, spleen, and iliac lymph nodes (ILN) (P < .01), but differences in frequencies did not reach statistical significance for cells isolated from the genital tract (GT).

Figure 5.

Phenotypes of Gag-specific CD8⁺ T cells. Mice preexposed to adenovirus vector of human serotype 5 (AdHu5) (i.m./i.m.) (gray lines) and nonpreexposed mice (-/i.m.) (black lines) were vaccinated with AdHu5gag. Cells from different compartments were analyzed 1, 6, and 12 weeks after vaccination. Cells were stained with a live cell stain, the Gag-specific tetramer, antibodies to CD8a, and the markers indicated. Graphs show expression of markers on tetramer (Tet)⁺CD8⁺ T cells isolated from vaccinated mice. Gray shaded areas show expression of the markers on tetramer −CD8⁺ T cells from naive mice. Graphs presenting samples from blood and spleens show cells isolated from 5 individual mice, and those from iliac lymph nodes (ILN) show samples pooled from 2 or 3 mice; data from these samples were concatenated using FlowJo software. Graphs of samples derived from the genital tract show results obtained with pooled lymphocytes from 5 mice. Student's t test was used to compare percentage differences in median fluorescence intensity between samples derived from 2 individual experiments, of which 1 is displayed here. *P < .05 for difference between immunized and naive mice; †P < .05 for difference between the 2 immunized groups. CCR5, chemokine (C-C motif) receptor 5.