Prime-boost vaccination with recombinant mumps virus and recombinant vesicular stomatitis virus vectors elicits an enhanced human immunodeficiency virus type 1 Gag-specific cellular immune response in rhesus macaques.

Intramuscular inoculation of rhesus macaques with one or more doses of recombinant vesicular stomatitis virus (rVSV) expressing human immunodeficiency virus type 1 (HIV-1) Gag (rVSVgag) typically elicits peak cellular immune responses of 500 to 1,000 gamma interferon (IFN-gamma) enzyme-linked immunospots (ELISPOTS)/10(6) peripheral blood lymphocytes (PBL). Here, we describe the generation of a novel recombinant mumps virus (rMuV) expressing HIV-1 Gag (rMuVgag) and measure the Gag-specific cellular immune responses detected in rhesus macaques following vaccination with a highly attenuated form of rVSV expressing HIV-1 Gag (rVSVN4CT1gag1) and rMuVgag in various prime-boost combinations. Notably, peak Gag-specific cellular immune responses of 3,000 to 3,500 ELISPOTS/10(6) PBL were detected in macaques that were primed with rMuVgag and boosted with rVSVN4CT1gag1. Lower peak cellular immune responses were detected in macaques that were primed with rVSVN4CT1gag1 and boosted with rMuVgag, although longer-term gag-specific responses appeared to remain higher in this group of macaques. These findings indicate that rMuVgag may significantly enhance Gag-specific cellular immune responses when administered with rVSVN4CT1gag1 in heterologous prime-boost regimens.

Effective mucosal immunization against respiratory syncytial virus using purified F protein and a genetically detoxified cholera holotoxin, CT-E29H.

We exploited the powerful adjuvant properties of cholera holotoxin (CT) to create a mucosally administered subunit vaccine against respiratory syncytial virus (RSV). A genetically detoxified mutant CT with an E to H substitution at amino acid 29 of the CT-A1 subunit (CT-E29H) was compared to wild type CT for toxicity and potential use as an intranasal (IN) adjuvant for the natural fusion (F) protein of RSV. When compared to CT the results demonstrated that: (1) CT-E29H binding to GM1 ganglioside was equivalent, (2) ADP-ribosylation of agmatine was 11.7%, and (3) toxicity was attenuated in both Y-1 adrenal (1.2%) and patent mouse gut weight assays. IN vaccination with F protein formulated with CT-E29H induced serum anti-CT and anti-F protein antibodies that were comparable to those obtained after vaccination with equivalent doses of CT. Vaccinations containing CT-E29H at doses of 0.1 microg were statistically equivalent to 1.0 microg in enhancing responses to F protein. Antigen-specific mucosal IgA and anti-RSV neutralizing antibodies were detected in nasal washes and sera, respectively, of mice that had received F protein and 0.1 or 1.0 microg of CT-E29H. Anti-F protein IgA was not detected in the nasal washes from mice IN vaccinated with 0.01 microg CT-E29H or IM with F protein adsorbed to AlOH adjuvant. In addition, the formulation of purified F protein and CT-E29H (0.1 and 1.0 microg) facilitated protection of both mouse lung and nose from live RSV challenge. Collectively, the data have important implications for vaccine strategies that use genetically detoxified mutant cholera holotoxins for the mucosal delivery of highly purified RSV antigens.

Induction of mucosal antibody responses by microsphere-encapsulated formalin-inactivated simian immunodeficiency virus in a male urethral challenge model.

Male rhesus macaques were immunized mucosally with microsphere-encapsulated formalin-inactivated simian immunodeficiency virus (SIV) particles in a test of immunogenicity and protection against mucosal SIV challenge. Tracheal boosting of animals that had been primed intramuscularly resulted in strong serum ELISA titers to SIV, and evidence of local IgA responses in broncho-alveolar washes. The bulk of the antibody response was against non-envelope epitopes. No neutralizing antibody was observed, and intraurethral challenge with cell-free rhesus-grown virus showed no evidence of protection against infection. Microsphere-based immunization efficiently raises local and system responses, but the resulting immunity to SIV is apparently not sufficient to protect against mucosal challenge.

Combined systemic and mucosal immunization with microsphere-encapsulated inactivated simian immunodeficiency virus elicits serum, vaginal, and tracheal antibody responses in female rhesus macaques.

We determined the efficacy of immunization with microsphere-encapsulated whole inactivated simian immunodeficiency virus (SIV) by combined systemic and mucosal administration to protect female rhesus macaques against vaginal challenge with homologous rhesus PBMC-grown SIVmac251. Animals in one group were primed and boosted intramuscularly. Two groups were primed intramuscularly and boosted either intratracheally or orally. A final group was primed by vaccinia/rgp140 scarification and subdivided for either intratracheal or oral boosting. Strong ELISA titers of circulating SIV-specific IgG and modest IgA responses were elicited in the animals primed intramuscularly. Intratracheal boosting in the intramuscularly primed macaques resulted in high bronchial alveolar wash (BAW) IgG and less pronounced IgA. SIV-specific vaginal wash (VW) IgG was also present in the intramuscular/intramuscular and intramuscular/intratracheal groups. Vaccinia/rgp140 priming gave low ELISA titers to whole SIV, and failed to elicit mucosal antibody regardless of the booster route. No animal in any group developed serum neutralizing antibody to homologous SIVmac251. On vaginal challenge none of the immunized groups was infected at a lesser frequency than the unimmunized controls. These data suggest that the use of microspheres in a combined parenteral and mucosal regimen is an effective method of eliciting IgG and IgA antibody at mucosal surfaces.

Induction of protective immune responses against Venezuelan equine encephalitis (VEE) virus aerosol challenge with microencapsulated VEE virus vaccine.

Venezuelan equine encephalomyelitis (VEE) virus, a member of the family Togaviridae, genus Alphavirus, causes disease in humans and equids. The virus is normally transmitted by the bite of an infected mosquito however, it can also be highly infectious by aerosol. The purpose of the present study was to determine the effectiveness of formalin-fixed, 60Co-irradiated VEE virus microencapsulated in poly DL-lactide-co-glycolide in inducing immune responses protective against aerosol challenge with virulent VEE virus. Balb/c mice were primed by subcutaneous injection of microencapsulated VEE virus vaccine, followed 30 days later by a single immunization with the same vaccine given via the oral, intratracheal (i.t.) or subcutaneous (s.c.) route. Mice boosted by the i.t. or s.c. route had higher plasma IgG anti-VEE virus levels than orally immunized animals. The responses in the former groups were similar in magnitude to those seen in mice primed and boosted by the i.t. route. Antibody activity was detected in bronchial-alveolar and intestinal washes, fecal extracts and saliva from immunized animals. The levels of IgG and IgA antibody activity in bronchial-alveolar wash fluids from mice boosted by the i.t. route were higher than those seen in animals immunized by the oral or s.c. route with the microsphere vaccine. Mice immunized with the microencapsulated VEE virus vaccine were protected from lethal VEE virus infection following aerosol challenge at approximately three months after the initial immunization. Mucosal immunization via the i.t. route appeared to be the most effective regimen, since 100% of the mice resisted aerosol challenge.

Development of antibody-secreting cells and antigen-specific T cells in cervical lymph nodes after intranasal immunization.

Intranasal (i.n.) immunization with bacterial protein antigens coupled to cholera toxin B subunit (CTB) effectively induces mucosal, especially salivary immunoglobulin A (IgA), and nonmucosal antibody responses in mice. To examine the regional distribution of antigen-specific B and T cells after i.n. immunization, antibody-secreting cells and antigen-responsive T cells in cervical lymph nodes (CLN) were compared with those found after intraoral or subcutaneous (in the neck) administration of the same antigen and with T cells found in mesenteric lymph nodes (MLN) and spleen after intragastric immunization. The i.n. immunization induced predominantly IgA antibody-secreting cells in salivary glands and IgA and IgG antibody-secreting cells in the superficial and central CLN; these responses were quantitatively enhanced if the antigen was coupled to CTB. Intraoral immunization also induced IgA and IgG antibody-secreting cells in the superficial and central CLN, but only if intact cholera toxin was included as an adjuvant. In contrast, subcutaneous (neck) immunization induced IgG antibody-secreting cells mainly in the draining facial lymph nodes. CLN cell populations resembled those of MLN, except that CLN lymphocytes had higher proportions of T cells and lower proportions of B cells and a slightly higher CD4+/CD8+ ratio among T cells than the MLN lymphocytes did. T cells that proliferated in response to antigen in vitro were found especially in central CLN 2 days after i.n. immunization and persisted for up to 6 months, whereas after intragastric immunization, responsive T cells were not found in the MLN for up to 14 days. After culture with antigen in vitro, T cells from the superficial CLN of i.n. immunized mice secreted both gamma interferon and interleukin-4. Therefore, after i.n. immunization, superficial and central CLN represent sites of regional lymphocyte development, and the central CLN in particular appear to be sites where memory T cells persist.

Host responses induced by co-infection with Porphyromonas gingivalis and Actinobacillus actinomycetemcomitans in a murine model.

In this study, evidence is presented that mixed infection with the periodontal pathogens Porphyromonas gingivalis and Actinobacillus actinomycetemcomitans results in a synergistic effect in their pathogenicity and in their ability to induce humoral and cellular host responses. BALB/c mice were injected subcutaneously on the back with P. gingivalis ATCC 53977, A. actinomycetemocomitans 75 or a mixture of both bacteria. Samples of blood and fluid from abscesses formed at the site of injection (first degree) or distant from the injection site were collected for microbiologic analysis. Serum and spleens were obtained for evaluation of humoral and cellular responses to P. gingivalis and A actinomycetemocomitans. Mice injected with A. actinomycetemcomitans had first-degree lesions only, whereas mice injected with P. gingivalis and A. actinomycetemcomitans had lesions at first- and second-degree sites from which both bacterial species were isolated. A serum anti-P. gingivalis response was induced in P. gingivalis-injected mice, which was higher in mice injected with P. gingivalis and A. actinomycetemcomitans. This pattern was not seen in the anti-A, actinomycetemcomitans response. Lymphoproliferative responses to phytohemagglutinin, Escherichia coli lipopolysaccharide and P. gingivalis of spleen cells from infected mice were decreased, especially following co-infection. Furthermore, co-infection of mice resulted in the greatest decrease in the number of CD5+, especially CD4+ lymphocytes.

Peritoneal cavity CD5 (Bla) B cells: cytokine induced IgA secretion and homing to intestinal lamina propria in SCID mice.

The mouse peritoneal cavity contains a unique population of B cells (Bla) with a high IgM/low IgD ratio, CD5+ (Ly1), MAC-1 + phenotype. These cells arise early in ontogeny, utilize a limited repertoire of immunoglobulin V genes, produce polyreactive IgM antibodies and have been implicated as the source of many auto-reactive immunoglobulins. Recent data from chimeric mice suggest that this B cell population also contains the precursors of many IgA plasma cells found in the lamina propria of the small intestine. In the present study we have investigated the potential of this cell population to secrete IgA (and IgG) in response to various cytokines. IL-5 alone, or in combination with IL-2, greatly enhanced secretion of both IgG and IgA. Cytokine-induced IgA secretion resulted from expansion of a subset of CD5 B cells co-expressing sIgA. Adoptive transfer of CD5 B cells while peripheral lymph nodes contained only IgM+ and some IgG+ B cells. Transfer of CD5+ B cells also reconstituted serum IgM, IgG and IgA and IgG, immunoglobulins characteristic of mucosal and anamnestic responses, when cultured in vitro with the appropriate cytokines. These cells also give rise to IgA plasma cells in the intestinal lamina propria following adoptive transfer to SCID mice, further supporting the hypothesis that cells of this lineage may be important in immune responses at mucosal surfaces.

IgA production by peritoneal cavity B cells is IL-6 independent: implications for intestinal IgA responses.

We have shown previously both in vitro and in vivo that IL-6 is an important factor for the development of IgA-producing B cells. However, despite the lack of this cytokine in mice with targeted disruption of the interleukin (IL)-6 gene (gene knockout mice), a substantial number of IgA-producing plasma cells occur in their intestinal mucosa. The experiments reported here indicate that there is a population of IgA-producing B cell precursors originating from the peritoneal cavity, distinguished from conventional Peyer's patch-derived precursors by their expression of CD5, and that IgA secretion by these cells is IL-6-independent. Further, there is an increase in CD5 expression among brightly staining IgA-producing cells obtained from the intestinal lamina propria of IL-6 gene-disrupted mice compared to normal controls. These data suggest an explanation for the persistence of IgA-producing plasma cells in the intestinal mucosa of IL-6-depleted mice and indicate the importance of IL-6 for development of conventional precursors of IgA-producing B cells, but not those derived from the peritoneal cavity pool.

Generation of monoclonal antibodies to Macaca mulatta (rhesus) IgA.

One IgG1 and five IgM murine monoclonal antibodies (mAb) specific for rhesus (Rh) IgA were generated. These mAbs bound to Rh IgA but not IgG or IgM when tested by enzyme-linked immunosorbent assay. Immunoblotting revealed that the mAbs reacted with the alpha heavy chain of Rh but not human IgA. The IgG1 anti-Rh IgA mAb detected IgA-producing cells in sections of monkey gut examined by immunofluorescent staining. These mAbs should be useful for characterizing IgA responses in the Rh monkey.

Enhancement of protective immune responses to Venezuelan equine encephalitis (VEE) virus with microencapsulated vaccine.

Venezuelan equine encephalomyelitis (VEE) virus is a mosquito-borne arbovirus of major human health significance in the New World. Currently two forms of VEE virus are used for immunization of humans and horses, i.e. a live attenuated and a formalin-inactivated vaccine. Clinical evidence suggests that these vaccines are not fully efficacious and may produce certain undesirable side-effects. In the present study, microspheres composed of biocompatible and biodegradable poly (DL-lactide-co-glycolide) (DL-PLG) were evaluated for their effectiveness as a delivery system of whole, inactivated VEE virus vaccine for the induction of protective immune responses. Mice receiving 50 micrograms VEE virus in microspheres composed of an equimolar ratio of DL-lactide and glycolide (50:50 DL-PLG) exhibited a primary circulating IgG antibody response which was approximately 32-times higher than the response induced with the same dose of unencapsulated (free) virus. A similar difference in responses was seen with antigen doses ranging from 3.1 to 50 micrograms. A rapid increase in antibody activity was seen after the secondary immunization (day 50). Formalin fixation of inactivated VEE virus was important for immunogenicity since the circulating anti-VEE virus antibody response induced with microencapsulated nonformalin-fixed virus vaccine was lower than that induced with microencapsulated formalin-fixed virus vaccine. Furthermore, at low antigen concentrations, DL-PLG microsphere vaccines prepared with the solvent methylene chloride induced higher antibody responses than those prepared using ethyl acetate as the solvent. Microencapsulated vaccine also induced higher VEE virus neutralization titers than did free virus vaccine. Finally, the microencapsulated virus was more effective than the free virus in inducing immune responses protective against systemic challenge with virulent VEE virus. These results demonstrate that DL-PLG microspheres containing formalin-fixed, inactivated VEE virus were effective in augmenting circulating IgG antibody levels and neutralization titers to the VEE virus following systemic immunization and in affording enhanced protection against systemic challenge with virulent VEE virus. The effects of antigen form and the microsphere processing solvent on the immunogenicity of the vaccine are discussed.

Peyer's patch CD8+ memory T cells secrete T helper type 1 and type 2 cytokines and provide help for immunoglobulin secretion.

Analysis of cytokine gene expression by reverse transcription-polymerase chain reaction (RT-PCR) demonstrated high spontaneous levels of transcripts for multiple cytokines in murine Peyer's patches (PP) compared to spleen and peripheral lymph nodes. This is consistent with the presence of active germinal centers in PP and their continuous exposure to lumenal antigen including bacterial endotoxin. RT-PCR analysis of cytokine transcripts in purified PP T cell populations revealed the presence of transcripts for interleukin-4 (IL-4), IL-5 and IL-10 in addition to interferon-gamma (IFN-gamma) in CD8+ cells purified by flow cytometry. The majority of PP CD8+ T cells were also CD45RBlo (MB23G2-), suggesting that these cells were activated/memory cells. CD8+ cells in spleen and mesenteric lymph nodes (MLN) were predominantly CD45RBhi (MB23G2+) consistent with a resting/naive phenotype. PP and MLN CD8+ T cells also secreted IL-5 and IL-10 when stimulated with anti-CD3 monoclonal antibody and when co-cultured with PP B cells enhanced secretion of both IgG and IgA. These studies suggest that CD8+ T cells at mucosal sites secrete T helper type 2 cytokines and can provide functional help for B cells in these tissues.

Contrasting IgA and IgG neutralization capacities and responses to HIV type 1 gp120 V3 loop in HIV-infected individuals.

Quantitative analysis for HIV-1-specific antibodies present in IgA and IgG preparations purified from the serum of HIV-seropositive individuals indicated that the proportion of HIV-specific antibodies present within the IgG isotype was seven times greater than the proportion of IgA HIV antibodies present within the IgA isotype. Dilution of IgA HIV-specific antibodies by nonspecific IgA was observed in patients with elevated serum IgA concentrations, whereas proportions of IgG HIV antibodies rose with increases in concentrations of serum IgG. Although proportions of IgA HIV antibodies were not observed to correlate with the CD4 counts of the individuals from whom immunoglobulins were purified, a significant association between the numbers of such cells and proportion of HIV antibodies present in the IgG isotype was found. Equivalent amounts of IgG were also more effective than IgA at inhibiting HIV-1IIIB infection of a susceptible T cell line. This may be due to the presence of higher proportions of IgG antibodies directed toward non-V3 determinants because reactivity against an HIV-1IIIB V3 peptide was low and did not differ significantly between these isotopes. IgA antibodies reacting against a V3 peptide containing the HIV consensus sequence could be detected in the majority of IgA samples purified from infected individuals. Proportions of IgG consensus V3-specific antibodies within the purified IgG samples were, however, much higher. The presence of accompanying increases in serum IgG concentration and proportions of IgG HIV antibodies, higher proportions of both HIV- and consensus V3-specific antibodies within this isotype, and more effective neutralization by IgG suggests that an HIV-driven response is dominated by B cells committed to production of this immunoglobulin isotype. The observed low proportions of HIV antigen-specific IgA antibodies with dilution in many individuals by elevations in non-HIV-specific IgA suggests that IgA B cells may be more susceptible to factors that mediate the polyclonal activation believed to be responsible for many of the B cell disorders characteristic of HIV infection.

Characterization of T and B cells isolated from mucosa-associated tissues of the rhesus macaque.

This study has assessed T cell subsets according to the expression of CD4 and CD8 and the isotype of surface Ig+ (sIg+) B cells and plasma cells in mononuclear cells (MC) isolated from mucosa-associated tissues of rhesus macaques in comparison to lymphoid cells from systemic tissues, i.e., spleen and peripheral lymph nodes (PLN). Using enzymatic and/or mechanical dissociation methods, mononuclear cells were isolated from lamina propria (LP) of the small (jejunum and ileum) and large (cecum) intestines, parotid glands (PG), and mesenteric lymph nodes (MLN). Approximately equal numbers of CD4+ and CD8+ T cells occurred in mucosa-associated tissues (CD4:CD8 ratios of 0.9-1.2), while slightly higher numbers of CD8+ T cells were seen in spleen and PLN (CD4:CD8 ratios of 0.6-0.9). When the isotypes of sIg+ B cells were assessed in MC isolated from mucosa-associated tissues, the highest frequency of sIgA+ B cells was seen in MLN, which may represent B cells which had recently migrated from IgA inductive sites. However, IgA effector tissues, such as intestinal LP and PG, contained frequencies of sIgA+ B cells lower than those of the MLN. When Ig-producing cells were examined, 50-90% of Ig-secreting cells were of the IgA class. This suggests that sIgA+ B cells differentiate into IgA plasma cells at higher rates in the mucosal effector tissues. Plasma cells of IgM isotype were also found in significant numbers in the LP of the intestine and PG, while IgG plasma cells were most prevalent in spleen and PLN. Taken together, mucosa-associated tissues of rhesus macaques are characterized by higher numbers of CD4+ T cells and Ig-secreting plasma cells of IgA and IgM isotypes than those of systemic lymphoid tissues.

New advances in vaccine delivery systems.

Successful application of the next generation of vaccines will require that protection be induced with a minimal number of administrations, and that a practical approach to inducing immunity at mucosal surfaces be developed. For these reasons, vaccine-containing microspheres were formulated from the biodegradable and biocompatible copolymer poly(DL-lactide-co-glycolide) [DL-PLG]. Subcutaneous immunization of mice with 1- to 10-microns microspheres containing a toxoid vaccine of staphylococcal enterotoxin B (SEB) induced a 500-fold potentiation of the circulating antitoxin response. Strong adjuvant activity was dependent on the microspheres being no more than 10 microns in diameter and required that the antigen was within the particles. The rate of DL-PLG biodegradation is a function of the ratio of lactide to glycolide, and the co-injection of SEB toxoid microspheres formulated with two different DL-PLG ratios stimulated both a primary and an anamnestic secondary antitoxin response. When it was administered by the oral or intratracheal (IT) route, microencapsulated SEB toxoid was found to be effective in the induction of concurrent circulating and disseminated mucosal antibody responses. Female rhesus macaques immunized with a microencapsulated simian immunodeficiency virus (SIV) vaccine produced high levels of circulating anti-SIV antibodies, and following oral or IT boosting, specific antibodies were found in vaginal wash fluids. Vaginal challenge with viable homologous SIV resulted in the infection of three out of four nonimmunized but only one out of seven microsphere-immunized macaques. Thus, DL-PLG microspheres are a promising approach to the delivery of vaccines, combining adjuvant activity with controlled release and effective presentation to mucosally associated lymphoid tissues (MALT).

Solution structure of the V3 loop of a Thailand HIV isolate.

The principal neutralizing determinant (PND) of human immunodeficiency virus (HIV) is located inside the third variable loop (designated the V3 loop) of the envelope glycoprotein gp120. The V3 loop is typically 35 amino-acids long, and the 1st and the 35th residues in the loop are invariant cystines involved in a disulfide-bridge. Although PNDs from different HIV isolates contain a conserved GPG-sequence, the amino acids flanking the conserved sequence show hypervariability among HIV isolates; the GPG and the two flanking regions are collectively referred to as the GPG-crest or the PND. The amino acid sequence variability in the GPG-crest gives rise to different antigenic specificities for different PNDs from different HIV isolates. By combining two-dimensional nuclear magnetic resonance (2D NMR) and molecular modeling techniques, we have developed a method to study (1) the global tertiary fold of the V3 loops of HIV and (2) the local structure of the PND at the tip of the V3 loop. In this article, we report the results of our structural studies on the V3 loop of a Thailand HIV isolate. The sequential assignment is made by combining DQF-COSY, TOCSY, and NOESY/ROESY experiments. Various intra- and inter-residue inter-proton distances are estimated by full-matrix analyses of the NOESY data at 100 and 400 ms of mixing times and of the ROESY data at 60 and 200 ms of mixing times. 100 inter-residue distances are used as structural constraints in a simulated annealing procedure to derive energetically stable structures. Two functional motifs in the V3 loop, i.e., the glycosylation site and the GPG-crest, form defined structures: a turn is located at the glycosylation site, and the GPG-crest forms a protruding domain with a type-II GPGQ turn. The other regions of the V3 loop are rather flexible--especially the C-terminal DIRKAYC-stretch. These flexible regions of the V3 loop lead to conformational flexure of the entire V3 loop without altering the local structures of the glycosylation site or the GPG-crest. However, the ROESY experiments revealed no slow exchange among different V3 loop conformations, and therefore the flexible conformations are in fast exchange within the NMR time scale. The extent of this conformational flexibility is also discussed.

Protection against vaginal SIV transmission with microencapsulated vaccine.

Although protection in animal models against intravenous challenges with simian immunodeficiency virus (SIV) has been reported, no previous vaccines have protected against a heterosexual route of infection. In this study, five of six macaques were protected against vaginal challenge when immunized with formalin-treated SIV in biodegradable microspheres by the intramuscular plus oral or plus intratracheal route. Oral immunization alone did not protect. After a second vaginal challenge, three of four intramuscularly primed and mucosally boosted macaques remained protected. The data suggest that protection against human immunodeficiency virus vaginal transmission could be provided by microsphere-based booster vaccines when used to immunize women who are systemically primed.