Search

Limited success of current HIV-1 vaccines warrants new approaches. We discuss feasibility and potential benefits of early life HIV-1 immunization followed by vaccine boosts during childhood that may enable maturation of vaccine-induced broad anti-HIV-1 immunity over several years. By initiating this immunization approach in the very young, well before sexual debut, such a strategy may dramatically reduce the risk of HIV-1 infection.

Due to the different mechanisms HIV-1 has evolved to escape from a neutralizing antibody response it has been extremely challenging to develop an effective anti-HIV-1 vaccine. The V3 region of the gp120 HIV-1 envelope glycoprotein has been considered as one of the possible targets for an anti-HIV vaccine. It is well known that the V3 region of gp120 is at least partially masked in circulating strains and becomes exposed only after CD4 binding. However, when the virus is bound to surface CD4, steric hindrance prevents effective neutralization by V3-directed antibodies. Here we have used a 27-residue CD4-mimetic peptide in combination with immune sera elicited by an optimally constrained V3 peptide to enhance neutralization of a panel of clade B viruses. We observed strong synergism between the immune sera and the CD4-mimetic in the neutralization of tier 1 and a representative tier 2 clade B virus suggesting that the constrained V3 peptide immunogen correctly mimics the V3 conformation even in tier 2 clade B viruses. This synergy should improve the potential of CD4-mimetic compounds for preexposure prophylaxis and in the treatment of HIV-1-infected patients who usually manifest high titers of V3-directed antibodies. Moreover, constrained V3 immunogens elicit immune sera that may neutralize HIV in synergy with CD4 binding site antibodies that expose V3 and the coreceptor binding site.

Mucosal and systemic transmission of HIV is prevalent. Therefore, mucosal followed by parenteral immunizations with chimeric vs. complete alphavirus-based replicon particles, encoding an HIV envelope glycoprotein, were tested. Female rhesus macaques were immunized intranasally and then intramuscularly. Following the immunizations, enhanced mucosal and systemic antibody responses were detected with the chimeric compared to the complete replicon particles. Although similar proportions of the same peripheral blood monocyte lineage target cells were infected with the chimeric vs. the complete replicon particles, the latter resulted in enhanced expression of the gene of interest, suggesting a possible mechanism of the enhanced immunogenicity.

In the present study, macaques were coimmunized with VEErep/SINenv chimeric alphavirus replicon particles expressing SIVp55Gag and HIVDeltaV2gp140Env or only with replicon particles expressing HIVDeltaV2gp140Env. All animals were subsequently immunized with recombinant trimeric HIVDeltaV2gp140Env protein. During alphavirus immunization, anti-SIVGag and anti-HIVEnv-specific interferon (IFN)-gamma responses, as well as high titers of anti-HIVEnv binding (gp120 but not gp41 specific) and anti-HIV neutralizing antibodies, were generated. The subsequent immunization with recombinant HIVDeltaV2gp140 enhanced the neutralizing antibody titers and Env-specific IFN-gamma responses. Following intravenous challenge with the R5- tropic SHIV(SF162P4) virus, significantly lower primary plasma viremia levels were recorded in the immunized animals, as compared to control animals immunized with replicon particles expressing influenza virus HA. Our results show that this method of immunization elicits both strong cellular immunity and neutralizing antibodies in primates and, thus, merits further investigation.

HIV-specific antibodies become detectable and continue to increase in frequency during primary infection. The effects of early antiretroviral treatment (ART) and its discontinuation on the evolution of this immune response have not been systematically analyzed. To investigate the associations between antibody titer, viral load, and ART, we used a less-sensitive enzyme-linked immunosorbant assay (LS-EIA) to measure changes in HIV-1-specific antibody levels in treated and untreated subjects undergoing primary infection. In this longitudinal study, antibody levels gradually increased in therapy-naive subjects, reaching a plateau approximately 40 weeks postinfection. In contrast, antibody titers remained low among subjects receiving ART. Subjects who discontinued ART exhibited a more rapid rise in antibody titers than therapy-naive subjects, suggesting the presence of an enhanced B cell response. These results demonstrate that early ART prevents the typical evolution of the HIV-1-specific antibody response and can alter the expected kinetics of this response in subjects discontinuing therapy.

Considerable evidence suggests both HIV-specific T cells and neutralizing antibodies (nAb) can, separately, assist control of viremia. T cell and nAb responses were studied in detail in three pigtail macaques protected from chronic simian/human immunodeficiency virus (SHIV) viremia by DNA prime/fowlpoxvirus boost vaccine regimens. Immunity was studied both after an initial intrarectal SHIV challenge, as well as during CD8 T cell depletion and a subsequent intravenous SHIV rechallenge. Remarkably, SHIV-specific CD4 and CD8 T cells were detectable in the absence of viremia following an initial SHIV challenge in one animal, subsequent to recovery from CD8 T cell depletion in all three animals, and following control of heterologous SHIV rechallenge in two animals. Neutralizing antibodies were also enhanced following CD8 depletion without recrudescence of viremia in all three animals. These observations, although in a small subset of animals, suggest the hypothesis that combinations of primed T cell immunity and neutralizing antibodies can maintain control of chronic primate lentiviral infections.

Immunologic and biologic factors associated with the progression to AIDS in HIV-1-infected chimpanzees were investigated. Chimpanzee C499 was euthanized in 1996 as a result of the development of AIDS approximately 11 years after infection with HIV-1. At the time of initial disease development (September 1995), blood from this animal was transfused to an uninfected chimpanzee, C455, resulting in a rapid loss of CD4(+) T cells. Virus isolates were derived from both animals and termed HIV-1(JC) (derived from C499 at the time of disease development; JC isolate) and HIV-1(NC) (derived from C455 1 month posttransfusion; NC isolate). In vitro studies demonstrate that the parental viruses used to inoculate C499 were susceptible to neutralization by serum from that animal. In contrast, serum from C499 at any time was unable to neutralize the JC or NC isolates. Similarly, the JC and NC isolates were highly resistant to neutralization by serum from C455. However, serum from C455 was also unable to neutralize either of the parental viruses or any of the normally neutralization sensitive isolates tested. Serum samples from the two additional chimpanzees that were inoculated with the NC isolate were also unable to neutralize these isolates. Coreceptor usage of the uncloned JC and NC isolates was somewhat expanded when compared with that of LAV1b and SF2. However, molecular clones derived from the JC and NC isolates (JC16 and NC7) displayed only a limited coreceptor repertoire despite having unique V3 loop sequences. The results suggest that the JC and NC isolates are neutralization escape mutants and display a different phenotype than the parental strains LAV1b and SF2.

One anticapsid (p26) mouse monoclonal antibody was developed after immunization with recombinant p26 Gag protein and was tested for reactivity with different HIV-1 and HIV-2 isolates by ELISA and Western blot analysis. This antibody, named R26.1, reacted with all HIV-2 isolates tested and with recombinant p26 proteins, but no HIV-1 isolates. The epitope of antibody R26.1 was mapped to residues 50-71 in the N-terminal domain of the capsid protein, a highly conserved region in all HIV-2 isolates sequenced to date. This monoclonal antibody may be useful for the detection of HIV-2, and for the discrimination between HIV-1 and HIV-2 infections. Likewise, the identified epitope may be useful for the detection of p26 antibodies in HIV-2-infected individuals.

Because the immune response to HIV depends on viral gene expression, we examined the HIV-specific immune responses in persons whose viral load after highly active antiretroviral therapy (HAART) was <400 on at least 3 occasions over a 12-month interval. Eleven patients were identified. While there was little change in mean HIV-binding antibody (Ab) titers in this group, two persons mounted increases in HIV envelope-specific binding antibody. Neutralizing antibody (NAb) titers against a panel of HIV-1 primary isolates (BZ167, US1, and CM237) increased post-HAART (80% neutralization titer against US1, p = 0.06; against CM237, p = 0.04). The two persons with large increases in binding antibody also had increases in primary isolate NAb. Roughly half of HAART recipients had significant increases in neutralizing antibody to the primary isolates US1 and CM237. Compared with CD4-matched, non-HAART controls, there were significant increases in NAb against the subtype B primary isolate US1 (p < 0.0009); no increases were seen against more easily neutralized primary isolate BZ167. There were no differences after HAART in antibody-directed cellular cytotoxicity (ADCC). HAART resulted in a partial restoration of lymphoproliferative responses to recall antigens (tetanus and diphtheria). New responses developed to HIV Gag p24. No patient responded to HIV Env gp160 or gp120 either before or after HAART. The data underscore the lack of functional reconstitution of HIV-specific, CD4-mediated responses despite durable suppression of viral replication. In the setting of stable anti-HIV Ab levels, the development of increased NAb in certain individuals suggests that control of the virus by HAART may assist in immune control of HIV.

Antibodies generated by candidate HIV-1 vaccines in a phase I clinical trial were assessed for neutralizing activity with a panel of eight well-characterized, genetically diverse clade B primary isolates having an R5 phenotype. The vaccines consisted of one of three different recombinant canarypox vectors expressing membrane-anchored HIV-1(MN)gp120 (ALVAC vCP205, vCP1433, and vCP1452) followed by boosting with a soluble gp160 hybrid consisting of MNgp120 and the majority of gp41 from strain IIIB. Serum samples from a subset of volunteers in each arm of the trial, containing moderate to high titers of neutralizing antibodies to HIV-1 MN, were analyzed. Competition assays with peptides revealed that the majority of neutralizing activity was specific for the MN-V3 loop. Despite MN-specific neutralization titers that sometimes exceeded 1:500, no neutralization of primary isolates was detected and, in some cases, mild infection enhancement was observed. In addition, little or no neutralization of the HIV-1 IIIB heterologous T cell line-adapted strain of virus was detected. These results reinforce the notion that monovalent HIV-1 ENV is a poor immunogen for generating cross-reactive neutralizing antibodies.

Several immunogens induce HIV-specific neutralization and in vitro lymphoproliferation in adults at low HIV-1 risk, but responses in persons at high HIV-1 risk are not known. We performed a multicenter, double-blinded, adjuvant-controlled trial with two gp120 vaccines in 296 HIV-1-uninfected volunteers, including 176 reporting higher HIV-1 risk activities. The immunogens were remarkably well tolerated. After three immunizations, 210 of 241 vaccinees (87%) developed neutralizing antibodies, which persisted in 59% after 2 years. The injection drug users receiving SF-2/gp120 had decreased antibody responses relative to the lower risk groups. Envelope-specific lymphoproliferation peaked after two immunizations, and 54% of vaccinees mounted a DTH reaction to gp120 after 4 years. In summary, these immunogens have low adverse reactogenicity and induce durable antibody and T cell responses to the prototype strains. Unexpected differences in antibody responses among diverse HIV-1 risk strata lend support to the conduct of expanded phase II trials in populations other than low-risk volunteers.

HIV-specific antibodies and CD8+ T cell antiviral responses were evaluated in three human immunodeficiency virus 1 (HIV-1) gp120 vaccine recipients who later became infected with HIV-1. Titers of neutralizing antibody to the HIV-1(SF2) vaccine isolate were boosted, but titers of antibody to the autologous infecting viruses were never high and required at least 6 months after HIV infection to develop. Similarly, a marginal noncytotoxic CD8+ T cell antiviral response was observed only in one of the three vaccinees 3 months after HIV-1 infection. The infecting virus isolates had several amino acid substitutions in the HIV-1 envelope V3 region but were similar to other regional HIV-1 clade B isolates. Viral loads were similar to those of other HIV-1-infected individuals who had not been vaccinated and transient CD4+ T cell declines were observed in each person, suggesting that the vaccine was not effective at controlling these prognostic markers early in infection.

The use of synthetic peptide antigens in human prophylaxis still suffers from the very important problem of finding suitable carriers devoid of side effects. A desirable carrier for use in humans would be poorly immunogenic by itself, yet it would enhance the immune response to the peptide antigen. In the study reported herein, we examined the role of polytuftsin (TKPR40), a synthetic polymer of the natural immunomodulator tuftsin, as a carrier for synthetic peptides of HIV derived from the gp41 and gp120 proteins. Chimeric immunogens were constructed by chemical linkage between synthetic peptides of HIV and polytuftsin. These were employed for immunization of mice of different MHC haplotypes, and the humoral and cellular immune responses developed against the peptides were assessed by measuring total IgG, IgG, subclasses, T-cell proliferation, and in vitro cytokine release. A significantly stronger immune response was observed in mice immunized with the peptide-polytuftsin conjugates than in mice receiving the peptide dimers (peptide-peptide). Peptide-polytuftsin conjugates induced IgG2a and IgG2b isotype switching after both primary and secondary immunization. In addition, there was a positive correlation between the amounts of cytokines and the shift in the IgG isotypes. These data suggest that the use of polytuftsin as a carrier may increase the immune response against poorly immunogenic synthetic peptides.

We compared antibody responses to various structural and accessory gene products of HIV-1 between long-term nonprogressors and patients who have progressed to AIDS. On the basis of our results, we performed epitope mapping of the Nef protein and identified a novel epitope. This novel epitope of the Nef protein was found to be exposed on the surface of HIV-1-infected cells. The antibody response against it correlated with CD4+ cell counts among HIV-1-infected patients (r = 0.457, p < 0.001). Although further research is necessary, we suspect that antibody response against the epitope may be protective against disease progression.

A live recombinant canarypox vector expressing HIV-1 gpl20 MN tm/gag/protease LAI (ALVAC-HIV, vCP205) alone or boosted by a p24E-V3 MN synthetic peptide (CLTB-36) was tested in healthy volunteers at low risk for HIV infection for their safety and immunogenicity. Both antigens were well tolerated. ALVAC-HIV (vCP205) induced low levels of neutralizing antibodies against HIV-1 MN in 33% of the volunteers. None of them had detectable neutralizing antibodies against a nonsyncytium-inducing HIV-1 clade B primary isolate (Bx08). After the fourth injection of vCP205, CTL activity was detected in 33% of the volunteers and was directed against Env, Gag, and Pol. This activity was mediated by both CD4+ and CD8+ lymphocytes. On the other hand, the CLTB-36 peptide was poorly immunogenic and induced no neutralizing antibodies or CTLs. Although the ALVAC-HIV (vCP205) and CLTB-36 prime-boost regimen was not optimal, further studies with ALVAC-HIV (vCP205) are warranted because of its clear induction of a cellular immune response and utility as a priming agent for other subunit antigens such as envelope glycoproteins, pseudoparticles, or new peptides.

DNA or nucleic acid immunization has been shown to induce both antigen-specific cellular and humoral immune responses in vivo. Moreover, immune responses induced by DNA immunization can be enhanced and modulated by the use of molecular adjuvants. To further engineer the immune response in vivo, we investigated the induction and regulation of immune responses from the codelivery of Thl cytokines (interleukin-2 [IL-2] and IL-12), Th2 cytokines (IL-4 and IL-10), and granulocyte-macrophage colony-stimulating factor (GM-CSF) genes along with a DNA vaccine construct encoding for simian immunodeficiency virus (SIV) gag/pol proteins. We observed that coinjection with IL-2, IL-4, IL-10, and GM-CSF resulted in increased levels of antigen-specific antibodies. In addition, we found that coinjection with cytokine genes drove the immune responses toward a more Thl or Th2 phenotype. We also observed that coadministration of IL-2, IL-12, and GM-CSF genes resulted in a dramatic enhancement of Th proliferation responses. Moreover, coimmunization with IL-12 genes resulted in a dramatic enhancement of antigen-specific cytotoxic T lymphocyte (CTL) responses. These results support the potential utility of molecular adjuvants in DNA vaccine regimens.

The aim of our study was to investigate the possible correlation of in vitro antibody production (IVAP) directed to the gp160 protein of HIV-1 with CD4+ slopes, plasma viremia, and disease progression in long-term nonprogressors (LTNPs). Nineteen subjects with a long-term nonprogressive HIV-1 infection were studied and followed for 2 years. During the follow-up, in vitro anti-gp160 producers showed negative CD4+ slopes in the majority of cases (9 of 12), whereas 5 of 7 nonproducers showed positive CD4+ slopes. Plasma viremia values, which were not significantly different in the two groups at baseline, became significantly higher in anti-gp160 producers when compared with nonproducers during the follow-up (p = 0.012). Finally, a trend toward progression was observed in the group of producers but not in nonproducers. These findings suggest that the in vitro production of anti-gp160 antibodies by peripheral B cells is not a correlate of protection, and may represent an early predictor of progression in LTNPs.

AIDSVAX (VaxGen, Inc., South San Francisco, CA), a possible vaccine to protect against human immunodeficiency virus type 1 (HIV-1) infection, is being tested for efficacy in phase 3 studies. It has been tested for potential efficacy in chimpanzees, and tested for safety and immunogenicity in human clinical studies. Four candidate vaccines, each with a different envelope protein antigen or combination of antigens, have been produced in alum formulations. In both design and clinical testing, AIDSVAX has an excellent safety profile. Because these highly purified proteins were prepared using recombinant DNA technology, there is no possibility of these vaccines causing HIV infection. Having been administered to over 1200 people, the only side effects attributable to AIDSVAX have been local pain and inflammation at the injection site. After immunization, essentially all recipients developed a robust antibody response, including binding and neutralizing antibodies. The neutralizing antibodies peaked after a 12-month boost. Excellent memory is induced. Two phase 3 trials of two bivalent formulations will evaluate their efficacy. One trial will use a bivalent subtype B formulation. This trial in North America will involve 5000 men who have sex with men and heterosexual women at high risk. The other study will use a bivalent subtype B/subtype E formulation. This trial in Thailand and will involve 2500 intravenous drug users. Both studies will be randomized, double-blinded and placebo controlled. The volunteers will be followed for 3 years. The end points of the studies are infection, as defined by seroconversion to standard diagnostic tests, and viral load, as defined by commercial polymerase chain reaction (PCR) tests.

Live attenuated vaccines prepared from simian immunodeficiency virus (SIV) have provided the best protective immunity in challenge experiments. In animals vaccinated with attenuated SIV, immune responses may be elicited owing to endogenous expression of native SIV proteins and/or antigen presentation in the native replication site of virus. However, replication-competent viral vaccines raise safety concerns for clinical trials in humans. To ensure the safety and maintain the immunogenicity of a live, attenuated vaccine, we have developed a replication-defective HIV pseudotyped with vesicular stomatitis virus G protein (VSV-G). The polymerase gene of HIV was truncated to construct the replication-defective HIV. This pseudotyped HIV can infect many cell types, including human and simian cells, and undergoes only one round of replication. Furthermore, antibody immune response can be detected in mice immunized with VSV-G-pseudotyped replication-defective HIV.

The fine specificity of the anti-V3 antibody responses induced in chimpanzees immunized by various human immunodeficiency type 1 (HIV-1) candidate vaccines and challenged by heterologous strains of HIV-1 was analyzed by enzyme-linked immunosorbent assay (ELISA) and Pepscan epitope mapping. Two chimpanzees immunized with the recombinant canarypox virus ALVAC-HIV (vCP125) expressing gp160MN and boosted with purified gp160MN/LAI alone, then with both immunogens in combination, were not protected against challenge with HIV-1 SF2. Their sera mainly recognized one epitope of the V3 loop, located in the NH2-terminal half. By contrast, immunization of two other chimpanzees with purified gp160MN/LAI and boosting with a synthetic V3MN peptide elicited a strong anti-V3 antibody response with a broader specificity directed against multiple epitopes all along the V3 loop. These chimpanzees were protected against infection by HIV-1 SF2. However, when these two chimpanzees were challenged later with a HIV-1 clade E strain virus, they became infected. We failed to detect any reactivity with the peptide of the ectodomain of gp41 of sera harvested after immunization with the various immunogens or after challenge with HIV-1 SF2 or HIV-1 90CR402. These results demonstrated that anti-V3 antibodies with a restricted fine specificity were induced in chimpanzees immunized with gp160 purified or expressed by recombinant canarypox confirming our previous results obtained in three different species (human, guinea pig and, macaque). In contrast, a boost with the V3 peptide broadened antibody responses, suggesting that the mode of presentation of the V3 loop to the immune system strongly influences the epitope specificity of the resulting antibody response.

We have previously reported on the generation of specific functional immune responses after inoculation of animals with expression vectors encoding HIV-1 genes. This article provides the details of the first application of this new technology to induce immune responses against HIV-2. This virus is molecularly and serologically distinct from HIV-1 and is in fact more closely related to the simian immunodeficiency virus (SIV). Anti-HIV-2 and SIV antibodies were induced in mice of three different haplotypes following a single intramuscular inoculation with an HIV-2/ROD envelope glycoprotein expression vector (pcEnv-2). Boosting of animals with pcEnv-2 induced both anti-HIV-2 neutralizing antibodies and T cell-proliferative responses against HIV-2 and SIVmac proteins. We compared the humoral and cellular immune responses of mice injected with pcEnv-2 and then boosted with either the homologous DNA construct or a recombinant Env protein. Animals boosted with pcEnv-2 generated B and T cell immune responses as strong as those of mice boosted with recombinant gp140 protein in adjuvant. Finally, cellular immune responses were significantly increased with the coadministration of pcEnv-2 and a plasmid expressing interleukin 12. We therefore conclude that DNA plasmid inoculation induces cross-reactive anti-HIV-2 and anti-SIVmac immune responses in mice. This technology should be further investigated as a potential vaccine component for this human pathogen.

Early in HIV-1 infection, B cell responses to T cell-dependent antigens are impaired. In addition to the receptor-ligand pair CD40/CD40L, CD27/CD70 also appears to be involved in T cell-dependent B cell stimulation. We have shown that CD70+ B cells are the main producers of Ig when stimulated in a T cell-dependent manner, and that CD70 upregulation is dependent on interaction of CD40L on T cells with CD40 on B cells. We confirm here that B cells from HIV-infected individuals are impaired in T cell-dependent Ig production in vitro. This dysfunction could partly be restored by adding allogeneic T cells to the culture. In contrast, IgG production induced by CD40 MAb, IgM MAb, and IL-10 was in the normal range. In line with this, CD70 upregulation on B cells from HIV-infected individuals was impaired after stimulation in vitro by activated T cells but not after stimulation with CD40 MAb and IgM MAb. Furthermore, CD40L expression was decreased on CD4+ T cells after stimulation in vitro. Finally, CD70 expression on freshly isolated B cells from HIV-infected individuals was decreased, and low CD70 expression correlated with low IgG production after T cell-dependent stimulation. In conclusion, our data strongly suggest that impaired B cell responses to T cell-dependent Ag in HIV-1 infection are due to a defect in T cells.

Vaginal anti-HIV antibody responses may be beneficial, and possibly required, for vaccine-induced protection against HIV infection acquired through receptive vaginal intercourse. We have previously determined that intranasal immunization with a hybrid HIV peptide and cholera toxin induced vaginal anti-HIV IgA responses in BALB/c and C57BL/6 mice. To determine if vaginal, gastric, or rectal boosting would enhance the induction of vaginal anti-HIV IgA responses over those observed with intranasal immunization only, C57BL/6 mice were intranasally immunized with the hybrid HIV peptide T1SP10MN(A) and cholera toxin (days 0 and 14) and boosted via the vaginal, gastric, or rectal route (days 7 and 28). Four intranasal immunizations was superior to all other immunizations evaluated for the induction of plasma anti-peptide IgG, vaginal anti-peptide IgG and IgA, and peptide-specific delayed-type hypersensitivity. In addition, intranasal priming with gastric boosting was associated with greatly elevated total serum IgE concentrations whereas intranasal immunization only was associated with only a modest increase in total serum IgE. These results suggest that intranasal immunization is a viable route of immunization for the induction of systemic and mucosal anti-HIV immune responses.

The HIV env-encoded synthetic peptide T1-SP10MN(A) contains immunodominant epitopes of the C4/V3 regions of gp120. The mucosal immunogenicity of this peptide in various vaccine preparations was first tested in rabbits using chronically isolated Thiry-Vella (T-V) ileal loops. Intestinal and serum samples collected from rabbits immunized via T-V loops demonstrated secretory IgA (S-IgA) and IgG anti-T1-SP10MN(A), respectively, when assayed by ELISA. Intranasal delivery of the peptide supplemented with cholera toxin (CT) resulted in serum IgG and S-IgA anti-T1-SP10MN(A) in vaginal and nasal secretions. This study further demonstrates the utility of rabbits as a convenient animal model for HIV vaccine research and the relationship between nasal immunization and vaginal immunity.

The safety and the immunogenicity of a recombinant hybrid envelope glycoprotein MN/LAI (rgp160) followed by booster injections of a V3 (MN) linear peptide were evaluated in HIV-negative adults at low risk for HIV infection. Volunteers received either rgp160 in alum at 0, 1, and 6 months (group A), rgp160 at 0 and 1 months followed by V3 at 3 and 6 months formulated in alum (group B), or in Freund's incomplete adjuvant (FIA) (group C). Local and systemic reactions were mild to moderate and resolved within the first 72 hr after immunization. No significant biological changes (routine tests and autoantibodies) were observed. One month after the last injection in either group, neutralizing antibodies (NAs) against the HIV-1 MN isolate were detected in 4 of 5 (A), 7 of 10 (B), and 10 of 10 (C) subjects with significantly higher geometric mean titers in the latter group. Four of nine sera with the highest NA titers against MN weakly cross-neutralized the HIV-1 SF2 isolate; none had NA against the HIV-1 LAI strain or against a North American primary isolate. Specific lymphocyte T cell proliferation to rgp160 was detected in 92% of the subjects after the second injection of rgp160 and in 80% of them 12 months after the first injection. A weak and short-lived envelope-specific CD(4+)-mediated cytotoxic lymphocyte activity was detected at certain time points in few subjects. This prime-boost vaccine approach using rgp160 followed by a V3 peptide was safe in humans, and was able to elicit high levels of neutralizing antibodies and a strong and persistent T cell lymphoproliferative response to rgp160 and to V3. However, the neutralization response was restricted to the homologous HIV-1 strain and little env-specific cytotoxic activity was induced.

Genetically engineered, noninfectious HIV-1-like particles containing processed envelope glycoproteins represent potential candidate immunogens for a vaccine against HIV-1. However, since the gp120 glycoprotein is known to be rapidly lost from the surface of infected cells and purified virions as a result of its low-affinity interaction with gp41, shedding of this extracellular subunit could compromise the immunogenic potential of particle-based HIV-1 vaccine candidates. In this study, we demonstrate for the first time the feasibility of producing fully assembled HIV-1-like particles containing only unprocessed gp160 glycoproteins. Monkey kidney Vero cells were transfected with an inducible, human metallothionein-based expression vector containing most of the HIV-1LAI coding sequences that were genetically modified to introduce safety mutations and destroy the major cleavage site of the HIV-1 envelope glycoprotein. A stably-transfected cell line was isolated and shown to secrete HIV-1-like particles containing unprocessed gp160. Immunization with these particles induced HIV-1 cross-neutralizing, syncytium-inhibiting and env-CD4 blocking antibodies. Thus, these novel HIV-1-like particles represent alternative candidate immunogens for the development of a particle-based AIDS vaccine.

Paired sera and cervicovaginal secretions or seminal fluids, obtained from HIV-1-infected, clinically asymptomatic women (n = 41) and men (n = 12), were investigated in order to test the hypothesis of a local synthesis of IgG to HIV in the female and male reproductive tracts. Anti-gp41 + p24 IgG was evaluated by an IgG immunocapture assay, and anti-gp160 IgG by an indirect ELISA. Estimation of anti-HIV IgG-specific activities was carried out after ponderal determination of total IgG and evaluation of anti-HIV IgG activity. IgG to gp41 + p24, as well as IgG to gp160, were specifically detected in all sera, cervicovaginal secretions, and seminal fluid samples from all tested HIV-1-infected subjects. The mean specific activities of IgG to gp41 + p24 in cervicovaginal secretions and in seminal fluids were about 33-fold (in women) and 16-fold (in men) that of the corresponding sera; similarly, the mean specific activities of IgG to gp160 in genital secretions were about 17-fold (in women) and 10-fold (in men) that of the corresponding sera. IgGs to HIV are constantly detected in genital secretions from HIV-1-infected subjects, and appear to be largely synthesized in situ within the genital tract of both genders.

Human anti-CD4 IgG antibodies from 3 HIV-1-infected patients were affinity purified and shown to inhibit HIV-1 binding and infection of HBP-T cells. Lymphocytes from patient A, whose anti-CD4 inhibited HIV-1 binding by 68% and infection by 72%, were cultured and transformed with EBV. A human monoclonal antiidiotype antibody against anti-HIV-1 gp120 (2B) was produced, which inhibited infection of HBP-T cells by 68% at 1 microgram/ml. Mice were immunized with 2B to determine whether this anti-CD4 could be an internal image antiidiotype against anti-HIV-1 gp 120 (Ab1). Two mice produced antisera reactive with rgp120 on ELISA, whereas immunization with normal IgG produced minimal reactivity compared to unreactive normal mouse sera. However, immunoblot competition studies in which affinity-purified anti-HIV-1 gp120 (Ab1) bound to the gp120 band on nitrocellulose strips in the presence of 2B demonstrated enhancement of signal (i.e., binding of Ab2 to Ab1), rather than inhibition of Ab1 binding. Thus 2B is not an internal image of the paratope of anti-HIV-1 gp120 but yet it is capable of inducing an antibody against rgp120. This indicates that the anti-CD4 (Ab2) does bind to the binding site of Ab1, but not as a complete internal image. These data indicate the production of a human monoclonal antiidiotype antibody that inhibits binding of HIV-1 to CD4 and induces the production of antibody against HIV-1 gp120, without being an internal image antiidiotype (Ab2 beta).

Recombinant adenovirus (Ad)-human immunodeficiency virus (HIV) vaccines expressing HIVIIIB Env and Gag proteins were evaluated for immunogenicity in chimpanzees following intranasal administration. When Ad7-, Ad4-, and Ad5-vectored vaccines were administered sequentially at 0, 24, and 52 weeks, respectively, to three chimpanzees, the inoculations resulted in limited virus replication in the nasopharynx, but extensive Ad-HIV replication occurred in the intestine. High-titered IgG serum antibody responses to Env and Gag that were nonneutralizing were induced following booster administration of Ad4-HIV recombinant viruses. Following the Ad5-HIV booster, low levels of neutralizing antibodies as well as V3 loop antibodies were induced in all three chimpanzees that persisted for several months. Administration of a gp160 subunit vaccine (baculovirus derived) in SAF-m 24 weeks later boosted broadly neutralizing serum antibodies that peaked within 1 month of the injection. Two additional subunit boosters 19 and 37 weeks later were progressively less effective at stimulating serum neutralizing antibody responses. Substantial local immune responses were induced in nasal, vaginal, and salivary secretions following the third Ad-HIV intranasal immunization. These responses were further boosted with the gp160 subunit vaccine, which also stimulated production of rectal antibodies. The predominant responses in all secretions tested were of the IgG isotype, although some IgA responses were also detected. Strong blastogenic responses to HIV recombinant Env and Gag proteins were induced after each immunization.

The potential for eliciting humoral and cytotoxic T lymphocyte (CTL) responses to HIV-1 gp120 by gene gun-based DNA immunization in mice was examined. We speculated that the induction of de novo antigen production in the epidermis of BALB/c mice following particle bombardment-based gene delivery would result in both MHC class I- and class II-mediated antigen presentation for the elicitation of CTL and antibody responses, respectively. Following epidermal delivery of microgram quantities of an expression plasmid, gp120 production resulted in the appearance of MHC class I-restricted, CD8+ CTL responses. gp120-specific CTL responses peaked following a booster immunization, then declined with the appearance of gp120-specific IgG responses when additional booster immunizations were administered. This qualitative progression in the nature of gp120-specific immune responses with subsequent immunizations was paralleled by a simultaneous shift in the interferon-gamma and interleukin 4 release profiles following antigen stimulation of splenocytes in vitro. The simultaneous shifts in immune responses and cytokine release profiles indicate that the progression of antigen-specific CTL and IgG responses in gp120 DNA-immunized mice may be mediated through changes in the in vivo production of cytokines, such as those associated with the Th1 and Th2 subsets of CD4+ cells.

T epitope mapping in human immunodeficiency virus proteins provides a useful tool for AIDS vaccine design. We have previously shown that four peptides selected from the Gag polyprotein of HIV-1 were able to prime mice for in vitro lymphoproliferative responses. These responses were shown to be MHC restricted, and a pool of these peptides was able to prime mice for a subsequent humoral response to HIV-1 Gag proteins. Here we show that two of these Gag peptides are able to prime the anti-HIV-1 IgG response to heat-inactivated HIV-1 in B10Sc.Cr mice. Furthermore, we extended this study in the nonhuman primate model, and show efficient priming of the IgG response to heat-inactivated HIV-1 using the pool of four Gag peptides in baboons. Further mapping of "nonself" peptides is extended to the HIV-1 Nef protein. Three potential Nef T epitopes located at positions 137-145, 98-107, and 81-95 are also shown to prime the IgG response to HIV-1 in the mouse model, although T cell proliferation to recall peptides in vitro was not detectable. Although they have not yet been defined as major helper T epitopes in humans, using classic in vitro stimulation assays, the fact that most of them are able to prime IgG responses in animals without detectable in vitro proliferative responses does not rule out their functional helper capacity in humans.

Monoclonal antibodies against a synthetic peptide (aa 138-152) from HIV-1 Nef protein were produced and characterized. Three hybridoma lines producing monoclonal antibodies (MAbs) against the synthetic peptide were generated by fusion between P3-X63 Ag8.653 myeloma cells and BALB/c splenocytes from mice immunized with the synthetic peptide coupled to keyhole limpet hemocyanin (KLH). The hybridomas were screened and selected by ELISA with the peptide coupled to bovine serum albumin (BSA) immobilized to the polystyrene surface and specificity for the peptide was confirmed by competitive ELISA with the peptide free in solution. The reactions of the MAbs with a 5-aa motif (WCYKL) included in the sequence were examined with synthetic peptides and two of the MAbs reacted with the motif. The recognitions of recombinant full-length Nef protein were also tested. One MAb reacted with the protein in both ELISA and dot blot, and one only in dot blot, whereas the last MAb did not recognize the recombinant full-length Nef protein.

Electrofusion and EBV transformation were studied by immortalizing human PBLs from blood of HIV-1-positive volunteers. A panel of 33 cell lines producing human monoclonal antibodies (Hu-MAbs) against HIV-1 was established by cell fusion or EBV transformation. For the first fusion experiments the source of B lymphocytes was peripheral blood of HIV-1-infected donors in CDC stages II or III with CD4 cell counts higher than 500/mm3. Later on, from these patients only, those with high anti-HIV titers were chosen as blood donors. By that means the yield of stable specific hybridomas was increased twofold. In our experiments electrofusion turned out to be a more efficient immortalization method than EBV transformation, due to a high and constant immortalization rate. The hybridomas were stable after intensive subcloning and could be cultivated over a period of 8 months without loss in monoclonal antibody production. Immunoglobulin class, subtype, reactivity against HIV-1 proteins, Western blot patterns, immunofluorescence, and epitopes were characterized. The subtype of all antibodies was IgG1 or IgG3. The light chain was predominantly kappa. All antibodies showed reactivity against HIV-1 envelope or core protein. All hybridomas were stable and suited for mass production. Several Hu-MAbs are becoming an important tool in the field of diagnosis, research, and immunotherapy.

A gene encoding a multiepitope polypeptide (MEP) has been synthesized. It contains the information for (1) an 11-amino acid (aa) epitope from the C1 region of gp120 of HIV-1 and (2) 3 epitopes of 15 amino acids each, from the central part of the V3 loop of isolates MN, SC, and WMJII. These four segments are linked by the short spacer peptide AGGGA. This gene was cloned in a plasmid vector and expressed in Escherichia coli as a fusion product with a 62-aa fragment of human IL-2. The recombinant protein TAB1 was purified by washed pellet procedures and reversed-phase HPLC. TAB1 was recognized in ELISAs by 25 of 27 sera from seropositive individuals. Mice were immunized and several hybridomas were obtained. Two of them secrete monoclonal antibodies that react with synthetic peptides from isolates MN, WMJI, WMJIII, and SC with an affinity constant in the range of 10(8) M-1. They also recognized peptides from isolates SF2 and WMJII, but at much lower affinity. The results obtained from peptide ELISAs indicate that the putative epitope recognized by these MAbs lies within the sequence IHIGPGRAFYT. Classic neutralization assays demonstrated that MAb 2C4 neutralizes 50% of the MN isolate at 0.6 micrograms/ml but fails to neutralize IIIB and SF2 strains. The presence of antibodies directed against every one of the component peptides in the sera of rabbits immunized with TAB1 was also documented.

A formaldehyde-fixed cell vaccine in adjuvant (syngeneic cells infected with a vaccinia virus recombinant expressing gp160: vacc-gp160) stimulated only nonneutralizing antibody when used on its own in four strains of mice, but a similar nonfixed cell vaccine stimulated neutralizing antibodies up to a titer of 1/320 in C57BL/6 (H-2b) mice previously infected with live vacc-gp160. Synthesis of ELISA antibodies to rgp120 or rgp160 did not correspond closely to the synthesis of neutralizing antibodies and should not therefore be used to monitor the production of neutralizing antibody. The ELISA antibody response produced by boosting with the cell vaccine made with the vaccinia virus expressing gp160 under the control of a T7 promoter (vacc-gp160-PT7) was as high as that in mice given an approximately 10-fold higher dose of purified rgp160. The ELISA antibody response to the cell vaccine made with vacc-gp160-PT7 was better than that in which gp160 was expressed under the vaccinia early/late promoter (vacc-gp160-P7.5). Nearly all mice (92%; 11 of 12) primed by infection with vacc-gp160 produced comparable levels of neutralizing antibodies after a single boost with rgp160, vacc-gp160-PT7-infected cells, or vacc-gp160-P7.5-infected cells. Neutralization titers peaked at around day 22 after boost, and declined by day 29. A second boost with the same vacc-gp160-infected cells gave increased neutralizing titers in all (eight of eight) mice.(ABSTRACT TRUNCATED AT 250 WORDS)

The successful development of an AIDS vaccine will require formulations that not only invoke the desired immunological response, but also are stable and easy to administer. A single shot MN rgp120 vaccine formulation comprised of MN rgp120 encapsulated in poly (lactic-coglycolic) acid (PLGA) microspheres was developed to provide an in vivo autoboost of antigen. These formulations were designed to yield an in vivo autoboost at 1, 2, 3 or 4-6 months. In addition, PLGA microspheres containing the adjuvant, QS21, were also prepared to provide an in vivo autoboost concomitant with antigen. In guinea pigs, these formulations yielded higher anti-MN rgp120 and anti-V3 loop antibody titers than alum formulations that were administered at higher antigen doses. Different doses of encapsulated MN rgp120 provided a clear and well-defined dose response curve for both anti-MN rgp120 and anti-V3 loop antibody titers. When soluble QS21 was mixed with the encapsulated MN rgp120, the antibody titers were increased by a factor of 5 over the titers with encapsulated MN rgp120 alone. An additional fivefold increase in antibody titers was observed for guinea pigs immunized with encapsulated MN rgp120 and QS21 on the same microspheres. These results suggest that the adjuvant properties of QS21 can be increased by microencapsulation in PLGA. Furthermore, antibodies induced by these preparations neutralized the MN strain of HIV-1. The neutralization titers for sera from animals immunized with MN rgp120-PLGA and soluble QS21 were greater than the titers obtained from guinea pigs that were treated with MN rgp120 and soluble QS21 at the same dose. Overall, these studies validate the in vivo autoboost concept, reveal a method for improving the adjuvant properties of QS21, and indicate the potential of future single shot vaccine formulations.

The effect of adjuvant and immunization schedule on the immunogenicity of HIV-1 envelope glycoprotein, MN rgp120, was optimized by using baboons. The novel adjuvant QS21 elicited earlier seroconversion than alum adjuvant, and the antibody titers to MN rgp120 for animals treated with QS21 were significantly greater than the titers obtained in animals treated with alum. The use of QS21 shifted the dose-response curve, resulting in less MN rgp120 required to achieve equivalent titers to those in the alum formulations. The in vitro virus neutralizing (VN) titers from animals treated with QS21 were 3- to 10-fold higher than with alum. The data presented herein point to the superiority of QS21 as adjuvant in primates for MN rgp120.

Accell particle-mediated gene delivery technology was employed for the intracellular delivery of antigen-encoding expression vectors in epidermal tissues in laboratory animals. Delivery of plasmid DNA-coated gold microparticles using the Accell gene delivery system resulted in de novo antigen expression in epidermal cells that stimulated the induction of antigen-specific humoral and cytotoxic cellular immune responses. Optimal DNA delivery conditions favoring maximal humoral responses required the delivery of 5 x 10(7) micron-sized gold particles containing 300 plasmid copies per particle (80 ng of vector total) into a 4-cm2 area of abdominal skin. Comparison of immune responses between animals that received intramuscular injections of relatively large quantities of vector DNA (100 micrograms) and those that received intracellular deliveries of submicrogram quantities of the same DNA to the epidermis demonstrated that the latter approach was considerably more effective at eliciting strong humoral responses. In addition, cytotoxic cellular immune responses were elicited to HIV-1 gp120 following epidermal delivery of HIV-1 gp160 or gp120 expression constructs. A qualitative shift from predominantly cytotoxic cellular to predominantly humoral immune responses with continued immunization indicated the potential for optimizing delivery conditions to favor specifically one type of response over the other.

We have compared in the macaque model the efficacy of the virion-derived glycoprotein of HIV-2ben (HIV-2 gp130) with that of SIVmac251/32H (SIV gp130). The latter vaccination trial was in part combined with vaccinia virus (VV) priming. Both antigen preparations induced a strong humoral, but a weak cellular, immune response. The first challenge was performed with autologous virus grown on a human T cell line. More than 50% of the monkeys immunized with HIV-2 gp130 (five of nine) and 63% of the monkeys immunized with SIV gp130 (five of eight) were protected. All such protected animals received one or two booster immunizations before they were rechallenged either with heterologous HIV-2SBL6669 grown on monkey peripheral blood mononuclear cells or with an ex vivo stock of SIVmac251/32H prepared from the spleen of an SIV-infected macaque and not passaged in vitro. Immunization with HIV-2 gp130 did not protect against the second challenge, but one animal showed limited infection as indicated by positive PCR only. Challenge of the SIV gp130-immunized monkeys with the spleen-derived virus led to infection of three animals; remarkably, one of these was only PCR positive. Two animals were completely protected. Thereby we can exclude the influence of cellular proteins on protective immunity. Priming with VV was not superior to immunization with gp130 alone. Neither at the first nor at the second challenge were the virus-specific humoral and cellular immune responses of the vaccinees clearly correlated with protection. However, neutralizing antibodies may have been important in the SIV gp130-immunized animals at first challenge.

We have made use of certain novel genetic elements of picornaviruses termed internal ribosomal entry sites (IRES) to construct a viral RNA with the following genetic order: PV 5' NTR-EMCV IRES-PV ORF-3' NTR (PV, poliovirus; NTR, nontranslated region; EMCV, encephalomyocarditis virus; ORF, open reading frame). Transfection of this RNA into HeLa cells yielded a poliovirus (W1-PNENPO) that contained two heterologous IRES elements (type 1 IRES of PV; type 2 IRES of EMCV) in tandem. The insertion of foreign coding sequences into the genome of W1-PNENPO between the IRES elements yielded viable polioviruses with the gene order PV 5' NTR-foreign ORF-EMCV IRES-PV ORF-3' NTR. The foreign ORFs we have employed in this study included the coding region for chloramphenicol acetyltransferase (CAT), or segments of either luciferase or the HIV-1 envelope glycoprotein gp120. W1-PV/V3-3, a dicistronic poliovirus that contained HIV-1-specific sequences that included the V3 domain of gp120, was used to infect transgenic mice (PVR+) that were engineered to express the poliovirus receptor. The genetic stability of the dicistronic viruses and the HIV-1-specific immune response in PVR+ mice after infection with these novel agents are discussed.

DNA, or genetic, inoculation mimics aspects of attenuated vaccines in that synthesis of specific foreign proteins is accomplished in the host. These proteins can be processed and presented on the relevant major histocompatibility complex (MHC) antigens and ultimately become the subject of immune surveillance. Very recently, we have described the use of the new technology to generate immune responses in mice against the human immunodeficiency virus type 1 (HIV-1) envelope using a gp160 DNA construct. Further analysis of this technology specifically in regard to HIV vaccine design is clearly important. In this report, we describe the analysis of additional HIV constructs as immunogens in both mice and report the use of this genetic immunization technology in nonhuman primates. In these studies, successful seroconversion occurs in more than 70% of the mice following the second immunization with 100 micrograms of construct DNA; three and four immunizations result in routinely 100% seroconversion of the mice. Furthermore, the same strategy has successfully seroconverted primates following their second inoculation, resulting in the generation of both antiviral and neutralizing antibodies in this animal species. These studies are the first report of which we are aware that demonstrate successful immunization of nonhuman primates through genetic vaccination technology and the first to describe genetic immunization of primates against HIV antigens. This technology has relevance for the development of safe and efficacious immunization strategies against HIV because it provides for relevant antigen production in vivo without the use of infectious agents.

We have used a human immunodeficiency virus type 1 (HIV-1)-specific IgG-Fc capture enzyme immunoassay (IgG-CEIA) to elucidate the dynamics of HIV-1 maternal antibody decay and de novo synthesis of HIV-1 antibodies in infants. Two hundred and thirty-nine serum specimens from 77 infants were analyzed by the IgG-CEIA and by two different conventional EIAs. With the IgG-CEIA, IgG was captured by an anti-human IgG monoclonal antibody (3C8) that reacts with all subclasses and was detected by recombinant HIV-1 envelope protein (CBre3)-peroxidase conjugate. Unlike the conventional EIAs, the IgG-CEIA showed a rapid decay of HIV-1-specific antibody in uninfected infants, with decline to background levels by 6 months (T1/2 [half-life] = 28-30 days). All 69 specimens collected from 39 uninfected infants between 6 and 15 months of age were negative by IgG-CEIA. However, HIV-1 antibodies remained high in infected infants; 20/22 infants (90.9%) with specimens between the ages of 6 to 23 months were positive by IgG-CEIA. Subtracting mean IgG-CEIA optical density values of seroreverting infants from those of HIV-1-infected infants in corresponding age groups provided a model for seroconversion in infected infants, with detectable IgG antibody synthesis starting about 3 months after birth. The IgG-CEIA may be a simple and important tool for early diagnosis of HIV-1 infection in infants at 6 months of age.