- Email: [email protected]
Design of engineered vaccines for systemic and mucosal immunity to HIV
Pathol Biol 2001 ; 49 : 466, 467
2001 Éditions scientifiques et médicales Elsevier SAS. Tous droits réservés S0369-8114(01)00167-5/FLA
Cellules dendritiques et infectiologie
Design of engineered vaccines for systemic and mucosal immunity to HIV J.A. Berzofsky Molecular Immunogenetics and Vaccine Research Section, Metabolism branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892, United States Summary To design vaccines for viruses such as HIV that do not elicit sufficient protective immunity, we first constructed cluster vaccines containing T helper, CTL and neutralizing antibody epitopes. For second generation vaccines, we increased responses by enhancing binding to Major Histocompatibility molecules or by incorporating cytokines. We found that high avidity CTL induce better viral clearance. We also induced anti-HIV mucosal T cell immunity by intrarectal administration. Such approaches may improve classic attenuated or killed pathogen vaccines. 2001 Éditions scientifiques et médicales Elsevier SAS HIV / mucosal immunity / peptides / T lymphocytes / vaccine
Résumé – Conception de vaccins pour une immunité systémique et muqueuse contre le VIH. Pour concevoir des vaccins contre des virus comme le VIH qui n’induisent pas d’immunité stérilisante, nous avons d’abord construit des vaccins contenant des épitopes auxiliaires, cytotoxiques et de neutralisation par anticorps. Pour une deuxième génération, nous avons augmenté les réponses en augmentant la liaison aux molécules d’histocompatibilité ou en incorporant des cytokines. Nous avons montré que des CTL de forte avidité induisent une meilleure stérilisation virale. Nous avons aussi induit une immunité muqueuse cellulaire T anti-VIH par administration rectale. De telles approches devraient apporter des améliorations par rapport aux vaccins classiques à base de pathogènes tués ou inactivés. 2001 Éditions scientifiques et médicales Elsevier SAS immunité muqueuse / lymphocytes T / peptides / vaccin / VIH
For viruses such as HIV that do not themselves elicit sufficient protective immunity, a vaccine may have to be more effective than natural infection. We have explored approaches to make more effective engineered vaccines. We have identified broadly recognized epitopes and tried to exclude epitopes that induce enhancing antibodies or autoimmune reactions that may contribute to immunodeficiency. The selected sequences included multideterminant regions of the HIV envelope protein sequence containing overlapping helper epitopes, from which we produced ‘cluster peptides’ spanning a series such overlapping epitopes to induce a helper T cell response in a broader population. We showed that an optimal cytotoxic T lymphocyte (CTL) response required association of a helper epitope with the CTL epitope, as had been known previously for B cell responses. Based on these principles, first generation vaccines were constructed by synthesizing peptides containing selected cluster T helper determinants, CTL epitopes and neutralizing antibody epitopes.
We then attempted to improve on the first generation synthetic peptide vaccines by several approaches. One approach for improving vaccine structures came out of our observations that adverse interactions play a major role in MHC binding and T-cell recognition, that Tcell receptors can distinguish broad classes of amino acids such as aromatic versus aliphatic, and that modifying the structure of an epitope can produce more potent immunogens and induce more broadly crossreactive CTL. We called this process of modifying the amino acid sequence to make a more effective vaccine ‘epitope enhancement’, and demonstrated proof of principle both for a peptide presented by a murine class II MHC molecule to helper T cells and for a peptide presented by a human class I molecule to CTL. First, we showed that modifying a helper epitope peptide of HIV to increase binding to the class II MHC molecule increased immunogenicity for a T cell proliferative response. When this enhanced helper epitope was coupled to a CTL epitope, the resulting vac-
467
Design of engineered vaccines
cine construct was more effective at inducing CTL against the unmodified CTL epitope than the original unmodified vaccine construct. The enhancement of class I-restricted CTL induction was mapped to the class II MHC molecule using congenic strains of mice. To test the applicability to a peptide presented by a human class I MHC molecule, we modified the sequence of an HLA-A2.1-binding epitope from the hepatitis C virus (HCV) core protein and showed that the enhanced epitope was more potent for recognition by human CTL from an infected patient, as well as murine CTL from HLA-A2.1-transgenic mice. To test immunogenicity in vivo, we immunized HLA-A2.1 transgenic mice with the wild type or enhanced epitope peptides and showed that the modified peptide induced higher levels of CTL than the wild type sequence, and thus was a more effective vaccine. A second approach focused on the role of CTL avidity in viral clearance. By adoptive transfer of CTL lines of varying avidity for the same HIV peptide-MHC complex into SCID mice, we found that high avidity CTL were more effective than low avidity CTL specific for the same HIV peptide-MHC complex at clearing infection in vivo by a vaccinia virus expressing the HIV-1 envelope protein gp160, even though both killed infected cells in vitro. Thus, quality of CTL was as important as quantity in protection against infection. We subsequently found that the mechanism explaining this difference relates to the ability of high avidity CTL to kill infected cells earlier in infection, before much viral progeny are produced. Thus, selective expansion of high avidity CTL for adoptive immunotherapy may be critical to success, and vaccines that selectively induce high avidity CTL may be more effective than ones that induce lower avidity CTL for the same target epitope. A third approach is the use of cytokines incorporated in the vaccine to amplify and steer the immune response. We first discovered in 1985 that IL-2 incorporated with antigen into adjuvants could be used to overcome low responsiveness. Therefore, more recently, we asked whether other cytokines could be incorporated to steer the response toward desired phenotypes, carrying out a matrix comparison of eight cytokines and seven immune responses. GM-CSF in the incomplete Freund’s adjuvant emulsion with a peptide antigen broadly enhanced most responses without altering the balance, whereas other cytokines such as IL-12 and IL-4 selectively steered the response toward Th1 or Th2 cytokines, and associated antibody isotypes. We found a synergy between IL-12 and
either GM-CSF or TNFα in increasing the CTL response, and recently showed that the three together are synergistic in increasing CTL and protection against virus challenge. A fourth approach is the induction of mucosal immunity to HIV-1. We found that intrarectal administration of a peptide vaccine, or a recombinant vaccinia vaccine, induced CTL in the mucosal lamina propria and Peyer’s patches as well as in the spleen, whereas subcutaneous administration induced CTL only in the spleen. Intrarectal was more effective than intragastric or intranasal administration. The intrarectal peptide vaccine protected mice for at least six months against mucosal transmission of a surrogate virus, recombinant vaccinia expressing HIV envelope protein, and the protection was shown to be dependent on CD8 cells, and to require the presence of CTL in the mucosal tissues. CTL in the spleen were not sufficient. The CTL level and the protection were enhanced by delivery of IL-12 mucosally with the vaccine. This was the first demonstration that protection against mucosal viral transmission required CTL to be present in the local mucosal site of transmission, and implies that for a vaccine to protect against a virus that is transmitted mucosally, such as HIV, it will be necessary to induce mucosal, not just systemic, CTL. We also took advantage of our finding that systemic immunization left the mucosal compartment naive, but mucosal immunization could induce systemic immunity, to circumvent the barrier of prior poxvirus immunity for the use of recombinant vaccinia vaccine vectors, by giving the vaccine mucosally. Thus, overall, our approach has been to design the building blocks of novel second generation vaccines that may improve on vaccine potency and efficacy, and allow selective induction of desired types of immune responses. One can potentially increase the magnitude of a response, broaden its specificity, steer the phenotype of the response, induce higher avidity effector cells, and target the response to the site of viral transmission. For pathogens that are not effective at inducing protective immunity themselves, such engineered vaccine approaches may improve on the classic methods of attenuated or killed pathogen vaccines.
REFERENCE 1 Berzofsky JA, Ahlers JD, Derby MA, Pendleton CD, Arichi T, Belyakov IM. Approaches to improve engineered vaccines for human immunodeficiency virus and other viruses that cause chronic infections. Immunol Rev 1999; 170: 151-72.
2001 Éditions scientifiques et médicales Elsevier SAS. Tous droits réservés S0369-8114(01)00167-5/FLA
Cellules dendritiques et infectiologie
Design of engineered vaccines for systemic and mucosal immunity to HIV J.A. Berzofsky Molecular Immunogenetics and Vaccine Research Section, Metabolism branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892, United States Summary To design vaccines for viruses such as HIV that do not elicit sufficient protective immunity, we first constructed cluster vaccines containing T helper, CTL and neutralizing antibody epitopes. For second generation vaccines, we increased responses by enhancing binding to Major Histocompatibility molecules or by incorporating cytokines. We found that high avidity CTL induce better viral clearance. We also induced anti-HIV mucosal T cell immunity by intrarectal administration. Such approaches may improve classic attenuated or killed pathogen vaccines. 2001 Éditions scientifiques et médicales Elsevier SAS HIV / mucosal immunity / peptides / T lymphocytes / vaccine
Résumé – Conception de vaccins pour une immunité systémique et muqueuse contre le VIH. Pour concevoir des vaccins contre des virus comme le VIH qui n’induisent pas d’immunité stérilisante, nous avons d’abord construit des vaccins contenant des épitopes auxiliaires, cytotoxiques et de neutralisation par anticorps. Pour une deuxième génération, nous avons augmenté les réponses en augmentant la liaison aux molécules d’histocompatibilité ou en incorporant des cytokines. Nous avons montré que des CTL de forte avidité induisent une meilleure stérilisation virale. Nous avons aussi induit une immunité muqueuse cellulaire T anti-VIH par administration rectale. De telles approches devraient apporter des améliorations par rapport aux vaccins classiques à base de pathogènes tués ou inactivés. 2001 Éditions scientifiques et médicales Elsevier SAS immunité muqueuse / lymphocytes T / peptides / vaccin / VIH
For viruses such as HIV that do not themselves elicit sufficient protective immunity, a vaccine may have to be more effective than natural infection. We have explored approaches to make more effective engineered vaccines. We have identified broadly recognized epitopes and tried to exclude epitopes that induce enhancing antibodies or autoimmune reactions that may contribute to immunodeficiency. The selected sequences included multideterminant regions of the HIV envelope protein sequence containing overlapping helper epitopes, from which we produced ‘cluster peptides’ spanning a series such overlapping epitopes to induce a helper T cell response in a broader population. We showed that an optimal cytotoxic T lymphocyte (CTL) response required association of a helper epitope with the CTL epitope, as had been known previously for B cell responses. Based on these principles, first generation vaccines were constructed by synthesizing peptides containing selected cluster T helper determinants, CTL epitopes and neutralizing antibody epitopes.
We then attempted to improve on the first generation synthetic peptide vaccines by several approaches. One approach for improving vaccine structures came out of our observations that adverse interactions play a major role in MHC binding and T-cell recognition, that Tcell receptors can distinguish broad classes of amino acids such as aromatic versus aliphatic, and that modifying the structure of an epitope can produce more potent immunogens and induce more broadly crossreactive CTL. We called this process of modifying the amino acid sequence to make a more effective vaccine ‘epitope enhancement’, and demonstrated proof of principle both for a peptide presented by a murine class II MHC molecule to helper T cells and for a peptide presented by a human class I molecule to CTL. First, we showed that modifying a helper epitope peptide of HIV to increase binding to the class II MHC molecule increased immunogenicity for a T cell proliferative response. When this enhanced helper epitope was coupled to a CTL epitope, the resulting vac-
467
Design of engineered vaccines
cine construct was more effective at inducing CTL against the unmodified CTL epitope than the original unmodified vaccine construct. The enhancement of class I-restricted CTL induction was mapped to the class II MHC molecule using congenic strains of mice. To test the applicability to a peptide presented by a human class I MHC molecule, we modified the sequence of an HLA-A2.1-binding epitope from the hepatitis C virus (HCV) core protein and showed that the enhanced epitope was more potent for recognition by human CTL from an infected patient, as well as murine CTL from HLA-A2.1-transgenic mice. To test immunogenicity in vivo, we immunized HLA-A2.1 transgenic mice with the wild type or enhanced epitope peptides and showed that the modified peptide induced higher levels of CTL than the wild type sequence, and thus was a more effective vaccine. A second approach focused on the role of CTL avidity in viral clearance. By adoptive transfer of CTL lines of varying avidity for the same HIV peptide-MHC complex into SCID mice, we found that high avidity CTL were more effective than low avidity CTL specific for the same HIV peptide-MHC complex at clearing infection in vivo by a vaccinia virus expressing the HIV-1 envelope protein gp160, even though both killed infected cells in vitro. Thus, quality of CTL was as important as quantity in protection against infection. We subsequently found that the mechanism explaining this difference relates to the ability of high avidity CTL to kill infected cells earlier in infection, before much viral progeny are produced. Thus, selective expansion of high avidity CTL for adoptive immunotherapy may be critical to success, and vaccines that selectively induce high avidity CTL may be more effective than ones that induce lower avidity CTL for the same target epitope. A third approach is the use of cytokines incorporated in the vaccine to amplify and steer the immune response. We first discovered in 1985 that IL-2 incorporated with antigen into adjuvants could be used to overcome low responsiveness. Therefore, more recently, we asked whether other cytokines could be incorporated to steer the response toward desired phenotypes, carrying out a matrix comparison of eight cytokines and seven immune responses. GM-CSF in the incomplete Freund’s adjuvant emulsion with a peptide antigen broadly enhanced most responses without altering the balance, whereas other cytokines such as IL-12 and IL-4 selectively steered the response toward Th1 or Th2 cytokines, and associated antibody isotypes. We found a synergy between IL-12 and
either GM-CSF or TNFα in increasing the CTL response, and recently showed that the three together are synergistic in increasing CTL and protection against virus challenge. A fourth approach is the induction of mucosal immunity to HIV-1. We found that intrarectal administration of a peptide vaccine, or a recombinant vaccinia vaccine, induced CTL in the mucosal lamina propria and Peyer’s patches as well as in the spleen, whereas subcutaneous administration induced CTL only in the spleen. Intrarectal was more effective than intragastric or intranasal administration. The intrarectal peptide vaccine protected mice for at least six months against mucosal transmission of a surrogate virus, recombinant vaccinia expressing HIV envelope protein, and the protection was shown to be dependent on CD8 cells, and to require the presence of CTL in the mucosal tissues. CTL in the spleen were not sufficient. The CTL level and the protection were enhanced by delivery of IL-12 mucosally with the vaccine. This was the first demonstration that protection against mucosal viral transmission required CTL to be present in the local mucosal site of transmission, and implies that for a vaccine to protect against a virus that is transmitted mucosally, such as HIV, it will be necessary to induce mucosal, not just systemic, CTL. We also took advantage of our finding that systemic immunization left the mucosal compartment naive, but mucosal immunization could induce systemic immunity, to circumvent the barrier of prior poxvirus immunity for the use of recombinant vaccinia vaccine vectors, by giving the vaccine mucosally. Thus, overall, our approach has been to design the building blocks of novel second generation vaccines that may improve on vaccine potency and efficacy, and allow selective induction of desired types of immune responses. One can potentially increase the magnitude of a response, broaden its specificity, steer the phenotype of the response, induce higher avidity effector cells, and target the response to the site of viral transmission. For pathogens that are not effective at inducing protective immunity themselves, such engineered vaccine approaches may improve on the classic methods of attenuated or killed pathogen vaccines.
REFERENCE 1 Berzofsky JA, Ahlers JD, Derby MA, Pendleton CD, Arichi T, Belyakov IM. Approaches to improve engineered vaccines for human immunodeficiency virus and other viruses that cause chronic infections. Immunol Rev 1999; 170: 151-72.