Effect of T-helper cytokine environment on specificity of T-cell responses to mycobacterial 65,000 MW heat-shock protein

Natural cutaneous anthrax infection, but not vaccination, induces a CD4(+) T cell response involving diverse cytokines

BACKGROUND

Whilst there have been a number of insights into the subsets of CD4(+) T cells induced by pathogenic Bacillus anthracis infections in animal models, how these findings relate to responses generated in naturally infected and vaccinated humans has yet to be fully established. We describe the cytokine profile produced in response to T cell stimulation with a previously defined immunodominant antigen of anthrax, lethal factor (LF), domain IV, in cohorts of individuals with a history of cutaneous anthrax, compared with vaccinees receiving the U.K. licenced Anthrax Vaccine Precipitated (AVP) vaccine.

FINDINGS

We found that immunity following natural cutaneous infection was significantly different from that seen after vaccination. AVP vaccination was found to result in a polarized IFNγ CD4+ T cell response, while the individuals exposed to B. anthracis by natural infection mounted a broader cytokine response encompassing IFNγ, IL-5, -9, -10, -13, -17, and -22.

CONCLUSIONS

Vaccines seeking to incorporate the robust, long-lasting, CD4 T cell immune responses observed in naturally acquired cutaneous anthrax cases may need to elicit a similarly broad spectrum cellular immune response.

Anthrax lethal factor as an immune target in humans and transgenic mice and the impact of HLA polymorphism on CD4+ T cell immunity

Bacillus anthracis produces a binary toxin composed of protective antigen (PA) and one of two subunits, lethal factor (LF) or edema factor (EF). Most studies have concentrated on induction of toxin-specific antibodies as the correlate of protective immunity, in contrast to which understanding of cellular immunity to these toxins and its impact on infection is limited. We characterized CD4⁺ T cell immunity to LF in a panel of humanized HLA-DR and DQ transgenic mice and in naturally exposed patients. As the variation in antigen presentation governed by HLA polymorphism has a major impact on protective immunity to specific epitopes, we examined relative binding affinities of LF peptides to purified HLA class II molecules, identifying those regions likely to be of broad applicability to human immune studies through their ability to bind multiple alleles. Transgenics differing only in their expression of human HLA class II alleles showed a marked hierarchy of immunity to LF. Immunogenicity in HLA transgenics was primarily restricted to epitopes from domains II and IV of LF and promiscuous, dominant epitopes, common to all HLA types, were identified in domain II. The relevance of this model was further demonstrated by the fact that a number of the immunodominant epitopes identified in mice were recognized by T cells from humans previously infected with cutaneous anthrax and from vaccinated individuals. The ability of the identified epitopes to confer protective immunity was demonstrated by lethal anthrax challenge of HLA transgenic mice immunized with a peptide subunit vaccine comprising the immunodominant epitopes that we identified.

Anthrax is of concern with respect to human exposure in endemic regions, concerns about bioterrorism and the considerable global burden of livestock infections. The immunology of this disease remains poorly understood. Vaccination has been based on B. anthracis filtrates or attenuated spore-based vaccines, with more recent trials of next-generation recombinant vaccines. Approaches generally require extensive vaccination regimens and there have been concerns about immunogenicity and adverse reactions. An ongoing need remains for rationally designed, effective and safe anthrax vaccines. The importance of T cell stimulating vaccines is inceasingly recognized. An essential step is an understanding of immunodominant epitopes and their relevance across the diverse HLA immune response genes of human populations. We characterized CD4 T cell immunity to anthrax Lethal Factor (LF), using HLA transgenic mice, as well as testing candidate peptide epitopes for binding to a wide range of HLA alleles. We identified anthrax epitopes, noteworthy in that they elicit exceptionally strong immunity with promiscuous binding across multiple HLA alleles and isotypes. T cell responses in humans exposed to LF through either natural anthrax infection or vaccination were also examined. Epitopes identified as candidates were used to protect HLA transgenic mice from anthrax challenge.

Differences between T cell epitopes recognized after immunization and after infection

Evidence suggests that cellular immune responses play a crucial role in the control of HIV and SIV replication in infected individuals. Several vaccine strategies have therefore targeted these CD8(+) and CD4(+) responses. Whether vaccination induces the same repertoire of responses seen after infection is, however, a key unanswered question in HIV vaccine development. We therefore compared the epitope specificity induced by vaccination to that present postchallenge in the peripheral blood. Intracellular cytokine staining of PBMC stimulated with overlapping 15/20-mer peptides spanning the proteins of SIV were measured after DNA/modified vaccinia Ankara vaccination of eight rhesus macaques. Lymphocytes from 8 animals recognized a total of 39 CD8 epitopes and 41 CD4 epitopes encoded by the vaccine. T cell responses were again monitored after challenge with SIVmac239 to investigate the evolution of these responses. Only 57% of all CD8(+) T cell responses and 19% of all CD4(+) T cell responses present after vaccination were recalled after infection as measured in the peripheral blood. Interestingly, 29 new CD8 epitopes and 5 new CD4 epitopes were recognized by PBMC in the acute phase. These new epitopes were not detected after vaccination, and only some of them were maintained in the chronic phase (33% of CD8 and no CD4 responses). Additionally, 24 new CD8 epitopes and 7 new CD4 epitopes were recognized by PBMC in the chronic phase of infection. The repertoire of the immune response detected in the peripheral blood after immunization substantially differed from the immune response detected in the peripheral blood after infection.

The route of administration of an immunodominant peptide derived from heat-shock protein 65 dramatically affects disease outcome in pristane-induced arthritis

Previous studies have shown that immunization of mice with an immunodominant epitope from heat-shock protein 65 (hsp 65) (amino acids 261-271) can protect from the development of pristane-induced arthritis (PIA) and this protection is mediated by an antigen-specific T helper type 2 (Th2) cytokine response. Here we confirm these findings and show that frequent intranasal administration of this peptide exacerbates disease. In naive mice given peptide intranasally an antigen-specific T-cell population is systemically activated similar to that induced by peptide immunization in incomplete Freund's adjuvant. Thus, a paradox exists whereby apparently similar peptide-specific populations are either associated with protection from, or exacerbation of, PIA. However, comparison of cytokine profiles reveals differences between these two cell populations. Peptide inhalation induces the production of Th1-type cytokines (interferon-gamma) whereas intraperitoneal immunization leads to the production of Th2-type cytokines (interleukin-4, interleukin-5 and interleukin-10) by splenic T cells upon stimulation with peptide. Thus, for the application of nasal 'tolerance' in clinical medicine, it is important to identify antigens and dosing regimes that counteract but do not activate adverse immune responses.