IAVI Report - February / March 2001
Richard Jefferys

Over 3,000 people attended the 8th Conference on Retroviruses and Opportunistic Infections in Chicago (4-8 February 2001), the largest annual HIV science meeting. While virology and antiretroviral therapy generally dominated the agenda at past conferences, this year's event continued the recent trend towards a more prominent role for immunology and vaccines.

Many presentations in Chicago added to the growing body of data indicating that broad, vaccine-induced cellular immune responses to SIV/SHIV can at least partially protect monkeys against pathogenic challenge viruses. The overall sense was therefore one of increasing conviction that vaccines which can control viral replication and prevent or delay disease are possible. 

Yet, although several researchers described ongoing work to develop vaccines analogous to these SHIVs and SIVs in humans, few new types of candidates have entered Phase I clinical testing yet, pointing to the still-significant gap between primate promise and human trials. 

Protection and Memory with a DNA/MVA Vaccine
One highlight of the meeting was Harriet Robinson's presentation on a prime--boost approach that induced impressive protection. Carried out at the Yerkes Regional Primate Research Center/Emory University in Atlanta, the work has since been published (Science 2001 Apr 6;292(5514):69-74). The research was published a month after the conference and made a significant media splash.

A key feature of the study was the long (seven month) lag between immunization and challenge, a contrast to the usual strategy of challenging the animals shortly after the last vaccine boost, when immune responses peak. Robin-son's study design ensured that vaccine-induced T-cell responses had matured into immunological memory, and therefore tested the critical question of whether resting T-cells can mediate protection against HIV. It also used a mucosal (rather than intravenous) challenge, mimicking the main route of HIV transmission in humans, and a two-component vaccine encoding multiple SIV/HIV proteins that induced broad immune responses.

The components are a DNA plasmid carrying eight viral genes (SIV gag, pol, vif, vpx and vpr and HIV-1 env, tat and rev) and a recombinant MVA (modified vaccinia virus Ankara) vector with HIV-1 env and SIV gag and pol, made by Bernard Moss at the US National Institutes for Allergy and Infectious Diseases (NIAID; Bethesda). Rhesus macaques were divided into four groups of six animals each and received the DNA vaccine at weeks 0 and 8, either intradermally (i.d.) or intramuscularly (i.m.), followed by a combined i.d./i.m. boost with MVA at week 24. Four animals (two mock-immunized and two naive) served as controls.

The vaccine elicited both CD4 and CD8 T-cell responses, detected during the peak response (one week after the rMVA boost) as well as in the memory phase (six months later). These responses were measured with ELISPOT assays for Gag-specific interferon-gamma production, carried out with the Centers for Disease Control (Atlanta). As expected, the post-boost peak in Gag-specific T-cells was followed by a 5 to 20-fold drop in the memory phase. 

Seven months after the boost, animals were given a highly pathogenic intrarectal (i.r.) challenge (SHIV 89.6P) and then monitored closely. After a transient spike of viremia, the geometric mean viral load levels in all vaccinated groups declined to 1,000 copies or less, while CD4 counts were preserved. Thereafter, only one vaccinated macaque (from the low-dose i.m. group) showed intermittent viral load increases to >100,000 copies and a drop in CD4 counts. Clinically, all immunized animals remained completely healthy. In contrast, four non-vaccinated controls showed the expected high virus levels (averaging two million copies eight weeks after challenge) and a rapid crash in CD4 counts; all died within 32 weeks.

Analyzing the data for possible immune correlates of protection, Robinson reported a highly significant (p=<0.001) inverse association between peak pre-challenge SIV-specific T-cell responses (as measured by ELISPOT) and viral load five weeks post-challenge. This correlation echoes data reported by Norman Letvin and colleagues, who demonstrated an inverse relationship between vaccine-induced CD8 CTL responses (measured by tetramer staining) and post-challenge viral load (Science 2000 Oct 20;290(5491):486-92). However, Robinson may be the first to correlate both CD4 and CD8 T-cell responses with post-challenge virologic outcome. 

Plans call for moving this vaccine approach into human trials through the HIV Vaccine Trials Network (HVTN) in early 2002. The test vaccine construct is based on HIV-1 subtype B and will include the gag, pol, env, tat and rev genes in the DNA prime and gag, pol and env in the rMVA boost. In a parallel project, Robinson is working with NIAID and CDC on extending the approach to HIV-1 subtype A vaccines.

VSV-based Vaccines in Monkeys
Over the past several years, John Rose and colleagues from Yale University (New Haven) have been developing a vaccine vector based on attenuated vesicular stomatitis virus (VSV; see J Virol 2000 Dec;74(23):10903-10). VSV can be pathogenic in farm animals but rarely causes disease in humans. Exposure to VSV is unusual in most human populations, although Rose mentioned that antibodies to VSV are found in up to 94% of residents in some tropical areas of the Americas where VSV infection of cattle is common. At the conference, Rose presented results of the first primate SIV challenge experiments, done with Preston Marx of the Tulane Regional Primate Research Center and Aaron Diamond AIDS Research Center (NY). 

Animals received three immunizations at two month intervals, each time using a different variant of the VSV vector (to circumvent the neutralizing antibodies that develop to the viral envelope after each vaccination round). The three variants, called "glycoprotein exchange vectors," encode different VSV envelopes but otherwise carry identical vector sequences, along with the HIV env (89.6) and SIV gag genes. Either two or six months after the last boost, five vaccinated macaques and six controls (given vector without HIV/SIV genes) were challenged i.v. with SHIV89.6P. 

In vaccinated animals, CD4 counts initially declined but showed a trend back towards normal levels within four months, and all of them remained clinically healthy. (Viral load data was not yet complete). Control monkeys showed high viral loads and a rapid loss of peripheral CD4 T-cells, although two animals gradually recovered CD4 counts and partially controlled their infections. Detailed analyses of SIV-specific T-cell responses are pending. Doug Nixon (Gladstone Institute, San Francisco) conducted preliminary ELISPOT assays on two vaccinees and found Env- and Gag-specific responses that increased after each boost.  

Combining these results with an earlier pilot study of two immunized animals (on a different schedule) and two control macaques, Rose concluded that 7/7 vaccinated macaques remained clinically healthy, showed relative preservation of CD4 counts, and maintained low or undetectable viral loads. In contrast, 4/8 control macaques developed symptoms of simian AIDS by 140 days post-challenge; 2/8 have high viral loads, low CD4 counts and are expected to develop symptomatic disease. The remaining 2/8 controls are slowly recovering CD4 cells and controlling viral load. Wyeth Ayerst is planning to work with Rose's group on the development of VSV vectors for HIV vaccines.  

Challenge Studies with VEE-based vaccines
Bob Johnston (University of North Carolina) presented new data from challenge studies on monkeys immunized with another viral vector-based vaccine, this one made with attenuated Venezuelan equine encephalitis virus (VEE; see IAVI Report, Nov-Dec 1999). The study included six animals vaccinated with a mixture of vector particles (encoding SIV gp160, gp140 or matrix/capsid proteins) at 0, 1, and 4 months, as well as six mock-vaccinated animals. One month after the last immunization, all animals were challenged i.r. with the pathogenic virus SIVE660. 

Data at 23 weeks post-challenge show evidence of vaccine-mediated protection. Four of the six immunized macaques vigorously controlled SIV viremia (two below the 400 copy cut-off for the assay and two <2,000 copies), compared to just one mock-immunized animal in this low range. (The remaining five controls and two vaccinees had loads well over 100,000 copies at this timepoint). On average, CD4 counts were increased to 136% of baseline in vaccine recipients, while controls showed a decline to 40% of baseline. All animals were euthanized at 41 weeks and complete data analysis is pending. The VEE-based vaccine is being developed for human clinical trials by an IAVI-sponsored partnership with South African scientists and AlphaVax (Durham), with trials planned through the HVTN's vaccine trial site in Durban. The trial will test vaccines containing gag, env and pol genes from a South African clade C HIV isolate.

Immune Control in  PEP-Treated Monkeys
Jeff Lifson (National Cancer Institute, Frederick, MD) gave an update on his studies of post-exposure antiretroviral therapy in macaques. Previous work has shown that the majority of monkeys treated with a short course of the drug PMPA within 24 hours of exposure to pathogenic SIVE660 do not seroconvert, and they exhibit sustained control of viremia after PMPA is withdrawn (see J Virol 2000 Mar;74(6):2584-93). These animals also show enhanced SIV-specific proliferative responses compared to untreated or suboptimally treated controls. Of three macaques rechallenged with SIVE660 six weeks after PMPA termination, two seroconverted but then controlled virus to undetectable levels, while a third showed a transient increase in SIV-specific T-cell proliferation but no viremia or antibody response. 

Lifson's group has now extended these studies by challenging the macaques with a heterologous pathogenic strain (SIVmac239). He reported that the animals all became infected with the challenge virus but strongly controlled viral replication. Depletion of CD8 cells by monoclonal antibody (mAb) led to a rapid increase in viral loads (up to 5 logs), with some animals showing replication of both the original E660 and the challenge mac239 viruses. As the effects of the mAb wore off (and CD8 cells returned), control of viremia was regained. This restriction of viral replication occurred without measurable neutralizing antibody and at cellular response levels (measured by lymphoproliferation and ELISPOTs for CD8 T-cell derived interferon-gamma production) that Lifson describes as "unremarkable." These encouraging findings suggest that, with a sufficient head start (potentially through vaccination), the immune system may be able to mediate prolonged containment of SIV and, by extension, HIV.

Antibody Booster
Ron Desrosiers (Harvard Medical School, Cambridge) reviewed his team's work on attenuated SIVs created by mutating the V1/V2 region of the envelope gene (see IAVI Report, Sept.-Nov. 2000). The attenuated variants cause only transient viremia and no apparent disease in infected macaques, and they are highly sensitive to neutralization by antibodies. Upon challenge with a pathogenic strain, these animals appeared completely protected from secondary infection.  

Desrosiers believes that neutralizing antibodies are largely responsible for the attenuated course of infection with these mutants. In support of this view, he said that depletion of B-cells (which produce antibodies) in six animals led to increased replication of the attenuated virus. Desrosiers is now collaborating with Larry Arthur and Jeff Lifson at the National Cancer Institute (Frederick, MD) on whole-killed derivatives of these viruses for testing as potential vaccines.

Towards a "World Clade" Vaccine
While vaccine designers usually focus on a specific HIV subtype and/or population, a few researchers are taking the opposite approach: designing vaccines to work as broadly as possible across the many different HIV subtypes and strains. Anne De Groot (Brown University and EpiVax, Providence) began by developing two bioinformatics tools: Conservatrix, which identifies potential CTL epitopes that are highly conserved across clades; and Epimatrix (available at http://tbhiv.biomed.brown.edu), a predictor of which epitopes are most likely to be immunogenic and to bind the common HLA molecules (the highly diverse molecules that help display epitopes to T-cells). De Groot then tests peptides selected in this way against CTLs from healthy, HIV-infected individuals. Of more than 100 peptides tested so far, 71% have turned out to bind the predicted MHC molecule. Her team is now analyzing DNA plasmids carrying the chosen epitopes by transfecting them into dendritic cells and testing whether they stimulate CTLs from HLA-matched, HIV-infected people.  

Identifying Epitopes 
Karina Yusim from Bette Korber's group (Los Alamos National Laboratory, CA) described a finding that might assist in CTL epitope prediction. After observing that CTL epitopes tend to cluster in certain regions of HIV proteins, she began looking at where HIV proteins- which must be chopped into small peptides for presentation to CTLs- are most likely to be cleaved. Drawing on hundreds of CTL epitopes from the Los Alamos HIV Genetic Sequence and Immunology Databases (hiv-web.lanl.gov), Yusim and coworkers found a clear correlation between the boundaries of published CTL epitopes and predicted protein cleavage sites in Nef and p17. Conversely, in variable regions of the virus with few known epitopes, Yusim found that protein sequences are cleaved differently, or not at all. This suggests that strategies for identifying useful CTL epitopes should consider where viral proteins are likely to be cleaved, and conversely, that HIV regions resistant to cleavage and/or MHC binding might be omitted from vaccine constructs without reducing immunogenicity.  

Exposed Seronegatives
Three posters featured data on seronegative individuals with documented exposures to HIV (ESNs). Ann Duerr (CDC, Atlanta) and colleagues looked for evidence of "resistance factors" in a cohort of ESN women married to HIV-infected men from Chang Mai (Thailand), by comparing them to wives who became infected. Factors that did not differ between the ESN and infected women included HLA class I alleles, CCR5 delta 32 genotype and the presence of SLPI (an HIV-inhibiting protein) in the cervico-vaginal lavage. 

Looking at samples from the husbands, investigators found no differences between the groups in seminal viral load, defects in HIV accessory genes or co-receptor use by the viral isolates. More than half the ESNs and all the infected women showed HIV-specific T-cell responses (measured by ELISPOT). The proportion of women with T-cells specific for Gag, Nef and Pol was similar in both groups, but only 18% of ESNs had detectable responses to Env, compared to 83% of the infected women

Two collaborations involving Sarah Rowland-Jones' group (Institute of Molecular Medicine, Oxford) studied a cohort of eight serodiscordant couples (with one HIV-infected and one uninfected partner). One poster was presented by Ruth Braganza (Imperial College School of Medicine at St. Mary's, London), the second by Susana Pinheiro from the Oxford team. All eight ESN partners had CD8 T-cell responses to HIV. The researchers sometimes saw responses to whole HIV proteins but not to known epitopes from that same protein, implying the existence of still-unidentified CTL epitopes. In one person, an expansion of CD8 T-cell responses was detected after a documented exposure (due to condom breakage) that did not lead to overt HIV infection. In five ESNs analyzed for CD4 T-cell responses, three showed HIV-specific activity directed against gag p24 and p16 in ELISPOT assays. 

Two years ago at the Retrovirus conference, Zhu and coworkers (Fred Hutchinson Cancer Research Center, Seattle) presented evidence of extremely low-level, cryptic HIV infections (0.1- 0.01 copies of DNA per million T-cells) in a cohort of ESN individuals. To search further for evidence of such low-level infection, Zhu has now studied macaques selected from previous vaccine challenge studies. None of the monkeys displayed detectable SIV on standard tests, but they all showed evidence of low levels of SIV DNA (average of 4.5 copies per million peripheral blood mononuclear cells). Zhu's work raises the question of whether other ESNs harbor latent HIV infections that are undetectable using standard methods.

CTL Responses to HIV  Regulatory Proteins
Bruce Walker's team at Harvard University (Cambridge) presented two posters on the epitope specificities of HIV-specific CD8 T-cells. The first study, led by Marylyn Addo, examined whether infected individuals show CTL responses to the HIV regulatory proteins Tat and Rev (this study was subsequently published in PNAS 98: 1781, 2001). Although these two proteins are small, Addo detected multiple CTL epitopes in each by performing ELISPOT assays with overlapping peptides spanning their entirety. Of 57 people studied, 19% showed CTL responses to Tat and 37% to Rev. The second study, led by Marcus Altfeld, used similar techniques to detect responses to two other regulatory proteins, Vif and Vpr. In 50 infected individuals, 33% demonstrated Vif-specific and 46% Vpr-specific CTL activity. Both posters stressed that these regulatory proteins are expressed early in the viral life cycle and may therefore be useful vaccine antigens. 

Visualizing Immune Activation 
David Schwartz (Johns Hopkins, Baltimore) presented a novel poster on using positron emission tomography (PET) scans to visualize lymphocyte activation, both in HIV-infected people and in vaccinees. Study participants were infused with a radiolabeled glucose analog ([18]-FDG) and then PET-scanned. This technique has been used to identify solid lymphomas and lymphocyte activation in SIV-infected macaques (see Scharko et al, Proc Natl Acad Sci U S A 1996 Jun 25;93(13):6425-30). In people infected with HIV, the degree of lymph node activation detected by PET was correlated with viral load. This correlation was seen both in recently infected individuals and in those infected for more than 10 years. Schwartz also applied the technique to seven uninfected recipients of a canarypox vaccine (ALVAC vCP205 or vCP1452). Interestingly, the only two individuals to show evidence of activation (5 days after the fourth immunization) were also the only ones with detectable CTL responses to the vaccine.

010310
IAVI2001-0304

©2001. The IAVI Report.

AEGiS is made possible through unrestricted grants from Boehringer Ingelheim, iMetrikus, Inc., John M. Lloyd Foundation, the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 2001. This material is designed to support, not replace, the relationship that exists between you and your doctor.

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1990, 2001. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content.