Adenoviral vectors persist in vivo and maintain activated CD8+ T cells: implications for their use as vaccines

Abstract

CD8(+) T cell-numbers rapidly expand and then contract after exposure to their cognate antigen. Here we show that the sustained frequencies of transgene product-specific CD8(+) T cells elicited by replication-defective adenovirus vectors are linked to persistence of low levels of transcriptionally active adenovirus vector genomes at the site of inoculation, in liver, and lymphatic tissues. Continuously produced small amounts of antigen maintain fully active effector CD8(+) T cells, while also allowing for their differentiation into central memory cells. The long-term persistence of adenoviral vectors may be highly advantageous for their use as vaccines against pathogens for which T-cell-mediated protection requires both fully activated T cells for immediate control of virus-infected cells and central memory CD8(+) T cells that, because of their higher proliferative capacity, may be suited best to eliminate cells infected by pathogens that escaped the initial wave of effector T cells.

Figure 1

Kinetics of the Gag-specific CD8⁺ T-cell response. (A) Top panels: Groups of 5 BALB/c mice injected intramuscularly with 10¹¹ vp of AdC68gag37 or 5 × 10¹¹ vp of AdHu5gag37 vector were killed at different times. Lymphocytes isolated from indicated compartments were tested for Gag-specific IFN-γ–producing CD8⁺ cells by intracellular cytokine staining and analyzed by 2-color flow cytometry. Results show frequencies of CD8⁺ cells producing IFN-γ as a percentage of all CD8⁺ cells from a representative experiment (top panels). Frequencies of Gag-specific CD8⁺ T cells in BALB/c mice immunized with 5 × 10¹¹ vp of AdC68gag37 and boosted 2 months later with 10⁶ plaque forming units (pfu) of a recombinant vaccinia virus expressing gag (VVgag). Weeks after immunization refer to the boost. Each experiment was conducted 2 to 4 times and the data shown are representative. (B) The same experiment was conducted with cells from mice immunized with 5 × 10¹¹ vp of Adhu5gag37 vector. C57Bl/6 mice were infected with 2 × 10⁵ pfu of LCMV intraperitoneally or immunized with 10¹⁰ vp of Adhu5 expressing LCMV glycoprotein intramuscularly. Db/GP33 tetramer-positive CD8 T cells were enumerated in the peripheral blood mononuclear cells by MHC tetramer staining at the indicated time points. n is 3 for LCMV and n is 2 for Adhu5-GP33. Similar results were observed for Adhu5-GP given intravenously or intraperitoneally (data not shown). (C) Mice were vaccinated with 10¹¹ vp of E1-deleted, E1-deleted, and E3-deleted or E1-deleted, E3-deleted, and E4-deleted AdC8 vector expressing Gag37. T-cell frequencies of spleens of individual mice were analyzed at different times. The graphs show frequencies of Gag-specific CD8⁺ T cells for individual mice (■) and mean frequencies (X) (± SD).

Figure 2

Persisting presence of vector sequences. (A) Adult female outbred (ICR) mice received 10¹¹v.p.AdC68rab.gp vector in saline, given once intramuscularly. On days 4, 30, 90, and 360 after vector application, mice were killed and perfused with cold PBS. Tissues were harvested from individual mice. DNA was isolated and the Gapdh sequences were amplified by a real-time PCR. The samples were adjusted to 10³ copies of Gapdh and the rabies virus glycoprotein gene was amplified by a real-time nested PCR. (B) Mice were vaccinated intramuscularly in the lower leg with 10¹¹ particles of an Ad vector expressing green fluorescent protein. Mice were killed 1 day and 39 days after vaccination and legs were illuminated with an Illumatool Lighting System. Digital photographs were taken using a Kodak DCS14N SLR camera with a 60-mm Micro Nikkor lens (Nikon). (C) DNA and RNA were isolated from spleens of mice immunized at different times previously with 1011 vp of AdC8gag37. RNA was reverse-transcribed. Gapdh was quantified from each sample by real-time PCR. Samples were adjusted before amplification to 6 × 10⁷ or 1.5 × 10⁹ copies of Gapdh DNA or cDNA, respectively, and amplified by a nested PCR. After amplification by the internal real-time PCR, each sample was analyzed by gel electrophoresis. The lower graph shows results for Gag DNA or cDNA. The arrow indicates the anticipated size of the Gag amplicon. These results are from a single experiment in which we ran multiple gels to accommodate the samples. Lines were added to show where the lanes were cut. (D) Mice were immunized with AdC68gag37. Twenty months later they were boosted with 10⁶ pfu of a vaccinia virus vector expressing Gag to increase frequencies of Gag-specific CD8⁺ T cells. In pre-experiments it was shown that Gag sequences from the vaccinia virus vector could not be amplified from spleens as of 1 week after inoculation. Five weeks after the boost, splenocytes were isolated. Cells were stained with a Gag-specific tetramer (tet) and an antibody to CD8. Cells were sorted into CD8⁺tet⁻ cells, CD8⁻tet⁻ cells and CD8⁺tet⁺ cells. Total cellular RNA was isolated and purified from each cell fraction. Complementary DNA was synthesized. The HIV Gag gene in each cell fraction was amplified first by regular PCR. The amplicon from the first PCR product was then used as template for a second real-time PCR to quantify the Gag gene in different cell fractions. The copy numbers of Gag in each cell fraction were normalized in comparison to Gapdh sequences quantified by a real-time PCR from the same samples. (E) Monkeys 18, 48, 140, and 145 were immunized intramuscularly with 10¹² vp of AdC7gag37 vector and boosted 8 months later with 10¹² vp of AdC6gag37 vector. Animal 164 was injected at the same time with the same vector backbone expressing the rabies virus glycoprotein. DNA from peripheral blood mononuclear cells harvested 99 days after the boost was tested for gag DNA as described in panel A on adjustments of samples to 3 × 10⁴ β-actin molecules. These results are from a single experiment; the lanes were rearranged to change the order of the samples. Lines were added to show where the lanes were cut. (F) The graph shows the gels from a hexon-specific nested PCR that was used to amplify vector sequences from peripheral blood mononuclear cells of vaccinated rhesus macaques. Lanes 1 and 2 show results from peripheral blood mononuclear cells harvested 6 (1) and 17 (2) weeks after intramuscular vaccination of a monkey with 10¹¹ vp of a mixture of 4 AdC68 vectors expressing HIV-1 Gag, gp140, 5′pol, or 3′pol⁺nef. Lanes 3 and 4 show results from peripheral blood mononuclear cells from a monkey harvested 6 (3) and 17 (4) weeks after intramuscular immunization with 10¹¹ vp of 4 AdHu5 vectors expressing the same antigens as used for results in lanes 1 and 2. Lanes 5 through 8 show results from peripheral blood mononuclear cells from 2 animals harvested 2 (5,7) and 14 (6,8) weeks after immunization with 10¹¹ vp of an AdHu5 vector expressing Gag. Lanes 9 and 10 show the negative control of the PCR reactions.

Figure 3

Vector persistence drives proliferation of antigen-specific T cells. C57Bl/6 Thy1.1 mice were immunized with 10¹⁰ vp of either AdC68NPOVA green fluorescent protein or AdC68rab.gp vector (3 mice per group). After 4 months, 3 × 10⁷ CFSE-labeled splenocytes isolated from OT1 mice were transferred into each of the immunized mice. One week after transfer the mice were killed and lymphocytes were isolated from the indicated tissues, stained with anti-CD8 and anti-Thy1.2 to identify donor cells, and the amount of CFSE in those cells was determined by flow cytometry. The numbers within the graphs show the percent of cells with low CFSE staining as indicated by the gates.

Figure 4

Cytolytic activity of Gag-specific CD8⁺ T cells. Groups of BALB/c mice were immunized with 5 × 10¹¹ vp of AdC68gag37 and tested 2 months later. (1) Additional mice were primed with 5 × 10¹⁰ vp of AdC68gag37 and boosted with 5 × 10¹⁰ vp of AdHu5gag37 and tested 3 months after the boost (2). Target cells (P815) were coated overnight with 5 μg/mL Gag peptide (■) or an equal concentration of a control peptide (□). Before the assay, frequencies of Gag-specific CD8⁺, T cells were determined by MHC-gag peptide tetramer staining. Lymphocytes were adjusted to equal numbers of the tetramer-positive CD8⁺ cells and cocultured at serial dilutions with ⁵¹Cr-labeled target cells for 6 hours. Supernatants were harvested and tested in a gamma counter. Percent specific lysis was calculated as described. The origin of the lymphocytes is indicated in the upper part of the graphs. LN, lymph nodes; PL, peritoneal lavage. Error bars show standard deviations for the percent of specific lysis tested in 3 separate samples.

Figure 5

Kinetics of memory CD8 T-cell differentiation. (A-C) Mice were either infected with LCMV Armstrong (2 × 10⁵ pfu intraperitoneally), or VV-GP33 (2 × 10⁶ pfu intraperitoneally), or immunized with 10¹⁰ vp of AdHu5-GP33 intramuscularly, intravenously, or intraperitoneally (similar results were observed by all routes). Two groups of mice were used. In the first group, a small number of naive splenocytes from P14 mice (≈5 × 10⁴ P14 cells) was adoptively transferred to naive B6 mice. The next day these P14 chimeras were infected or immunized. These are designated T-cell receptor Tg (P14). A second group of wild-type B6 mice was also used for confirmation of the P14 results (B6 non-Tg). Db/GP33-specific CD8 T-cell responses were monitored in the peripheral blood mononuclear cells of individual mice over time by MHC tetramer staining in conjunction with staining for CD62L and CD127. (A,B) The kinetics of CD62L and CD127 reversion on the population of Db/GP33 tetramer-positive CD8 T cells is plotted for each condition including P14 chimeras and B6 mice. For each group, n is 2 to 9. Error bars in panels A and B show standard deviations between samples from individual nice. (C) Examples of individual Db/GP33 tetramer staining and CD62L expression on Db/GP33 tetramer⁺ CD8 T cells are shown for each condition at 2 time points. The numbers show percent double positive cells over all cells positive for a given marker for the dot blots and percent of cells that were expressed high levels of a given marker as indicated by the brackets in the histograms. (D) Groups of BALB/c mice were immunized intramuscularly with 10¹⁰ vp of AdC68HIVgag37 for the acute cohort and 5 × 10¹¹ vp for the memory cohort; they were killed at day 10 (acute) or 15 months (memory) after immunization. Groups of C57BL/6 mice were immunized intraperitoneally with 2 × 10⁵ pfu LCMV Armstrong and killed on day 8 (acute) and 7 months (memory) after immunization. Lymphocytes isolated from indicated tissues were stained with a HIVgag tetramer for the AdC68HIVgag37–immunized mice and a GP33 tetramer for the LCMV mice. Cells were also stained with anti-CD8 and the indicated surface markers, and analyzed by flow cytometry. Results shown are gated on tetramer-positive cells (black line) or naive cells (filled curve).