Search

An antibody response capable of neutralizing not only homologous but also heterologous forms of the CXCR4-tropic human immunodeficiency virus type 1 (HIV-1) MNp and CCR5-tropic primary isolate HIV-1 JR-CSF was achieved through sequential immunization with a combination of synthetic peptides representing HIV-1 Env V3 sequences from field and laboratory HIV-1 clade B isolates. In contrast, repeated immunization with a single V3 peptide generated antibodies that neutralized only type-specific laboratory-adapted homologous viruses. To determine whether the cross-neutralization response could be attributed to a cross-reactive antibody in the immunized animals, we isolated a monoclonal antibody, C25, which neutralized the heterologous primary viruses of HIV-1 clade B. Furthermore, we generated a humanized monoclonal antibody, KD-247, by transferring the genes of the complementary determining region of C25 into genes of the human V region of the antibody. KD-247 bound with high affinity to the "PGR" motif within the HIV-1 Env V3 tip region, and, among the established reference antibodies, it most effectively neutralized primary HIV-1 field isolates possessing the matching neutralization sequence motif, suggesting its promise for clinical applications involving passive immunizations. These results demonstrate that sequential immunization with B-cell epitope peptides may contribute to a humoral immune-based HIV vaccine strategy. Indeed, they help lay the groundwork for the development of HIV-1 vaccine strategies that use sequential immunization with biologically relevant peptides to overcome difficulties associated with otherwise poorly immunogenic epitopes.

By immunizing mice sequentially with six different V3 peptides we obtained a murine monoclonal antibody (MAb) C25, and its humanized counterpart KD-247. The MAb recognizes the sequence IGPGRA at the tip of the V3 loop and displays broad neutralizing activity against a variety of HIV-1 isolates. KD-247 was tested in an ex vivo neutralization assay to determine its capability to contain the spread of a quasi species population of clade B HIV-1 derived from two patients. The epitope of KD-247 was generally matching with the V3 sequences of various clones isolated from plasma and peripheral blood mononuclear cells (PBMC) of two patients. Complete or strong inhibition of viral replication was observed when the patients' PBMC were cultured with a high concentration of KD-247. Some neutralization escape variants, which had mutations in the V3 or outside of the V3 loop, emerged only at a low concentration of the MAb. These results suggest that KD-247 could be a good candidate for immunotherapy against HIV-1 in vivo.

By using the hepatitis B core (HBc) protein gene as a carrier, HIV-1 env V3 gene was inserted into the carrier gene, and the HIV gene was expressed inside a chimeric HIV-HBc particle (HIV-HBc), which was a unique candidate for induction of HIV-specific CTL activity. This was seen significantly in mice without the need of an adjuvant, because other responses specific for the HIV peptide such as T-cell proliferation and antibody production were not induced. However, when hemagglutinating virus of Japan (HVJ) protein was incorporated into an anionic liposome containing HIV peptide (HIV-HVJ-liposome) and was used as a booster immunization in HIV-HBc primed animals, the HIV-specific T-cell response and enhanced CTL activity were clearly induced in consecutively immunized animals. Furthermore, the HIV-specific humoral immune response was also induced and a neutralization activity was detected in the immune sera. Thus, when an HIV peptide antigen is expressed inside the virus like a particle of HBc, it can induce both cellular and humoral immunities when an HVJ-HIV-liposome, but not an HIV-liposome, is inoculated as the booster antigen. The HVJ-stimulated splenocytes secreted IL-18 and IL-12 to synergistically enhance the secretion of IFN-gamma in vitro. These findings suggest that the HVJ protein is effective at inducing the HIV-specific immunities, if used as part of a booster antigen in the consecutive immunization regimen.

We previously described a Sendai virus (SeV)-based expression system for the recombinant gp120 of HIV-1 subtype B (rgp120-B), which has permitted the production of antigenetically and functionally authentic gp120 at a concentration as high as 6 microg/ml of culture supernatant (Yu D et al.: Genes Cells 1997;2:457-466). Here the same procedure was successfully applied to the production of HIV-1 subtype E gp120 (rgp120-E). The remarkable production of the proteins by the SeV expression system enabled us to use crude culture supernatants for serological and functional studies of gp120s. The immunological authenticity of rgp120-E was verified by patient sera and anti-V3 loop monoclonal antibodies specific for HIV-1 subtypes B and E. CD4-binding properties were corroborated by FACS analyses. The rgp120s were then used in an enzyme immunoassay (rgp120-EIA) to detect antibodies in the sera of HIV-1-infected individuals, and the performance was assessed in comparison with a conventional V3 loop peptide EIA (V3-EIA). The initial evaluation of a serum panel (n = 164) consisting of 76 subtype E and 88 subtype B sera revealed that the rgp120-EIA was nearly 1000-fold more sensitive than the V3-EIA and was able to detect subtype-specific antibody with 100% sensitivity and with a complete correlation with the genotypes, whereas the V3-EIA failed to detect 9 and 24% of the same subtype E and B sera, respectively. Furthermore, a study employing a panel of 28 international sera with known genotypes (HIV-1 subtypes A through F) confirmed the remarkable specificity of this method. An EIA reactivity higher than 1.0 was an unambiguous predictor of HIV-1 subtype E and B infections. The data imply the presence of strong subtype-specific epitopes for antibody bindings to these rgp120s.

The sequences of the V3 loop and surrounding regions of human immunodeficiency virus type-1 from a father-to-mother-to-infant trimmer were studied and the horizontal and vertical transmissions compared. The father's virus was variable for reactivity with neutralizing antibody and sequences of the V3 loop central core sequence. In contrast, the mother's viral sequences were much less diverse and reacted with a virus neutralizing antibody. The infant's viral sequences were also less diverse than those of the father, and N-glycosylation sites were conserved. By phylogenetic analysis, the major clone, of which V3-peptide reacted with the neutralizing antibody, was found to be transmitted from the mother to her infant; however, the mutated minor clones did not bind to the antibody. These findings suggest that both horizontal and vertical virus transmission were selective, and that the clonally transmitted virus in infants mutates more rapidly than viruses in the mother, to whom the virus was horizontally transmitted.

A chimeric mouse-human antibody (C beta 1) was constructed that recognized the principal neutralizing domain (PND) of human immunodeficiency virus type 1 (HIV-1) gp120. The constant (C) immunoglobulin regions (C gamma 1 and C kappa) of a mouse monoclonal antibody, 0.5 beta, were substituted for the human C gamma 1 and C kappa by recombining the DNA modules encoding variable or C regions. The DNA constructs were then transfected into X63 Ag8.653 myeloma cells. A clone with a high production of the chimeric antibody (C beta 1) was selected. This antibody was tested for its biological activity against HIV-1. It bound to the surface of HTLV-IIIB-infected cells and reacted with gp120/160 with equal affinity and specificity to that of the parental 0.5 beta murine monoclonal antibody in a Western blot assay. Neutralization and/or enhancement of HIV infection were evaluated with C beta 1 and 0.5 beta. Both C beta 1 and 0.5 beta neutralized cell-to-cell infection and cell-free virus infection by HTLV-IIIB. Antibody-dependence enhancement of HIV infection was not observed with either C beta 1 or 0.5 beta in the presence or absence of human complement. Antibody-dependent cell-mediated cytolysis (ADCC) and antibody-dependent complement-mediated cytolysis (ACC) were observed with C beta 1 but not with the parental 0.5 beta. These findings suggest that the neutralizing antibodies to PND may neutralize but not enhance HIV infection. Furthermore, the high levels of ACC and ADCC shown against HIV-infected cells by C beta 1 indicate that the clinical application of such monoclonal antibodies may be possible.

The acquired immunodeficiency syndrome (AIDS) is the late-stage clinical manifestation of long-term persistent infection with the human immunodeficiency virus type 1 (HIV-1). Immune responses directed against the virus and against virus-infected cells during the persistent infection fail to mediate resolution of the infection. As a result, a successful AIDS vaccine must elicit an immune state that will prevent the establishment of the persistent infection following introduction of the virus into the host. The third hypervariable (V3) domain of the HIV-1 gp120 envelope glycoprotein is a disulphide-linked closed loop of about 30 amino acids which binds and elicits anti-HIV-1 type-specific virus-neutralizing antibodies. The in vitro characteristics of anti-V3 domain antibody suggest that this antibody could by itself prevent HIV-1 infection in vivo, an idea supported by chimpanzee challenge studies in which protection against the HIV-1 persistent infection seemed to correlate with the presence of anti-V3 domain antibody. Here we directly demonstrate the protective efficacy of anti-V3 domain antibody in vivo and propose that this antibody is potentially useful as both a pre- and post-exposure prophylactic agent.

The principal neutralizing determinants (PNDs) of 29 human immunodeficiency virus type 1 (HIV-1) isolates in Japan were analyzed using polymerase chain reactions. The viruses were isolated from 16 hemophiliacs, 11 individuals infected by their sexual transmission and 1 patient infected by blood transfusion (total 28 patients). Two virus isolates which were obtained from the same individual at different periods were also analyzed. All individuals were Japanese except one. The results produced 32 different PND sequences. A highly conserved central core sequence (GPG) was present in 27 of 32 patients, similar to the number reported in the United States, despite the marked heterogeneity in flanking regions of PNDs. The PNDs of all the 16 HIV-1 isolates obtained from patients with coagulation disorders had GPG sequences. Secondary structure prediction of PNDs by a joint method suggested that they were composed of coil-beta strand-coil-beta strand-alpha helix. It is suggested that the conserved core sequence has a type I turn. These findings may be useful in planning further clinical trials for passive vaccination.