Research Briefs

By Roberto Fernandez-Larsson, Ph.D.*  

Env immunogens contribute to AIDS vaccine protection in simian models

Following infection with HIV, neutralizing antibodies and cytotoxic T lymphocytes (CTLs) are raised to specific Env epitopes. But global Env diversity poses a significant challenge for the development of an effective AIDS vaccine, and the utility of including an Env immunogen as a vaccine component has been questioned.

In previous studies, vaccine-elicited Env-specific cellular immune responses have protected monkeys challenged with both simian immunodeficiency virus (SIV) and simian-human immunodeficiency virus (SHIV) expressing Envs identical to the immunogen. Norman Letvin and colleagues have now looked to see if an HIV Env immunogen contributes to protective immunity against challenge with a pathogenic SHIV-89.6P having a genetically disparate Env (J. Virol. 78, 7490; 2004).

The vaccination strategy employed a plasmid DNA prime and a recombinant replication-defective adenovirus (rADV) boost. To determine whether an Env immunogen induces protection against a genetically disparate challenge SHIV, six monkeys were vaccinated with one of three different DNA plus rADV strategies: (i) Mock (Gag-Pol-Nef with no Env), (ii) Env-matched (Gag-Pol-Nef with SHIV-89.6P Env), and (iii) Env-mismatched (Gag-Pol-Nef with HXB2/Bal Env). All monkeys were then challenged 12 weeks later with SHIV-89.6P. A control group of six unvaccinated naïve monkeys were also challenged with SHIV-89.6P.

All controls suffered a profound loss in CD4+ T cells, which is the usual outcome of SHIV-89.6P infection in naïve animals. Four of six monkeys vaccinated with Gag-Pol-Nef with no Env were able to mitigate CD4+ T-cell loss after SHIV-89.6P challenge, as expected from previous studies. The two groups of vaccinated monkeys that received Env (matched or mismatched) in addition to Gag-Pol-Nef immunogens putatively demonstrated a significantly better mitigation of CD4+ T-cell loss than the group of monkeys that received only Gag-Pol-Nef immunogens without Env. Viral replication in monkeys after SHIV-89.6P challenge was monitored by quantitating viral RNA load in plasma by using a bDNA assay. Here they found that the unvaccinated control animals had significantly higher peak viral loads than the vaccinated animals. However, there was no significant difference between the three vaccinated groups, whether they received Env immunogen or not.

Env-immunized monkeys developed high-titer antibodies to their cognate Env (SHIV-89.6P Env or SHIV-HXB2/Bal Env), but plasma samples from them failed to neutralize SHIV-89.6P virus in vitro, so they were unable to demonstrate that neutralizing antibodies directed against SHIV-89.6P contributed to viral containment after challenge. However, peripheral blood mononuclear cells from monkeys that received either matched or mismatched Env immunogens developed several-fold higher ELISPOT responses to SHIV-89.6P and HXB2/Bal, respectively, than did the monkeys that received the mock Env injections.

The authors conclude that the results suggest a strong association between the generation of postchallenge Env-specific T-cell immunity and the inclusion of either matched or mismatched Env immunogens in the vaccine regimes of the tested monkeys. But given the limits of the small numbers of animals providing comparison in the paper, the only apparent advantage of including Env is the broadening of the vaccine-elicited antiviral cellular immune responses. It is also of interest that the responses are cross-reactive.

Vaccine-induced CTLs contain highly pathogenic immunodeficiency virus infection

The significance of CTLs in the control of immunodeficiency virus infections has been demonstrated by associations between CTL activity and by control of viremia in primary HIV infection in humans, and CD8+ T-cell depletion experiments in SIV-infected macaque models. However, it remains unclear if vaccine-induced CTL responses could control chronic SIV disease progression or, ultimately, HIV replication in humans.

Tetsuro Matano and colleagues (J. Exp. Med. 199, 1709; 2004) now provide evidence that vaccine-induced CTLs can result in the containment of SIVmac239 infection in non-Indian macaques; the vast majority of previous studies have used Indian macaques. The authors say that the challenge virus they use, SIVmac239, is a more realistic challenge virus than SHIV-89.6P because the former induces chronic disease progression (analogous to HIV infection) rather than the acute CD4+ T-cell depletion induced by the latter.

Eight rhesus macaques were vaccinated with a DNA prime followed by a single boost with a recombinant Sendai virus vector (SeV) 6 weeks later; both vaccine components expressed SIVmac239 Gag. All macaques were challenged intravenously with SIVmac239 13 weeks after the SeV boost. Four unvaccinated control macaques developed high peak viremia on day 10 after challenge and maintained relatively high plasma viral concentrations, while five out of the 8 vaccinated macaques controlled replication of the highly pathogenic challenge virus; plasma viremia became undetectable after week 5 and peripheral CD4+ T cells were maintained.

At week 2 after challenge, the investigators detected anamnestic Gag-specific CD8+ T-cell responses in all of the vaccinated macaques, indicating efficient secondary responses during the acute phase of infection. They found no neutralizing activities in plasma against SIVmac239 in any of the controls or the vaccinees at weeks 5 or 12 after challenge, suggesting that neutralizing antibodies were not essential for the control of SIV replication observed. The SIV gag region in the viral genomes obtained from plasma RNA at week 5 after challenge was sequenced to determine whether vaccine-induced Gag-specific T-cell responses exerted a selective pressure on the virus. The numbers of amino acid changes per clone in the vaccinated macaques were significantly higher than those in the unvaccinated, which may reflect the immune pressure exerted by vaccine-induced Gag-specific T-cell responses.

All of the macaques that controlled SIVmac239 replication showed consistent amino acid changes in Gag, some of which conferred diminished replication efficiency of the viruses, in vitro as well as in vivo, compared with the wild-type SIVmac239.

The authors conclude that vaccine-elicited CTLs can “cripple” the virus by imposing a fitness cost, and results in the containment of replication of a neutralization-resistant, highly pathogenic immunodeficiency virus that is not contained in the natural course of chronic infections. The data in this paper are notably different from data that have been previously presented with the SIVmac239 challenge model. It has typically been notoriously difficult to demonstrate that vaccine-elicited cellular immune responses can mitigate infection with this challenge virus. However, the reported observations could likely be influenced by the use of non-Indian macaques in which infection with SIVmac239 is attenuated compared with the usually used Indian macaques. Nonetheless, the data support a positive role for vaccine-elicited cellular immune responses in the mitigation of the SIV challenge infection.

Persons who acquire HIV-1 infection while enrolled in Phase I and II prime-boost AIDS vaccine trials with canarypox vectors (ALVAC)

Many researchers think that AIDS vaccine candidates that predominantly elicit cellular immune responses may not achieve sterilizing immunity but may modulate disease progression. So far, data on human subjects who became infected incidentally with HIV while participating in AIDS vaccine clinical trials were from trials of recombinant envelope subunit vaccines, not viral vector-based immunogens. Celum and colleagues reported on US National Institute of Allergy and Infectious Diseases (NIAID)-sponsored safety and immunogenicity trials of several canarypox vector (ALVAC) AIDS vaccine candidates constructed with clade B sequences (J. Infect. Dis. 190, 903; 2004). They looked into the natural history of early HIV infection among participants who received an ALVAC vaccine with or without a booster dose of recombinant gp120 or gp160, compared to that among placebo recipients.

A total of 1,497 study participants enrolled in nine Phase I and two Phase II canarypox AIDS vaccine prime-boost trials, of whom 1,257 were vaccinees and 240 were placebo recipients. Overall, 30 participants (2%) became HIV infected after enrollment, 18 of whom consented to enroll in breakthrough-infection protocols.

HIV seroincidence rates among canarypox HIV vaccine and placebo recipients were similar and comparable to those in the HIVNET Vaccine Preparedness Study (1.38 HIV infections/100 person-years). Vaccinees and placebo recipients who became infected with HIV were not different with respect to the proportion with symptomatic seroconversion and rate of disease progression, suggesting that the natural history of early HIV infection among vaccinees is similar to that of placebo recipients and historical cohorts.

The authors caution that reliable estimates of efficacy in reducing HIV infection rates cannot be derived given the small sample size of Phase I and II trials and the differences in immunogens and immunization schedules across protocols. They also point out that a series of large, randomized, controlled studies in diverse populations will be required to define the effect of preexisting HIV immunity on the long-term sequelae of HIV infection.

It is important to note that the authors of this article did not include any data on the vaccine-elicited immune responses in the volunteers, because the data could be potentially misinterpreted. Therefore, while the infection data are quite discouraging with regard to this particular test vaccine, the data do not address the antiviral potential of a vaccine-elicited cellular immune response.

Specific immune responses in primates vaccinated with adenoviral vector vaccines

Recombinant serotype 5 human adenoviruses (Ad) are promising vectors for delivering HIV antigens. Ertl and colleagues (J. Virol. 78, 7392; 2004) report on new chimpanzee E1 deletion Ad vectors (AdC6 and AdC7) constructed to express a truncated form of HIV-1 Gag and developed to circumvent preexisting immunity to Ad 5 in many adults. The vaccines were tested for induction of T-cell immunity responses in mice and nonhuman primates (NHP) in triple immunization protocols that included a heterologous human E1 and E3 deletion Ad 5 vector with the same truncated Gag. The E1 deletion reduces transcription of Ad antigens which results in loss of cell death upon infection and thus sustained antigen presentation, but also in an immune response that is more focused on the transgene product.

In previous studies, simian Ad vectors of serotypes C68 and C6 have induced potent transgene-specific CD8+ T-cell responses in mice and these responses could be boosted using heterologous vectors. Ertl and colleagues expanded these studies to two triple immunization protocols.

In mice, each booster immunization increased the frequencies of CD8+ T cells by two to threefold. The Gag-specific CD8+ T cells were stable, especially after the third immunization. In NHP, Gag-specific T-cell responses were readily detectable after the first booster immunization in all animals, although the frequencies of responses differed between the animals. They assumed this variability was in part due to the short Gag transgene used in this proof-of-principle study, encoding a limited number of T-cell epitopes. The longevity of the T-cell responses in NHP also differed but was remarkably stable over the four month period between the first and the second boosters.

The second booster immunization in NHP transiently increased gamma interferon-producing CD8+ T cells but overall the yield was not as impressive as the results observed in mice. The frequency of IL-2-producing T cells (presumably CD4+ cells) was significant in most animals after three immunizations. IL-2-secreting cells are viewed as critical to the success of AIDS vaccines.

The authors conclude that, within the limitation of Gag-encoding Ad vaccines for which preclinical challenge models are not available, the results of these studies are promising because frequencies of Gag-specific CD8+ T cells obtained with the triple immunization protocols were markedly higher in the high responder animals than those obtained with previous SIV-HIV protocols.

*Roberto Fernandez-Larsson, Ph.D., is the IAVI Report Web editor.