The infection of macaques by simian immunodeficiency virus (SIV) results in many of the characteristics of HIV infection and progression to AIDS in humans, including loss of CD4+ T cells and susceptibility to opportunistic infections. Our studies have used this SIV macaque model to focus on various vaccine approaches. 1. Role of cellular antigens in AIDS vaccines. Macaques immunized with uninfected human cells have been shown to be protected from challenge with SIV propagated in human cells. In collaboration with Drs. L. Arthur and L. Henderson, we have shown that various cellular antigens (including beta-2 microglobulin (beta-2M), HLA class I and HLA class II DR) are specifically associated with purified preparations of HIV and SIV. To identify the potential antigens involved in protection, macaques were immunized with beta-2M or with HLA class I or class II DR purified from H9 cells. Although all macaques developed humoral immune responses to these proteins, only the macaques immunized with HLA DR were protected from a subsequent challenge with HIV grown in H9 cells. This was the first demonstration that immunization with a cellular protein can protect from an AIDS virus infection. A second challenge, this time with SIV grown in macaque cells, led to infection of all animals, presumably because of the presence of a different HLA molecule on the surface of the virus used for the second challenge. The study described above represents xenoimmunizations?the use of human HLA molecules to immunize macaques. These studies have now been extended to alloimmunization studies - a study of the role of macaque class II antigens in protecting from AIDS virus infection (alloimmunization studies). In collaboration with Dr. M. Carrington, we have identified class II alleles in the pigtailed macaque colony at the University of Washington, and obtained class II DR from microvesicles harvested from Herpesvirus papio transformed macaque B-lymphocytes. Macaques that share similar and distinct class II alleles (as determined by oligotyping) were immunized with microvesicles containing DR, and then recently challenged with SIV/Mne grown in lymphocytes from the animal from which the class II antigen was purified. Animals are being monitored for evidence of viral infection, plasma loads of viral RNA and for the development of anibodies to SIV. These studies will further address whether antibodies to class II correlate with protection from infection in an alloimmunization regimen. 2. A DNA vaccine producing whole non-infectious SIV particles: Immunization and challenge. Full-length SIVMne proviral DNA with a 12-bp deletion in the Gag nucleocapsid (NC) coding domain was cloned into a pBR322 DNA plasmid vector (SIV ori/neo-M14). The 12-bp deletion removes Cys-Trp-Lys-Cys, which contains two of the four zinc ligand binding residues (Cys thiols) required for metal ion coordination and formation of the second zinc finger of the native NC protein. Cells transfected with SIV ori/neo-M14 produce normal levels of Gag, Gag-Pol, Env and auxiliary proteins, and yield budding virus-like particles that are deficient in genomic RNA. Mutant particles budding from transfected cells show immature and mature virus morphology by electron microscopy analysis. The particles have well-defined surface structures (spikes), strongly suggesting a native-like gp120 and gp41 orientation. The mutant particles can bind to and enter CD4 cells but are non-infectious and are incapable of establishing a proviral state. Thus, the characterization of these non-infectious mutant particles suggests that they mimic live virus in their assembly, budding, maturation, attachment, and entry stages of normal virus replication and therefore, if produced in vivo, could serve as immunogens for the presentation of native structures. To test this hypothesis, five macaques were inoculated intramuscularly with SIV ori/neo-M14 DNA. Anti-SIV antibodies were detected in two of the immunized animals, showing that the DNA had expressed viral proteins. Four control animals received vector plasmid DNA lacking the SIV construct. All animals were challenged with 10 macaque-infectious units of SIVmne. The data have now been analyzed through 106 weeks after infection. All four control animals became persistently infected: they had high plasma RNA viral levels and virus could be readily isolated from peripheral blood mononuclear cells (PBMC). Three of the four control animals show markedly declining CD4 cells, and three have died at 60 to 102 weeks after challenge. Four of the five DNA-vaccinated animals had plasma virus levels that were either undetectable or 2 to 3 logs less than control animals. These four animals continue to show normal CD4 cell counts and are healthy 2 years after challenge. The results of this preliminary study suggest that DNA vaccines may be an effective and safe route of immunization leading to protection from an AIDS virus challenge. 3. Recombinant SIV vaccines. We previously reported that immunization with recombinant SIVmne envelope (gp160) vaccines protected macaques against intravenous challenge by the cloned homologous virus, E11S, but protection was only partially effective against an uncloned stock of SIVmne. In the present study, we examined the protective efficacy of this immunization regimen against infection by a mucosal route. We found that gp160-based vaccines were highly effective against intrarectal infection not only by the E11S clone, but also against the uncloned SIVmne. Protection against mucosal infection is therefore achievable by parenteral immunizations. Protection appears to correlate with the presence of serum-neutralizing activities and high levels of SIV-specific antibodies. To understand the basis for the differential efficacy against the uncloned virus by the intravenous vs. the intrarectal routes, we examined viral sequences recovered from the PBMC of animals early after infection by both routes. We previously showed that the majority (85%) of the uncloned SIVmne challenge stock contained V1 sequences homologous to the molecular clone from which the vaccines were made (E11S type), with the remainder (15%) containing multiple conserved changes (the variant types). In contrast to intravenously infected animals, from which either E11S-type or the variant V1 sequences could be recovered in significant proportions, animals infected intrarectally had predominantly E11S-type sequences. Preferential transmission or amplification of the E11S-type viruses may account in part for the enhanced efficacy of the recombinant gp160 vaccines against the uncloned virus challenge by the intrarectal route as compared to the intravenous one. 4. Role of cell-mediated immunity in AIDS vaccines. A possible role for cell-mediated immunity (CMI) in protection against HIV infection is suggested by the finding that a significant number of potentially HIV-exposed individuals from different risk groups, who show no evidence of infection, exhibit a strong HIV-specific CMI. To test this hypothesis in the SIV-macaque model, we inoculated macaques with graded doses of SIV and examined these animals for SIV viral sequences and the development of immune responses. Macaques inoculated with doses of SIV below the threshold required for virus isolation or seroconversion exhibited lymphocyte proliferation and were protected from an infectious SIV challenge delivered 16 months later. Additional studies are planned that will compare both inactivated and live virus delivered at a subinfectious dose.
