These studies demonstrate striking qualitative differences in the lineages of the cellular responses elicited by each of the evaluated vaccine modalities. The rAd5-induced responses predominantly involved CD8+ T lymphocytes, while the plasmid DNA-elicited cells were predominantly CD4+ T lymphocytes. Although these biases did not change following homologous boosting, they changed substantially following heterologous boosting, with the development of high-frequency, balanced responses.
We also evaluated the functional profiles of the virus-specific CD8+ T cells by measuring their IFN-γ, IL-2, and TNF-α production. The expression of several other molecules has been used by other investigators to assess the function of T cells, including β-chemokines (macrophage inflammatory protein 1β [MIP-1β]) and molecules that are associated with cytolytic activity (CD107). Previous studies from our laboratory and the laboratories of others have shown that as HIV/SIV-induced disease progresses, HIV/SIV-specific CD8+ T cells first lose their ability to produce IL-2 and later TNF-α, while their ability to produce IFN-γ and MIP-1β and their expression of CD107 can be preserved at the very late stages of disease. We therefore decided not to evaluate CD107 and MIP-1β expression in the present study, and we used a relatively limited set of anti-cytokine MAbs to evaluate the functional profile of the virus-specific T cells.
Monkeys receiving rAd5, rPox, and plasmid DNA developed virus-specific polyfunctional CD8+
T-lymphocyte responses following the initial administration of a single vaccine immunogen, although there was some bias toward IFN-γ-only production in the rPox-immunized monkeys. This finding of an IFN-γ bias differs from recent reports of polyfunctionality in rPox-induced cellular immune responses in humans (16
). However, importantly, when rAd5 or rPox vectors were employed as boosting immunogens following plasmid DNA priming, the profile of the CD8+
T-lymphocyte response was mostly polyfunctional.
Although the different vaccination regimens generated qualitatively different virus-specific T-cell populations, those differences were lost following the virus challenge. While the T lymphocytes of the control vaccinees made only low-frequency virus-specific responses, both the CD8+ and CD4+ T-lymphocyte responses in all groups of experimentally vaccinated monkeys were robust. Further, the profile of cytokine production by the virus-specific T lymphocytes in the control monkeys was heavily biased toward cells that produce only IFN-γ, while the virus-specific T lymphocytes of all of the experimentally vaccinated monkeys following challenge were uniformly polyfunctional. Most importantly, we observed no significant differences between any of the cohorts of vaccinated monkeys following virus challenge in the magnitudes of their virus-specific cellular immune responses, the biases of those responses to CD8+ T lymphocytes, and the functional profiles of those cells. This uniformity of virus-specific T lymphocytes in the vaccinated monkeys likely reflects the overwhelming influence on lymphocyte differentiation of the high levels of viral antigen present in these animals.
Consistent with these findings, we observed lower plasma viral RNA levels and better preservation of naïve CD4+ T lymphocytes in the vaccinated than in the control monkeys. However, there was no significant difference in these clinical parameters between the various groups of experimentally vaccinated monkeys. Thus, while the immunologic profiles of the vaccine-elicited T lymphocytes differed considerably between these groups of vaccinees, the critical cellular immune responses that were mounted following virus infection and the clinical consequences of those infections were indistinguishable between the groups.
Interestingly, when all animals that received different vaccine immunizations were grouped together and divided into two groups based on the magnitudes of their peak plasma viral RNA levels, highly significant differences were observed in both the quantity and quality of the vaccine-elicited virus-specific CD8+ T cells between these two groups. This finding was not apparent when analyzing each cohort of monkeys separately. These observations suggest that both the quantity and quality of the vaccine-induced immunodeficiency virus-specific CD8+ T-cell responses were associated with control of viral replication.
CD8+ T cells mediate multiple effector functions during acute HIV infection but become exhausted and lose their ability to produce some cytokines with the persistence of viral antigenemia during chronic HIV infection. Although many studies provide insights into how the functional capacity of the HIV-specific CD8+ T cells correlates with the clinical course of disease progression, it remains unclear whether the maintenance of CTLs with a “polyfunctional” profile in long-term nonprogressors is an epiphenomenon associated with good viral control or is the mechanism responsible for the good clinical status. Although we found that both the magnitude and polyfunctionality of the vaccine-elicited CD8+ T-cell responses were predictive of the viral set point after infection, it is still possible that polyfunctionality is a consequence of vaccine take in these monkeys.
Virus replication is contained by cellular immune responses during the first days following infection. The magnitude of that cellular response is modified by prechallenge vaccine-elicited cellular immune responses. However, the findings in the present study suggest that the comparable viral antigen load in all cohorts of vaccinated monkeys during the primary infection was associated with comparable functional repertoires of the cellular immune responses generated in response to the infection. These findings likely explain why no significant differences were observed between the different cohorts of vaccinated monkeys in the magnitudes and the functional profiles of their virus-specific CD8+ T cells following viral infection.
There are certainly important caveats that must be acknowledged when interpreting these findings. The available macaque challenge systems do not perfectly model HIV-1 infection in humans. More specifically, the CXCR4-tropic SHIV-89.6P used in these challenge studies causes a disease that is very different than that caused by CCR5-tropic strains of SIV and HIV-1. SHIV-89.6P preferentially infects CXCR4+ naïve CD4+ T cells, whereas SIVmac251 and HIV-1 mainly target activated memory CD4+ T cells. In macaque models, T-cell-based vaccines provide long-term protection against SHIV-89.6P. However, the same live recombinant vaccines only provide short-term control of viral replication after a SIVmac251 challenge. Moreover, the recent STEP human clinical trial suggests that the dramatic protection seen in the SHIV-89.6P macaque model may not correlate with clinical protection against HIV infection in humans. Nevertheless, the findings in the present study raise the possibility that differences in both the magnitude and quality of vaccine-elicited CD8+ T cells generated by different vaccine modalities may be of importance for controlling virus replication.