Tài liệu: Ai cần tớ giúp....

116. Lee KH, Wucherpfennig KW, Wiley DC. Structure of a human insulin peptide-HLA-DQ8 complex and susceptibility to type 1 diabetes. Nat Immunol 2001;2: 501-511.

117. Quartey-Papafio R, et al. Aspartate at position 57 of nonobese diabetic I-Ag7 beta-chain diminishes the spontaneous incidence of insulin-dependent diabetes mellitus. J Immunol 1995;154: 5567-5573.

118. Peterson DA, DiPaolo RJ, Kanagawa O, Unanue ER. Negative selection of immature thymocytes by a few peptide-MHC complexes: Differential sensitivity of immature and mature T cells. J Immunol 1999;162: 3117-3120.

119. Peterson DA, DiPaolo RJ, Kanagawa O, Unanue ER. Quantitative aspects of the T cell repertoire that escapes negative selection. Immunity 1999;11: 453-462.

120. Reay PA, Matsui K, Haase K, Wulfing C, Chien YH, David MM. Determination of the relationship between T cell responsiveness and the number of MHC-peptide complexes using specific monoclonal antibodies. J Immunol 2000;164: 5626-5634.

121. Harding CV, Unanue ER. Quantitation of antigen-presenting cell MHC class II/peptide complexes necessary for T-cell stimulation. Nature 1990;346: 574-576.

122. Kimachi K, Croft M, Grey HM. The minimal number of antigen-major histocompatibility complex class II complexes required for activation of naÃve and primed T cells. Eur J Immunol 1997;27: 3310-3321.

123. Lederberg J. Genes and antibodies. Science 1959;129: 1649-1653.

124. Pape KA, Khoruts A, Mondino A, Jenkins MK. Inflammatory cytokines enhance the in vivo clonal expansion and differentiation of antigen-activated CD4+ T cells. J Immunol 1997;159: 591-601.

125. DiPaolo RJ, Unanue ER. The level of peptide-MHC complex determines the susceptibility to autoimmune diabetes: studies in HEL transgenic mice. Eur J Immunol 2001;31: 3453-3459.

126. Hanahan D. Heritable formation of pancreatic beta-cell tumours in transgenic mice expressing recombinant insulin/simian virus 40 oncogenes. Nature 1985;315: 115-122.

127. Ohashi PS, et al. Ablation of 'tolerance' and induction of diabetes by virus infection in viral antigen transgenic mice. Cell 1991;65: 305-317.

128. Di Paolo RJ, Unanue ER. Chemical dominance does not relate to immunodominance: studies of the CD4+ T cell response to a model antigen. J Immunol 2002;169: 1-4.

129. Sercarz EE, Lehmann PV, Ametani A, Benichou G, Miller A, Moudgil K. Dominance and crypticity of T cell antigenic determinants. Annu Rev Immunol 1993;11: 729-766.

130. Busch DH, Pilip IM, Vijh S, Pamer EG. Coordinate regulation of complex T cell populations responding to bacterial infection. Immunity 1998;8: 353-360.

131. Viner NJ, Nelson CA, Unanue ER. Identification of a major I-Ek-restricted determinant of hen egg lysozyme: Limitations of lymph node proliferation studies in defining immunodominance and crypticity. Proc Natl Acad Sci USA 1995;92: 2214-2218.

132. Viner NJ, Nelson CA, Deck B, Unanue ER. Complexes generated by the binding of free peptides to class II MHC molecules are antigenically diverse compared with those generated by intracellular processing. J Immunol 1996;156: 2365-2368.

133. Cirrito TP, Pu Z, Deck MB, Unanue ER. Deamidation of asparagine in an MHC-bound peptide affects T cell recognition but does not explain type B reactivity. J Exp Med 2001;194: 1165-1169.

134. Pu Z, Carrero JA, Unanue ER. Distinct recognition by two subsets of T cells of an MHC-class-II-peptide complex. Proc Natl Acad Sci USA 2002;99: 8844-8849.

135. Unanue ER. Studies in Listeriosis show the strong symbiosis between the innate cellular system and the T cell response. Immunol Rev 1997;158: 11-25.

136. Janeway CA Jr, Medzhitov R. Innate immune recognition. Ann Rev Immunol 2002;20: 197-216.


Concay

116. Lee KH, Wucherpfennig KW, Wiley DC. Structure of a human insulin peptide-HLA-DQ8 complex and susceptibility to type 1 diabetes. Nat Immunol 2001;2: 501-511.

117. Quartey-Papafio R, et al. Aspartate at position 57 of nonobese diabetic I-Ag7 beta-chain diminishes the spontaneous incidence of insulin-dependent diabetes mellitus. J Immunol 1995;154: 5567-5573.

118. Peterson DA, DiPaolo RJ, Kanagawa O, Unanue ER. Negative selection of immature thymocytes by a few peptide-MHC complexes: Differential sensitivity of immature and mature T cells. J Immunol 1999;162: 3117-3120.

119. Peterson DA, DiPaolo RJ, Kanagawa O, Unanue ER. Quantitative aspects of the T cell repertoire that escapes negative selection. Immunity 1999;11: 453-462.

120. Reay PA, Matsui K, Haase K, Wulfing C, Chien YH, David MM. Determination of the relationship between T cell responsiveness and the number of MHC-peptide complexes using specific monoclonal antibodies. J Immunol 2000;164: 5626-5634.

121. Harding CV, Unanue ER. Quantitation of antigen-presenting cell MHC class II/peptide complexes necessary for T-cell stimulation. Nature 1990;346: 574-576.

122. Kimachi K, Croft M, Grey HM. The minimal number of antigen-major histocompatibility complex class II complexes required for activation of naÃve and primed T cells. Eur J Immunol 1997;27: 3310-3321.

123. Lederberg J. Genes and antibodies. Science 1959;129: 1649-1653.

124. Pape KA, Khoruts A, Mondino A, Jenkins MK. Inflammatory cytokines enhance the in vivo clonal expansion and differentiation of antigen-activated CD4+ T cells. J Immunol 1997;159: 591-601.

125. DiPaolo RJ, Unanue ER. The level of peptide-MHC complex determines the susceptibility to autoimmune diabetes: studies in HEL transgenic mice. Eur J Immunol 2001;31: 3453-3459.

126. Hanahan D. Heritable formation of pancreatic beta-cell tumours in transgenic mice expressing recombinant insulin/simian virus 40 oncogenes. Nature 1985;315: 115-122.

127. Ohashi PS, et al. Ablation of 'tolerance' and induction of diabetes by virus infection in viral antigen transgenic mice. Cell 1991;65: 305-317.

128. Di Paolo RJ, Unanue ER. Chemical dominance does not relate to immunodominance: studies of the CD4+ T cell response to a model antigen. J Immunol 2002;169: 1-4.

129. Sercarz EE, Lehmann PV, Ametani A, Benichou G, Miller A, Moudgil K. Dominance and crypticity of T cell antigenic determinants. Annu Rev Immunol 1993;11: 729-766.

130. Busch DH, Pilip IM, Vijh S, Pamer EG. Coordinate regulation of complex T cell populations responding to bacterial infection. Immunity 1998;8: 353-360.

131. Viner NJ, Nelson CA, Unanue ER. Identification of a major I-Ek-restricted determinant of hen egg lysozyme: Limitations of lymph node proliferation studies in defining immunodominance and crypticity. Proc Natl Acad Sci USA 1995;92: 2214-2218.

132. Viner NJ, Nelson CA, Deck B, Unanue ER. Complexes generated by the binding of free peptides to class II MHC molecules are antigenically diverse compared with those generated by intracellular processing. J Immunol 1996;156: 2365-2368.

133. Cirrito TP, Pu Z, Deck MB, Unanue ER. Deamidation of asparagine in an MHC-bound peptide affects T cell recognition but does not explain type B reactivity. J Exp Med 2001;194: 1165-1169.

134. Pu Z, Carrero JA, Unanue ER. Distinct recognition by two subsets of T cells of an MHC-class-II-peptide complex. Proc Natl Acad Sci USA 2002;99: 8844-8849.

135. Unanue ER. Studies in Listeriosis show the strong symbiosis between the innate cellular system and the T cell response. Immunol Rev 1997;158: 11-25.

136. Janeway CA Jr, Medzhitov R. Innate immune recognition. Ann Rev Immunol 2002;20: 197-216.


Concay

coi như tui gửi tăng bạn 5 bài review tương đối mới về antigene presentation, lưu ý là có nhiếu trường phái khác nhau, nên một số quan niệm của họ không được thống nhất.
bạn không nên bận tâm là các bài này lấy từ tạp chí nào, vì bạn có hỏi tui cũng không trả lời (đó là nguyên tắc).
chúc bạn tìm được nhiều thông tin bổ ích.
Thân
Concay

coi như tui gửi tăng bạn 5 bài review tương đối mới về antigene presentation, lưu ý là có nhiếu trường phái khác nhau, nên một số quan niệm của họ không được thống nhất.
bạn không nên bận tâm là các bài này lấy từ tạp chí nào, vì bạn có hỏi tui cũng không trả lời (đó là nguyên tắc).
chúc bạn tìm được nhiều thông tin bổ ích.
Thân
Concay

Peptide vaccines against hepatitis B virus: from animal model to human studies
Olivier B. Englera, Wen Juan Daia, Alessandro Settee, Isabelle P. Hunzikerb, Jẳrg Reichenc, Werner J. Pichlera and Andreas Cerny, , d
a Clinic of Rheumatology and Clinical Immunology/Allergology, University Hospital, Inselspital Bern, 3010 Bern, Switzerland
b Department of Neuropharmacology, The Scripps Research Institute, La Jolla, CA 92037, USA
c Institute of Clinical Pharmacology, University of Bern, 3010 Bern, Switzerland
d Clinica Medica, Ospedale Civico, 6903 Lugano, Switzerland
e Epimmune Inc., San Diego, CA 92121, USA
Available online 5 December 2001.
Volume 38, Issue 6 , December 2001, Pages 457-465 cỏằĐa mỏằTt tỏĂp chư khoa hoc nào 'ó
Abstract
An estimated 400 million people are chronically infected with the hepatitis B virus (HBV). Chronic viral hepatitis infection incurs serious sequelae such as liver cirrhosis and hepatocellular carcinoma. Prevention and treatment, thus, represent an important target for public health. Preventive vaccines using HBsAg alone or combined with other antigens allow for the generation of neutralizing antibodies which effectively prevent infection in immunocompetent individuals. Cell-mediated immunological mechanisms are thought to be crucial in determining viral persistence or viral elimination. Therapeutic approaches aiming to shift cellular immunity towards viral elimination have been on the research agenda for many years. This paper summarizes pre-clinical and clinical results obtained with the use of immunogenic peptides formulated as vaccines to selectively boost cellular immune responses. Such vaccines are capable of generating cellular immune responses in animal models as well as in humans and represent an important step towards the development of a therapeutic vaccine against chronic hepatitis.
Author Keywords: Adjuvants; Hepatitis B virus; Peptide vaccines

Article Outline
1. Introduction
2. Studies in animal models
2.1. CTL epitope
2.2. T helper epitope support CTL induction and maintenance
2.3. Universal T helper epitopes
2.4. Adjuvants for CTL induction
2.5. Liposomes and virosomes
2.6. Cytokines that support CTL induction
2.7. Microbial components that support priming of CTL
2.8. Human studies
3. Conclusions
Acknowledgements
References
1. Introduction
Hepatitis B virus (HBV) infection is caused by a small enveloped DNA virus that infects the liver, causing immune mediated hepatocellular necrosis and inflammation (Tiollais et al., 1985). The infection can be either acute or chronic. Clinical severity may range from (a) asymptomatic and completely resolving to (b) symptomatic with progressive and even fatal illness or (c) occasional fulminant hepatic failure. The course of the infection appears to be determined by the host's immune response. In most immunocompetent adults, acute infection leads to an acute hepatitis followed by rapid clearance of the virus and the development of lifelong immunity. If, however, the infection occurs in the neonatal period or in the first years of life, infection with HBV usually becomes persistent. Virus clearance crucially depends on the activation of appropriate helper T cells which secrete antiviral and cytolytic cytokines and also, most importantly, on the presence of activated peptide-specific CD8+ cytotoxic T lymphocytes (CTL). Through their T cell receptors (TCR) these CTL specifically recognize viral peptides that are presented by specific MHC class I molecules and cause lysis of the infected cells ( Chisari and Ferrari, 1995). Biologically relevant epitope sequences have been identified for animal and humans. This review summarizes efforts to design biologically active vaccines making use of the peptide principle, which allows induction of highly specific cellular immune responses against the HBV.
2. Studies in animal models
CD8+ T lymphocytes are an important component of host defense mechanisms responsible for successful HBV clearance. In patients with acute hepatitis B, multispecific CTL responses appear to be associated with spontaneous resolution of acute infection. An HBV transgenic mouse model showed that the transfer of activated CTL can lead to the elimination of expression of HBV antigen (Guidotti et al., 1996). The fact that virus-specific CTL are capable of both preventing the establishment of chronic viral infection and eliminating the infection once it has become established was demonstrated in a murine lymphocytic choriomeningitis virus model ( Byrne and Oldstone, 1984).
Induction of CTL response can be achieved in vivo by various means including priming with replicating virus or recombinant DNA. But immunization with live virus may cause disease in immunosuppressed individuals and there are strong doubts about the use of DNA for human application. Means have been sought for many years to induce virus-specific CTL responses in vivo with non-replicating synthetic virus antigens such as peptides (Aichele and Goa)
2.1. CTL epitope
The shortest possible amino acid sequences, consisting only of the CTL epitope with the ability to bind without further processing to the MHC class I molecule, are termed minimal CTL epitopes. The identification of new CTL epitopes in a protein can be done by screening libraries of synthetic peptides covering the whole protein sequence for good MHC class I binding characteristics (Vitiello et al., 1997). Alternatively putative MHC class I binding peptides can be predicted from sequence analysis on the basis of known binding motifs for determined MHC alleles. A different approach is to recover MHC class I bound peptides by the acid elution technique from cell lines stable or transiently expressing the protein of interest ( Rảtzschke et al., 1990). In a second step, the epitopes are analyzed for their immunogenicity by immunizing animals with the minimal CTL epitope or with the whole protein. Thereby, different categories of peptides have been identified: the non-epitopes which consist of peptides that are not able to bind to the MHC class I molecules, cryptic epitopes, subdominant and dominant epitopes. An epitope is considered to be cryptic if a response can be elicited by peptide immunization, but is not detected when intact whole protein is used as an antigen. An explanation for crypticity could be that the epitope may not be generated in the course of antigen processing or may not be transported to the cell surface. In this case, cryptic epitopes would not be useful for vaccine development as they would not be presented on the MHC class I molecules of a virus infected cell. In contrast, the subdominant and the dominant epitopes are both processed from whole protein and presented on the cell surface and are, therefore, valuable targets for CTL stimulation during a viral infection. Dominant epitopes are presented at high levels on the surface of a cell that expresses the whole protein and they efficiently prime CTL when presented in an immunogenic context. In contrast, the subdominant epitopes are found at much lower levels on an APC and poorly prime CTL ( Vitiello et al., 1997). Exogenously added peptide can bind directly to the MHC class I molecules present on the cell surface of an APC without the need of further processing ( Widmann et al., 1991) and can, therefore, efficiently be presented to TCR of CTL specific for subdominant epitopes. Chronic HBV infection is known to be associated with T cell hyporesponsiveness and tolerance. It is possible that subdominant epitopes might be less effective tolerogens because they are naturally expressed at lower levels in hepatocytes. Targeting subdominant epitopes may be an effective way to overcome CTL tolerance ( Sette et al., 2001) In ad***ion, the peptide based immunization allows the immune response to focus against highly conserved epitopes ( Loing et al., 2000).
The final repertoire of CTL specificities is significantly determined by MHCõ?"peptide binding capacity, stability of MHCõ?"peptide complex, availability of specific TCR, and antigen processing (Vitiello et al., 1997). For the priming of CTL and lysis of target cells, the number of peptides expressed on MHC class I is relevant ( Mottez et al., 1995). To determine the immunodominance of a peptide, MHC binding affinity plays the central role. By analyzing HLA-A2 binding peptides it was found that the majority of immunogenic peptides corresponded to those with binding affinities of 50 nM or less. Peptides that bound below the 500 nM affinity threshold never elicited reproducible CTL responses (Sette et al., 1994). Increasing the affinity to the MHC class I molecule of several low affinity HLA-A2-restricted peptides by adapting the anchor amino acid produced more stable complexes. Thus, peptides were presented at sufficiently high levels to induce CTL without altering the specificity of the induced CTL response ( Sarobe et al., 1998).
Another reason for the inability of some peptides to induce a response in animals could be due to their rapid degradation by proteases and peptidases in the extracellular fluid. In vitro studies have shown that the degradation of peptides in serum varies very strongly depending on the peptide sequence. For the most unstable peptides, antigenic potency decreased 10,000-fold after overnight incubation in FCS at 37 °C. Degradation was much less for most peptides and they seemed to be protected once bound to the MHC class I molecule. In vivo studies have shown that exogenously added peptides remained on the cell surface for more than 3 days (Widmann et al., 1991). An important problem for peptide-based vaccines is that the minimal CTL epitopes are not immunogenic per se and in general require the ad***ion of T helper epitopes and adjuvants to efficiently stimulate CTL.
2.2. T helper epitope support CTL induction and maintenance
The requirement of T cell help in CTL induction is still discussed and controversial (Hiranuma et al., 1999). Although much evidence for a helper requirement in CTL induction exists, there is also evidence for CTL response independent of T helper cells ( Shirai et al., 1994). Studies by Leclerc showed that for three different CTL epitopes from LCMV, adenovirus and vesicular stomatitis virus (VSV) using short optimal-length peptides in incomplete Freunds adjuvants (IFA) could stimulate a CTL response in the absence of CD4+ T cell help. The CD4 T cell independence was proven by using CD4 T cell depleted mice for immunization studies and minimal CD8 T cell epitopes that were shown not to stimulate proliferation in vitro ( Fayolle et al., 1996). Similar results were obtained by Dyall et al. (1995). It was hypothesized that optimal fit of these short peptides into the antigen binding groove of class I molecules increases the density of peptides presented. This may control the capacity of antigen-presenting cells to directly activate CD8+ T cells independently of CD4+ T cell help. If the required density threshold is not reached CD4+ T cells may be required to give appropriate signals to activate CD8+ T cells ( Fayolle et al., 1996).
Although certain CTL epitopes on their own can elicit a CTL response, repeated injections of the peptide are often necessary. They must also be spaced at optimal time intervals (Walden and Eisen, 1990). Immunization experiments with epitopes such as the ovalbumin peptide (SIINFEKL) indicate that Th epitopes are required to prime CTL for less immunogenic epitopes in vivo. In this experiment, a mix of T helper and CTL epitope was as effective as covalently bound T helperõ?"CTL (Hõ?"CTL) fusion peptides. The helper function was similar for an intrinsic T helper epitope derived from the same protein or an extrinsic T helper epitope, taken from an independent protein ( Borges et al., 1994). The notion that CD4 T helper cells are critical not only for priming but also for the maintenance of CTL and chronic control of viremia come from murine studies of specific strains of LCMV. In particular, under con***ions of high dose infection or infection of rapidly replicating and disseminating LCMV strains, CD4 cells are required to prevent CTL exhaustion. A critical role for CD4 cells has also been shown after immunization: vaccination of mice deficient in CD4 cells led to less efficient protection from subsequent viral challenge ( Von Herrath et al., 1996). A progressive loss of CTL in the absence of adequate helper function has also been demonstrated from murine -herpes virus ( Cardin et al., 1996). The evidence that a helper site is beneficial for CTL induction is quite strong ( Widmann and Lasarte), but it is still not clear whether the helper and CTL sites need to be linked. Immunization studies from Shirai et al. (1994) show that the immunodominant HIVõ?"CTL epitope P18 formulated in the adjuvant QS21 gave only marginal CTL induction and that the Th peptides from the HIV envelope protein gp160 had to be covalently attached to the CTL determinant on a single peptide for efficient priming of CTL in vivo. These results were confirmed in immunization experiments based on HBV- and HCV-related epitopes, where the requirement for covalent linkage was seen for the tetanus helper epitope coupled to the HBV core CTL epitope or for the HCV NS5 CTL epitope that was only immunogenic as a fusion peptide with a T helper epitope. Thereby an extrinsic T helper peptide stimulating a Th1 like cytokine pattern was much more efficient in providing help for CTL induction as helper peptides producing a Th0 like cytokine profile ( Lopez-Diaz de Cerio et al., 1999). The results indicate that covalent attachment of a Th determinant to the CTL determinant on a single peptide facilitates collaboration among APC and CD4+ T cell help for priming of CTL response in vivo. It was hypothesized that if the same APC presents both the CTL and Th epitope through MHC class I and class II, respectively CTL are more efficiently primed because it brings either the Th cell and the CTL precursor closer together for a more effective transmission of small amounts of lymphokines ( Gill and Matloubian) or because the APC is activated by the helper cell and then in turn is more effective at presenting the antigen to the CTL. It is thought that large peptides would bind to MHC class II expressing cells probably at the injection site or draining lymph nodes and are cleaved by cell surface proteases or by proteases in extracellular fluid so the same cell may present both the CTL and the Th cell epitopes through MHC class I and class II, respectively. The observation that in vitro CTL restimulation required only IL-2 ad***ion when the minimal epitope was used for restimulation and not when the combined Hõ?"CTL epitope was added supports the idea that IL-2 or another activation signal can be provided by helper cells ( Shirai et al., 1994).
In this context, results from previous studies, in which covalent linkage of helper and CTL epitope was not obligatory, could be explained through physical association of the peptides of adjuvant emulsion oil microdroplets or that application of multiple high doses of peptide was able to overcome the inherent disadvantages of the unlinked mixture (Shirai et al., 1995).
2.3. Universal T helper epitopes
MHC class II molecules show a high degree of polymorphism. In this context, the development of a universal T helper epitope with broad MHC class II binding specificity as described by Alexander et al. (1994) is of special interest. The so called universal DR binding peptides hemagglutinin (HA) 307-319, HBVnc50-69, tetanus toxoid (TT) 830-843 have been found to bind to several DR types, but they were by no means completely cross-reactive in their DR binding capacity. Therefore, totally synthetic peptides were designed to bind a broad DR repertoire. PanDR-binding peptides were originally developed by introducing the main DR anchor residues necessary for binding to a representative set of common DR molecules into a polyalanine backbone. PanDR peptides bound to all ten of ten DR molecules tested with affinities mostly in the nanomolar range. To engineer powerful immunogens, bulky and charged residues were introduced at crucial peptide positions, which indeed yielded an extremely powerful immunogen denoted PADRE. These peptides were up to 1000-fold more powerful than natural T cell epitopes in their capacity to deliver help during induction of antigen-specific CTL response in mice ( Alexander et al., 1994).
Large carrier proteins are powerful sources of helper epitopes, but suffer from several disadvantages such as a relatively high production cost and that they do not allow for focusing on certain epitopes. Specifically in the case of HBV-S subunit vaccine, it has been shown that laboratory mice can be classified as responders and non-responders as a function of their genetic background at the MHC class II locus, indicating a deficient HTL function of these mice in recognition of the HBV-S antigen. Improved HTL function of these mice would increase the CTL induction. The use of a non-HBV-derived T helper epitope in a therapeutic HBV vaccine may have two advantages. First, the crucial therapeutic variable is the induction of HBV-specific CTL response. Second, a potential T cell tolerance at the level of class II HBV epitopes could be circumvented. In an HBV transgenic mouse model where PADRE was incorporated into a lipopeptide construct named Theradigm, it was observed that tolerance at the CTL level was broken by a PADREõ?"CTL lipopeptide, but not by similar constructs containing a conventional T helper epitope (Alexander et al., 1998).
Extrinsic T cell help is a good choice to use in vaccines against infections where the CD4+ T cells are not directly involved in controlling viral infection by producing cytokines such as IFN-, IL-2, IL-4, and others or by direct interaction. In certain infections, disease related epitopes are required such as for chronic HIV and HCV infection where positive clinical outcomes have been associated with the recognition of particular helper T lymphocyte (HTL) epitopes (Alexander et al., 1998).
Concay
ĐặỏằÊc Milou sỏằưa vào 06:34 ngày 18/06/2003

Peptide vaccines against hepatitis B virus: from animal model to human studies
Olivier B. Englera, Wen Juan Daia, Alessandro Settee, Isabelle P. Hunzikerb, Jẳrg Reichenc, Werner J. Pichlera and Andreas Cerny, , d
a Clinic of Rheumatology and Clinical Immunology/Allergology, University Hospital, Inselspital Bern, 3010 Bern, Switzerland
b Department of Neuropharmacology, The Scripps Research Institute, La Jolla, CA 92037, USA
c Institute of Clinical Pharmacology, University of Bern, 3010 Bern, Switzerland
d Clinica Medica, Ospedale Civico, 6903 Lugano, Switzerland
e Epimmune Inc., San Diego, CA 92121, USA
Available online 5 December 2001.
Volume 38, Issue 6 , December 2001, Pages 457-465 cỏằĐa mỏằTt tỏĂp chư khoa hoc nào 'ó
Abstract
An estimated 400 million people are chronically infected with the hepatitis B virus (HBV). Chronic viral hepatitis infection incurs serious sequelae such as liver cirrhosis and hepatocellular carcinoma. Prevention and treatment, thus, represent an important target for public health. Preventive vaccines using HBsAg alone or combined with other antigens allow for the generation of neutralizing antibodies which effectively prevent infection in immunocompetent individuals. Cell-mediated immunological mechanisms are thought to be crucial in determining viral persistence or viral elimination. Therapeutic approaches aiming to shift cellular immunity towards viral elimination have been on the research agenda for many years. This paper summarizes pre-clinical and clinical results obtained with the use of immunogenic peptides formulated as vaccines to selectively boost cellular immune responses. Such vaccines are capable of generating cellular immune responses in animal models as well as in humans and represent an important step towards the development of a therapeutic vaccine against chronic hepatitis.
Author Keywords: Adjuvants; Hepatitis B virus; Peptide vaccines

Article Outline
1. Introduction
2. Studies in animal models
2.1. CTL epitope
2.2. T helper epitope support CTL induction and maintenance
2.3. Universal T helper epitopes
2.4. Adjuvants for CTL induction
2.5. Liposomes and virosomes
2.6. Cytokines that support CTL induction
2.7. Microbial components that support priming of CTL
2.8. Human studies
3. Conclusions
Acknowledgements
References
1. Introduction
Hepatitis B virus (HBV) infection is caused by a small enveloped DNA virus that infects the liver, causing immune mediated hepatocellular necrosis and inflammation (Tiollais et al., 1985). The infection can be either acute or chronic. Clinical severity may range from (a) asymptomatic and completely resolving to (b) symptomatic with progressive and even fatal illness or (c) occasional fulminant hepatic failure. The course of the infection appears to be determined by the host's immune response. In most immunocompetent adults, acute infection leads to an acute hepatitis followed by rapid clearance of the virus and the development of lifelong immunity. If, however, the infection occurs in the neonatal period or in the first years of life, infection with HBV usually becomes persistent. Virus clearance crucially depends on the activation of appropriate helper T cells which secrete antiviral and cytolytic cytokines and also, most importantly, on the presence of activated peptide-specific CD8+ cytotoxic T lymphocytes (CTL). Through their T cell receptors (TCR) these CTL specifically recognize viral peptides that are presented by specific MHC class I molecules and cause lysis of the infected cells ( Chisari and Ferrari, 1995). Biologically relevant epitope sequences have been identified for animal and humans. This review summarizes efforts to design biologically active vaccines making use of the peptide principle, which allows induction of highly specific cellular immune responses against the HBV.
2. Studies in animal models
CD8+ T lymphocytes are an important component of host defense mechanisms responsible for successful HBV clearance. In patients with acute hepatitis B, multispecific CTL responses appear to be associated with spontaneous resolution of acute infection. An HBV transgenic mouse model showed that the transfer of activated CTL can lead to the elimination of expression of HBV antigen (Guidotti et al., 1996). The fact that virus-specific CTL are capable of both preventing the establishment of chronic viral infection and eliminating the infection once it has become established was demonstrated in a murine lymphocytic choriomeningitis virus model ( Byrne and Oldstone, 1984).
Induction of CTL response can be achieved in vivo by various means including priming with replicating virus or recombinant DNA. But immunization with live virus may cause disease in immunosuppressed individuals and there are strong doubts about the use of DNA for human application. Means have been sought for many years to induce virus-specific CTL responses in vivo with non-replicating synthetic virus antigens such as peptides (Aichele and Goa)
2.1. CTL epitope
The shortest possible amino acid sequences, consisting only of the CTL epitope with the ability to bind without further processing to the MHC class I molecule, are termed minimal CTL epitopes. The identification of new CTL epitopes in a protein can be done by screening libraries of synthetic peptides covering the whole protein sequence for good MHC class I binding characteristics (Vitiello et al., 1997). Alternatively putative MHC class I binding peptides can be predicted from sequence analysis on the basis of known binding motifs for determined MHC alleles. A different approach is to recover MHC class I bound peptides by the acid elution technique from cell lines stable or transiently expressing the protein of interest ( Rảtzschke et al., 1990). In a second step, the epitopes are analyzed for their immunogenicity by immunizing animals with the minimal CTL epitope or with the whole protein. Thereby, different categories of peptides have been identified: the non-epitopes which consist of peptides that are not able to bind to the MHC class I molecules, cryptic epitopes, subdominant and dominant epitopes. An epitope is considered to be cryptic if a response can be elicited by peptide immunization, but is not detected when intact whole protein is used as an antigen. An explanation for crypticity could be that the epitope may not be generated in the course of antigen processing or may not be transported to the cell surface. In this case, cryptic epitopes would not be useful for vaccine development as they would not be presented on the MHC class I molecules of a virus infected cell. In contrast, the subdominant and the dominant epitopes are both processed from whole protein and presented on the cell surface and are, therefore, valuable targets for CTL stimulation during a viral infection. Dominant epitopes are presented at high levels on the surface of a cell that expresses the whole protein and they efficiently prime CTL when presented in an immunogenic context. In contrast, the subdominant epitopes are found at much lower levels on an APC and poorly prime CTL ( Vitiello et al., 1997). Exogenously added peptide can bind directly to the MHC class I molecules present on the cell surface of an APC without the need of further processing ( Widmann et al., 1991) and can, therefore, efficiently be presented to TCR of CTL specific for subdominant epitopes. Chronic HBV infection is known to be associated with T cell hyporesponsiveness and tolerance. It is possible that subdominant epitopes might be less effective tolerogens because they are naturally expressed at lower levels in hepatocytes. Targeting subdominant epitopes may be an effective way to overcome CTL tolerance ( Sette et al., 2001) In ad***ion, the peptide based immunization allows the immune response to focus against highly conserved epitopes ( Loing et al., 2000).
The final repertoire of CTL specificities is significantly determined by MHCõ?"peptide binding capacity, stability of MHCõ?"peptide complex, availability of specific TCR, and antigen processing (Vitiello et al., 1997). For the priming of CTL and lysis of target cells, the number of peptides expressed on MHC class I is relevant ( Mottez et al., 1995). To determine the immunodominance of a peptide, MHC binding affinity plays the central role. By analyzing HLA-A2 binding peptides it was found that the majority of immunogenic peptides corresponded to those with binding affinities of 50 nM or less. Peptides that bound below the 500 nM affinity threshold never elicited reproducible CTL responses (Sette et al., 1994). Increasing the affinity to the MHC class I molecule of several low affinity HLA-A2-restricted peptides by adapting the anchor amino acid produced more stable complexes. Thus, peptides were presented at sufficiently high levels to induce CTL without altering the specificity of the induced CTL response ( Sarobe et al., 1998).
Another reason for the inability of some peptides to induce a response in animals could be due to their rapid degradation by proteases and peptidases in the extracellular fluid. In vitro studies have shown that the degradation of peptides in serum varies very strongly depending on the peptide sequence. For the most unstable peptides, antigenic potency decreased 10,000-fold after overnight incubation in FCS at 37 °C. Degradation was much less for most peptides and they seemed to be protected once bound to the MHC class I molecule. In vivo studies have shown that exogenously added peptides remained on the cell surface for more than 3 days (Widmann et al., 1991). An important problem for peptide-based vaccines is that the minimal CTL epitopes are not immunogenic per se and in general require the ad***ion of T helper epitopes and adjuvants to efficiently stimulate CTL.
2.2. T helper epitope support CTL induction and maintenance
The requirement of T cell help in CTL induction is still discussed and controversial (Hiranuma et al., 1999). Although much evidence for a helper requirement in CTL induction exists, there is also evidence for CTL response independent of T helper cells ( Shirai et al., 1994). Studies by Leclerc showed that for three different CTL epitopes from LCMV, adenovirus and vesicular stomatitis virus (VSV) using short optimal-length peptides in incomplete Freunds adjuvants (IFA) could stimulate a CTL response in the absence of CD4+ T cell help. The CD4 T cell independence was proven by using CD4 T cell depleted mice for immunization studies and minimal CD8 T cell epitopes that were shown not to stimulate proliferation in vitro ( Fayolle et al., 1996). Similar results were obtained by Dyall et al. (1995). It was hypothesized that optimal fit of these short peptides into the antigen binding groove of class I molecules increases the density of peptides presented. This may control the capacity of antigen-presenting cells to directly activate CD8+ T cells independently of CD4+ T cell help. If the required density threshold is not reached CD4+ T cells may be required to give appropriate signals to activate CD8+ T cells ( Fayolle et al., 1996).
Although certain CTL epitopes on their own can elicit a CTL response, repeated injections of the peptide are often necessary. They must also be spaced at optimal time intervals (Walden and Eisen, 1990). Immunization experiments with epitopes such as the ovalbumin peptide (SIINFEKL) indicate that Th epitopes are required to prime CTL for less immunogenic epitopes in vivo. In this experiment, a mix of T helper and CTL epitope was as effective as covalently bound T helperõ?"CTL (Hõ?"CTL) fusion peptides. The helper function was similar for an intrinsic T helper epitope derived from the same protein or an extrinsic T helper epitope, taken from an independent protein ( Borges et al., 1994). The notion that CD4 T helper cells are critical not only for priming but also for the maintenance of CTL and chronic control of viremia come from murine studies of specific strains of LCMV. In particular, under con***ions of high dose infection or infection of rapidly replicating and disseminating LCMV strains, CD4 cells are required to prevent CTL exhaustion. A critical role for CD4 cells has also been shown after immunization: vaccination of mice deficient in CD4 cells led to less efficient protection from subsequent viral challenge ( Von Herrath et al., 1996). A progressive loss of CTL in the absence of adequate helper function has also been demonstrated from murine -herpes virus ( Cardin et al., 1996). The evidence that a helper site is beneficial for CTL induction is quite strong ( Widmann and Lasarte), but it is still not clear whether the helper and CTL sites need to be linked. Immunization studies from Shirai et al. (1994) show that the immunodominant HIVõ?"CTL epitope P18 formulated in the adjuvant QS21 gave only marginal CTL induction and that the Th peptides from the HIV envelope protein gp160 had to be covalently attached to the CTL determinant on a single peptide for efficient priming of CTL in vivo. These results were confirmed in immunization experiments based on HBV- and HCV-related epitopes, where the requirement for covalent linkage was seen for the tetanus helper epitope coupled to the HBV core CTL epitope or for the HCV NS5 CTL epitope that was only immunogenic as a fusion peptide with a T helper epitope. Thereby an extrinsic T helper peptide stimulating a Th1 like cytokine pattern was much more efficient in providing help for CTL induction as helper peptides producing a Th0 like cytokine profile ( Lopez-Diaz de Cerio et al., 1999). The results indicate that covalent attachment of a Th determinant to the CTL determinant on a single peptide facilitates collaboration among APC and CD4+ T cell help for priming of CTL response in vivo. It was hypothesized that if the same APC presents both the CTL and Th epitope through MHC class I and class II, respectively CTL are more efficiently primed because it brings either the Th cell and the CTL precursor closer together for a more effective transmission of small amounts of lymphokines ( Gill and Matloubian) or because the APC is activated by the helper cell and then in turn is more effective at presenting the antigen to the CTL. It is thought that large peptides would bind to MHC class II expressing cells probably at the injection site or draining lymph nodes and are cleaved by cell surface proteases or by proteases in extracellular fluid so the same cell may present both the CTL and the Th cell epitopes through MHC class I and class II, respectively. The observation that in vitro CTL restimulation required only IL-2 ad***ion when the minimal epitope was used for restimulation and not when the combined Hõ?"CTL epitope was added supports the idea that IL-2 or another activation signal can be provided by helper cells ( Shirai et al., 1994).
In this context, results from previous studies, in which covalent linkage of helper and CTL epitope was not obligatory, could be explained through physical association of the peptides of adjuvant emulsion oil microdroplets or that application of multiple high doses of peptide was able to overcome the inherent disadvantages of the unlinked mixture (Shirai et al., 1995).
2.3. Universal T helper epitopes
MHC class II molecules show a high degree of polymorphism. In this context, the development of a universal T helper epitope with broad MHC class II binding specificity as described by Alexander et al. (1994) is of special interest. The so called universal DR binding peptides hemagglutinin (HA) 307-319, HBVnc50-69, tetanus toxoid (TT) 830-843 have been found to bind to several DR types, but they were by no means completely cross-reactive in their DR binding capacity. Therefore, totally synthetic peptides were designed to bind a broad DR repertoire. PanDR-binding peptides were originally developed by introducing the main DR anchor residues necessary for binding to a representative set of common DR molecules into a polyalanine backbone. PanDR peptides bound to all ten of ten DR molecules tested with affinities mostly in the nanomolar range. To engineer powerful immunogens, bulky and charged residues were introduced at crucial peptide positions, which indeed yielded an extremely powerful immunogen denoted PADRE. These peptides were up to 1000-fold more powerful than natural T cell epitopes in their capacity to deliver help during induction of antigen-specific CTL response in mice ( Alexander et al., 1994).
Large carrier proteins are powerful sources of helper epitopes, but suffer from several disadvantages such as a relatively high production cost and that they do not allow for focusing on certain epitopes. Specifically in the case of HBV-S subunit vaccine, it has been shown that laboratory mice can be classified as responders and non-responders as a function of their genetic background at the MHC class II locus, indicating a deficient HTL function of these mice in recognition of the HBV-S antigen. Improved HTL function of these mice would increase the CTL induction. The use of a non-HBV-derived T helper epitope in a therapeutic HBV vaccine may have two advantages. First, the crucial therapeutic variable is the induction of HBV-specific CTL response. Second, a potential T cell tolerance at the level of class II HBV epitopes could be circumvented. In an HBV transgenic mouse model where PADRE was incorporated into a lipopeptide construct named Theradigm, it was observed that tolerance at the CTL level was broken by a PADREõ?"CTL lipopeptide, but not by similar constructs containing a conventional T helper epitope (Alexander et al., 1998).
Extrinsic T cell help is a good choice to use in vaccines against infections where the CD4+ T cells are not directly involved in controlling viral infection by producing cytokines such as IFN-, IL-2, IL-4, and others or by direct interaction. In certain infections, disease related epitopes are required such as for chronic HIV and HCV infection where positive clinical outcomes have been associated with the recognition of particular helper T lymphocyte (HTL) epitopes (Alexander et al., 1998).
Concay
ĐặỏằÊc Milou sỏằưa vào 06:34 ngày 18/06/2003

2.4. Adjuvants for CTL induction
It has been shown that CTL induction is possible in vivo with chemically modified peptides linked to lipid components. The first lipopeptides used were peptides conjugated to tripalmitoyl-S-glycerylcysteinyl-seryl-serine (P3CSS) synthesized according to the immunologically active N-terminal sequence of the principle lipoprotein of Escherichia coli (Deres et al., 1998). The lipopeptide used in later experiments may only have one or two palmitoyl groups. Experiments on lipidated peptides in the HLA-A2 transgenic mouse model showed that induction of CTL responses was crucially dependent on the presence of HTL and most efficient in the case of lipidated covalently linked HTLâ?"CTL epitope constructs. These last were considerably more immunogenic than mixtures of unlipidated HTLâ?"CTL epitope emulsified in IFA. The magnitude of the CTL response was similar to the naturally occurring response associated with clearance of acute hepatitis B ( Oseroff and Deres). High CTL responses compatible with protection were induced by immunizing primates and humans with lipopeptides derived from SIV and HBV sequences ( Bourgault-Villada et al., 1997). The lipid seems to serve three different roles: (a) it leads to the persistence of peptides at the site of injection or in the draining lymph node for sufficient time to induce a CTL response; (b) lipid modification may help to translocate peptides across the plasma membrane into the cytoplasm; and c) the lipid moiety may also have intrinsic adjuvant activity ( Vitiello et al., 1995). Local concentration of antigen in lymphoid organs seems to be particularly important for peptide immunizations as peptides may have the adverse effect of inducing peptide specific tolerance since rapid systemic distribution of the peptide may result in presenting peptides by too many non-professional APC ( Toes et al., 1996).
Studies to compare adjuvant efficiencies indicate that CTL induction is also possible with peptides emulsified with Montanide ISA 720, containing biodegradable oils, polyoxyethylene (Poe) lipid molecule (Poe-attached lipopeptide) or IFA. CTL were induced most efficiently with lipopeptides where one dose was sufficient and generated long-lived CTL. Peptide formulated in Montanide ISA 720 primed for CTL, but it required multiple boosts to be administered. Peptides attached to Poe-induced only short term CTL and were not able to stimulate long term CTL even after several immunizations (Hioe et al., 1996). A second study based on the malaria-derived minimal epitope SYIPSAEKI comparing the adjuvants IFA, NAGO (a mix of neuraminidase and galactose oxidase), QS21 (a saponin from the bark of the plant Quillaja saponaria), AF (an adjuvant formulation containing squalene, Tween 80 and pluronic L121), and lipopeptides concluded that lipopeptides were the most efficient way to induce CTL ( Allsopp et al., 1996). Other adjuvants such as immunostimulating complexes (ISCOMs) containing lipids, the saponin adjuvant Quil A, and aliphatic antigen have also been used successfully for CTL inductions ( Tarpey et al., 1991).
2.5. Liposomes and virosomes
Liposomes and virosomes are reconstituted membrane vesicles in the case of virosomes containing the viral envelope glycoproteins HA and neuraminidase. Liposome- or virosome-encapsulated peptides are delivered to the cytosol of the cell and may, thus, gain access to the MHC class I presentation pathway. The integration into membrane vesicles protects peptides from degradation by proteases in the extracellular fluid. Studies in the LCMV system by Ludewig et al. (2000) show that liposome-encapsulated immunodominant LCMV peptide gp33 is highly immunogenic when administered intradermally. Liposomes appear particularly well suited in targeting antigens to dentritic cells. Upon peripheral subcutaneous application, liposomes distribute preferentially via the lymph and reach local organized lymphoid tissue ( Oussoren et al., 1997) and are able to deliver their antigen into the cytosol of DC and stimulate CTL ( Allsopp et al., 1996).
Immunopotentiating reconstituted influenza virosomes (IRIV) contain the influenza virus HA and neuraminidase (Mengiardi et al., 1995). It is likely that HA in the virosomal membrane mediates cell entry of the virosomes through receptor-mediated endocytosis and subsequent fusion of virosomal and endosomal membrane ( Wilschut and Bron, 1993) The immunological helper activity of HA is presumably involved as well. Immunization studies by Arkema et al. (1999) based on whole ovalbumin protein demonstrate that fusion-active virosomes are very efficient at delivering antigens to the cytoplasm of DC and induce CTL. The CTL induction was not macrophages dependent as the response was not abrogated in macrophage-depleted mice. Two i.m. injections with 0.5 g virosome-encapsulated peptide were sufficient to raise a strong CTL response where 100 g of the same peptide emulsified in IFA was unsuccessful. Influenza-derived virosomes stimulate potent antibody responses to covalently linked proteins on the surface of the virosome and have found application in human hepatitis A, hepatitis B, and Influenza vaccine. Our group is investigating the potential to use virosomes as a carrier and adjuvant to induce efficient cellular immune response against HCV epitopes. Since immune responses are initiated in organized lymphoid tissues, immunogenic peptide formulations should, therefore, be designed to facilitate localized release and transport to organized lymphoid tissue. The surface of liposomes and virosomes may be specifically modified to optimize binding to DC for example via specific sugar moieties or antibodies (Fukasawa et al., 1998). Serre et al. (1998) showed that targeting of liposome-encapsulated antigen to DC in vitro may be augmented more than 1000-fold by antibody modifications of the liposomes.
2.6. Cytokines that support CTL induction
Cytokines such as IL-2, IL-12, IL-18 or GM-CSF or their combinations are now very often included in adjuvants used for experimental CTL induction. Cytokines responsible for delivering the growth and differential signals to the APC are important for the priming of a CTL response and can often replace the need for helper T cell epitope. Cytokines not only increase the magnitude of the immune response, but also steer it towards a desired phenotype (Ahlers et al., 1997). Berzofsky et al. (1999) documented that GM-CSF had very broad activity inducing Th1 as well as Th2 cytokines leading to both induced CTL response and antibody response. In contrast, IL-12 enhanced only the CTL and Th1 response and decreased Th2 response. The combination of the cytokines GM-CSF and IL-12 worked synergistically and stimulated strong CTL response to peptide vaccine. Three cytokines together, GM-CSF, IL-12, and TNF- enhanced the CTL response more than the combination of any two of them. The cytokines IL-2 and IFN- increased T cell proliferation and IFN- production but did not substantially increase CTL response ( Ahlers et al., 1997). GM-CSF seems to increase the APC activity in the draining lymph nodes ( Belyakow et al., 2000) possibly enhancing the presentation of administered peptide vaccine in the local lymph tissues. In contrast, IL-12 is known to be a key cytokine for optimal differentiation of CTL precursors and IFN- producing cells. Recently IL-18 gained interest as a cytokine able to upregulate cytotoxic activities of NK cells and cytotoxic T cells. It has been shown to strongly support the stimulation of tumor-specific CTL in vitro ( Tanaka et al., 2000) IL-18 may stimulate a Th2 immune response in the absence of IL-12, but combined application of IL-12 and IL-18 synergistically induces CTL priming ( Nakanishi et al., 2001). We are currently evaluating the possibility of using IL-18 in a virosome-based peptide vaccine.
2.7. Microbial components that support priming of CTL
E. coli heat labile toxin (LT) and the cholera toxin (CT) are widely used mucosal adjuvants. Mucosal immunization, in contrast to systemic application, leads to the generation of both mucosal and systemic immunity (Berzofsky et al., 1999). Intrarectal immunization with an HIV peptide vaccine containing the adjuvant LT- or CT-stimulated cytotoxic response in Peyer's patches, draining lymph nodes and the spleen after four applications. The CTL induction was synergistically enhanced by coadministration of the two cytokines GM-CSF and IL-12, reducing the required applications from four to two. Mutant forms of E. coli heat LT such as LT (R192G) or LTK63 are far less toxic compounds than the native LT or the CT and can be used for human application. LT mutants that retain some cAMP activity elicit Th1-type response and are suitable as adjuvants for vaccines whose main purpose is the induction of a Th1 T cell response. In contrast, the CT downregulates IL-12 production and, therefore, requires the complementation with exogenous IL-12 to be as effective as the LT. An immunization study based on the RSV minimal epitope M2 82-90 showed that LTK63 was able to induce long lasting systemic (spleen) and local (lymph node) immunity after two intranasal applications without the need for any cytokine. The mechanism by which bacterial enterotoxin mediate their adjuvant effects remains unclear but it is suggested that the E. coli heat LT and the CT directly modulate dendritic cells in the mucosa associated lymphoid tissue which makes them competent for productive T cell priming (Belyakow et al., 2000).
Many microbial components have been found to activate DC in a CD4 Th-independent way and are, therefore, of great help in situations where T cell help is difficult to provide such as in HIV-related diseases (Moroi et al., 2000). Activation of DC often operates via recognition of molecular patterns associated with microbial pathogens such as certain bacterial cell wall components. As members of the innate immune system, DC express pattern recognition receptors, members of the Toll-like receptor family, that bind pathogen-derived ligands and initiate a signaling cascade that lead to DC activation, inducing T cell costimulatory molecules and cytokine release ( Vabulas et al., 2000). Synthetic oligodinucleotides (ODNs) with CpG motifs imitating bacterial DNA show the capacity in vitro and in vivo to activate and to induce maturation of APC such as B cells, macrophages, and dendritic cells. The existence of a surface receptor able to selectively bind ODN has been postulated. Surface bound CpGâ?"ODN seem to be taken up by endocytosis ( Oxenius et al., 1999) inducing cellular activation processes which finally upregulate costimulatory molecules (CD40, CD80, and CD86), MHC class I and class II molecules, and the production of cytokines such as IL-12 ( Sparwasser et al., 1998). Subcutaneous administration of the immunodominant CTL epitope gp33 from LCMV together with CpGâ?"ODN conferred complete protection against LCMV challenge infection. Peptide-induced vaccinations for the generation of protective CTL responses with CpGâ?"ODN as adjuvant were at least as efficient as with complete Freunds adjuvant (CFA) ( Oxenius et al., 1999).
Srivastava and Amato as well as others have demonstrated that heat shock proteins (hsp) strongly enhance the immune reaction to associated antigen (Srivastava and Amato, 2001). The hsp70 or gp96 preparations isolated from tumor cells are able to induce specific CTL response and tumor protection. Other reports have shown that immunization with recombinant hsp70 and a peptide epitope from LCMV induced specific CTL responses and virus protection ( Moroi et al., 2000). The problem that not all potentially immunogenic peptides bind to hsp proteins could be circumvented by coupling the antigenic peptide to a hsp70 binding domain. That such hybrid peptides efficiently induce CTL was shown for an immunodominant CTL epitope from ovalbumin. The priming of CTL is, at least in part, likely due to the capacity of hsp70 to deliver peptides to MHC class I pathways of antigen-presenting cells ( Suzue et al., 1997). In ad***ion, bacterial hsp60 has been shown to activate monocytic cell lines, suggesting that hsp can activate antigen-presenting cells for competence to prime CTL response ( Moroi et al., 2000). In ad***ion to CTL induction hsp can also be used to stimulate T helper cells. For the CD4 T helper epitope of influenza virus nucleoprotein (NP) it was shown that the coupling to hsp70 allowed a strong CD4+ T cell response in mice ( Roman and Moreno, 1996).
Other microbial molecules such as OmpA (Jeannin et al., 2000) from bacterial cell wall or double stranded (ds) RNA, a viral replication intermediate, show similar effects on APC maturation and licensing for CTL priming. Their exact value in a peptide vaccine has yet to be evaluated. The ds RNA is very effective as an adjuvant in mice as well as in humans, but provokes severe side effects including fever, chills and dramatic weight loss ( Saravolac et al., 2001). With CpG, the toxic cytokine syndrome is much less marked and appears to occur only for a very short period. Future animal experiments and human trials will allow identification of the best suitable adjuvant in terms of immunostimulatory capacity and safety issues.
2.8. Human studies
Even before the discovery of MHC class I and class II binding motifs, it was shown that peptides could be used for primary in vitro induction of cytotoxic T lymphocytes (Carbone et al., 1988). After learning more about the molecular basis of CTL recognition through the identification and characterization of viral encoded peptides eluted from MHC class I molecules, it became possible to test the repertoire of naive sero-negative persons for the presence of HBV-specific cytolytic T cell precursors, which could readily be detected using repetitive in vitro restimulation ( Cerny et al., 1995). This was possible even in the absence of specific T helper cell stimulation by using a relatively high concentration of IL-2. This approach was taken from the laboratory bench to clinical application using a lipopeptide construct containing the immunodominant epitope HBV core 18-27 which binds to HLA-A2.1 with high affinity linked to a HTL epitope. The HTL epitope was covalently linked to the CTL epitope and contained a palmitic acid moiety at the N terminus. The TT peptide 830-843 was selected as the helper T cell epitope. The administration of four injections of up to 500 g per dose was found to yield CTL precursor frequencies similar to those previously observed in patients who successfully cleared HBV. The responses lasted for more than 1 month after the last injection and the presence of a TT helper T cell response was significantly associated with the development of memory CTL activity (Livingston and Vitiello). In an attempt to expand the limitations set by a particular HLA allele, Sette and co-workers demonstrated that the HBV core 18-27 peptide was capable of binding to a series of HLA alleles grouped together as supertypes. The broad binding activity of that particular epitope (HBV core 18-27) to different HLA class I was correlated to its in vivo immunogenicity in persons with different HLA class I alleles ( Livingston et al., 1999). This promising approach was taken even further to a pilot study lead by Heathcote et al. (1999), in which patients with chronic hepatitis B received up to four doses of the CY-1899 lipopeptide vaccine (ranging from 0.05 to 15 mg) within 6 weeks intervals. This vaccine consists of a palmitic acid lipid tail linked to the TT 830-843 helper T cell epitope which is linked to the cytotoxic T cell epitope core 18-27 with a spacer of 3-alanines. A total of 19 patients with chronic hepatitis B infection were immunized and cytotoxic T cell responses induced were low and dose dependent. Cytotoxic T cell responses remained lower than those observed following resolution of acute hepatitis B infection or those that could be induced by the vaccine to HBV negative individuals. Elevation of transaminases or e-antigen negativization was not increased over controls. This trial may serve as a proof of principle for a peptide-based immunotherapeutic approach demonstrating a dose-dependant augmentation of the cytotoxic T cell response to the HBV. The failure to reach viral elimination may be related to CD4 T cell dysfunction in chronic hepatitis B or to the fact that an insufficiently intense and polyspecific cytotoxic T cell response was induced. Future efforts will, thus, aim to induce polyspecific CTL responses respecting the need for more potent helper T cell induction.
Such an approach may be combined with antiviral treatment, thus, reducing the viral load.
Concay

2.4. Adjuvants for CTL induction
It has been shown that CTL induction is possible in vivo with chemically modified peptides linked to lipid components. The first lipopeptides used were peptides conjugated to tripalmitoyl-S-glycerylcysteinyl-seryl-serine (P3CSS) synthesized according to the immunologically active N-terminal sequence of the principle lipoprotein of Escherichia coli (Deres et al., 1998). The lipopeptide used in later experiments may only have one or two palmitoyl groups. Experiments on lipidated peptides in the HLA-A2 transgenic mouse model showed that induction of CTL responses was crucially dependent on the presence of HTL and most efficient in the case of lipidated covalently linked HTLâ?"CTL epitope constructs. These last were considerably more immunogenic than mixtures of unlipidated HTLâ?"CTL epitope emulsified in IFA. The magnitude of the CTL response was similar to the naturally occurring response associated with clearance of acute hepatitis B ( Oseroff and Deres). High CTL responses compatible with protection were induced by immunizing primates and humans with lipopeptides derived from SIV and HBV sequences ( Bourgault-Villada et al., 1997). The lipid seems to serve three different roles: (a) it leads to the persistence of peptides at the site of injection or in the draining lymph node for sufficient time to induce a CTL response; (b) lipid modification may help to translocate peptides across the plasma membrane into the cytoplasm; and c) the lipid moiety may also have intrinsic adjuvant activity ( Vitiello et al., 1995). Local concentration of antigen in lymphoid organs seems to be particularly important for peptide immunizations as peptides may have the adverse effect of inducing peptide specific tolerance since rapid systemic distribution of the peptide may result in presenting peptides by too many non-professional APC ( Toes et al., 1996).
Studies to compare adjuvant efficiencies indicate that CTL induction is also possible with peptides emulsified with Montanide ISA 720, containing biodegradable oils, polyoxyethylene (Poe) lipid molecule (Poe-attached lipopeptide) or IFA. CTL were induced most efficiently with lipopeptides where one dose was sufficient and generated long-lived CTL. Peptide formulated in Montanide ISA 720 primed for CTL, but it required multiple boosts to be administered. Peptides attached to Poe-induced only short term CTL and were not able to stimulate long term CTL even after several immunizations (Hioe et al., 1996). A second study based on the malaria-derived minimal epitope SYIPSAEKI comparing the adjuvants IFA, NAGO (a mix of neuraminidase and galactose oxidase), QS21 (a saponin from the bark of the plant Quillaja saponaria), AF (an adjuvant formulation containing squalene, Tween 80 and pluronic L121), and lipopeptides concluded that lipopeptides were the most efficient way to induce CTL ( Allsopp et al., 1996). Other adjuvants such as immunostimulating complexes (ISCOMs) containing lipids, the saponin adjuvant Quil A, and aliphatic antigen have also been used successfully for CTL inductions ( Tarpey et al., 1991).
2.5. Liposomes and virosomes
Liposomes and virosomes are reconstituted membrane vesicles in the case of virosomes containing the viral envelope glycoproteins HA and neuraminidase. Liposome- or virosome-encapsulated peptides are delivered to the cytosol of the cell and may, thus, gain access to the MHC class I presentation pathway. The integration into membrane vesicles protects peptides from degradation by proteases in the extracellular fluid. Studies in the LCMV system by Ludewig et al. (2000) show that liposome-encapsulated immunodominant LCMV peptide gp33 is highly immunogenic when administered intradermally. Liposomes appear particularly well suited in targeting antigens to dentritic cells. Upon peripheral subcutaneous application, liposomes distribute preferentially via the lymph and reach local organized lymphoid tissue ( Oussoren et al., 1997) and are able to deliver their antigen into the cytosol of DC and stimulate CTL ( Allsopp et al., 1996).
Immunopotentiating reconstituted influenza virosomes (IRIV) contain the influenza virus HA and neuraminidase (Mengiardi et al., 1995). It is likely that HA in the virosomal membrane mediates cell entry of the virosomes through receptor-mediated endocytosis and subsequent fusion of virosomal and endosomal membrane ( Wilschut and Bron, 1993) The immunological helper activity of HA is presumably involved as well. Immunization studies by Arkema et al. (1999) based on whole ovalbumin protein demonstrate that fusion-active virosomes are very efficient at delivering antigens to the cytoplasm of DC and induce CTL. The CTL induction was not macrophages dependent as the response was not abrogated in macrophage-depleted mice. Two i.m. injections with 0.5 g virosome-encapsulated peptide were sufficient to raise a strong CTL response where 100 g of the same peptide emulsified in IFA was unsuccessful. Influenza-derived virosomes stimulate potent antibody responses to covalently linked proteins on the surface of the virosome and have found application in human hepatitis A, hepatitis B, and Influenza vaccine. Our group is investigating the potential to use virosomes as a carrier and adjuvant to induce efficient cellular immune response against HCV epitopes. Since immune responses are initiated in organized lymphoid tissues, immunogenic peptide formulations should, therefore, be designed to facilitate localized release and transport to organized lymphoid tissue. The surface of liposomes and virosomes may be specifically modified to optimize binding to DC for example via specific sugar moieties or antibodies (Fukasawa et al., 1998). Serre et al. (1998) showed that targeting of liposome-encapsulated antigen to DC in vitro may be augmented more than 1000-fold by antibody modifications of the liposomes.
2.6. Cytokines that support CTL induction
Cytokines such as IL-2, IL-12, IL-18 or GM-CSF or their combinations are now very often included in adjuvants used for experimental CTL induction. Cytokines responsible for delivering the growth and differential signals to the APC are important for the priming of a CTL response and can often replace the need for helper T cell epitope. Cytokines not only increase the magnitude of the immune response, but also steer it towards a desired phenotype (Ahlers et al., 1997). Berzofsky et al. (1999) documented that GM-CSF had very broad activity inducing Th1 as well as Th2 cytokines leading to both induced CTL response and antibody response. In contrast, IL-12 enhanced only the CTL and Th1 response and decreased Th2 response. The combination of the cytokines GM-CSF and IL-12 worked synergistically and stimulated strong CTL response to peptide vaccine. Three cytokines together, GM-CSF, IL-12, and TNF- enhanced the CTL response more than the combination of any two of them. The cytokines IL-2 and IFN- increased T cell proliferation and IFN- production but did not substantially increase CTL response ( Ahlers et al., 1997). GM-CSF seems to increase the APC activity in the draining lymph nodes ( Belyakow et al., 2000) possibly enhancing the presentation of administered peptide vaccine in the local lymph tissues. In contrast, IL-12 is known to be a key cytokine for optimal differentiation of CTL precursors and IFN- producing cells. Recently IL-18 gained interest as a cytokine able to upregulate cytotoxic activities of NK cells and cytotoxic T cells. It has been shown to strongly support the stimulation of tumor-specific CTL in vitro ( Tanaka et al., 2000) IL-18 may stimulate a Th2 immune response in the absence of IL-12, but combined application of IL-12 and IL-18 synergistically induces CTL priming ( Nakanishi et al., 2001). We are currently evaluating the possibility of using IL-18 in a virosome-based peptide vaccine.
2.7. Microbial components that support priming of CTL
E. coli heat labile toxin (LT) and the cholera toxin (CT) are widely used mucosal adjuvants. Mucosal immunization, in contrast to systemic application, leads to the generation of both mucosal and systemic immunity (Berzofsky et al., 1999). Intrarectal immunization with an HIV peptide vaccine containing the adjuvant LT- or CT-stimulated cytotoxic response in Peyer's patches, draining lymph nodes and the spleen after four applications. The CTL induction was synergistically enhanced by coadministration of the two cytokines GM-CSF and IL-12, reducing the required applications from four to two. Mutant forms of E. coli heat LT such as LT (R192G) or LTK63 are far less toxic compounds than the native LT or the CT and can be used for human application. LT mutants that retain some cAMP activity elicit Th1-type response and are suitable as adjuvants for vaccines whose main purpose is the induction of a Th1 T cell response. In contrast, the CT downregulates IL-12 production and, therefore, requires the complementation with exogenous IL-12 to be as effective as the LT. An immunization study based on the RSV minimal epitope M2 82-90 showed that LTK63 was able to induce long lasting systemic (spleen) and local (lymph node) immunity after two intranasal applications without the need for any cytokine. The mechanism by which bacterial enterotoxin mediate their adjuvant effects remains unclear but it is suggested that the E. coli heat LT and the CT directly modulate dendritic cells in the mucosa associated lymphoid tissue which makes them competent for productive T cell priming (Belyakow et al., 2000).
Many microbial components have been found to activate DC in a CD4 Th-independent way and are, therefore, of great help in situations where T cell help is difficult to provide such as in HIV-related diseases (Moroi et al., 2000). Activation of DC often operates via recognition of molecular patterns associated with microbial pathogens such as certain bacterial cell wall components. As members of the innate immune system, DC express pattern recognition receptors, members of the Toll-like receptor family, that bind pathogen-derived ligands and initiate a signaling cascade that lead to DC activation, inducing T cell costimulatory molecules and cytokine release ( Vabulas et al., 2000). Synthetic oligodinucleotides (ODNs) with CpG motifs imitating bacterial DNA show the capacity in vitro and in vivo to activate and to induce maturation of APC such as B cells, macrophages, and dendritic cells. The existence of a surface receptor able to selectively bind ODN has been postulated. Surface bound CpGâ?"ODN seem to be taken up by endocytosis ( Oxenius et al., 1999) inducing cellular activation processes which finally upregulate costimulatory molecules (CD40, CD80, and CD86), MHC class I and class II molecules, and the production of cytokines such as IL-12 ( Sparwasser et al., 1998). Subcutaneous administration of the immunodominant CTL epitope gp33 from LCMV together with CpGâ?"ODN conferred complete protection against LCMV challenge infection. Peptide-induced vaccinations for the generation of protective CTL responses with CpGâ?"ODN as adjuvant were at least as efficient as with complete Freunds adjuvant (CFA) ( Oxenius et al., 1999).
Srivastava and Amato as well as others have demonstrated that heat shock proteins (hsp) strongly enhance the immune reaction to associated antigen (Srivastava and Amato, 2001). The hsp70 or gp96 preparations isolated from tumor cells are able to induce specific CTL response and tumor protection. Other reports have shown that immunization with recombinant hsp70 and a peptide epitope from LCMV induced specific CTL responses and virus protection ( Moroi et al., 2000). The problem that not all potentially immunogenic peptides bind to hsp proteins could be circumvented by coupling the antigenic peptide to a hsp70 binding domain. That such hybrid peptides efficiently induce CTL was shown for an immunodominant CTL epitope from ovalbumin. The priming of CTL is, at least in part, likely due to the capacity of hsp70 to deliver peptides to MHC class I pathways of antigen-presenting cells ( Suzue et al., 1997). In ad***ion, bacterial hsp60 has been shown to activate monocytic cell lines, suggesting that hsp can activate antigen-presenting cells for competence to prime CTL response ( Moroi et al., 2000). In ad***ion to CTL induction hsp can also be used to stimulate T helper cells. For the CD4 T helper epitope of influenza virus nucleoprotein (NP) it was shown that the coupling to hsp70 allowed a strong CD4+ T cell response in mice ( Roman and Moreno, 1996).
Other microbial molecules such as OmpA (Jeannin et al., 2000) from bacterial cell wall or double stranded (ds) RNA, a viral replication intermediate, show similar effects on APC maturation and licensing for CTL priming. Their exact value in a peptide vaccine has yet to be evaluated. The ds RNA is very effective as an adjuvant in mice as well as in humans, but provokes severe side effects including fever, chills and dramatic weight loss ( Saravolac et al., 2001). With CpG, the toxic cytokine syndrome is much less marked and appears to occur only for a very short period. Future animal experiments and human trials will allow identification of the best suitable adjuvant in terms of immunostimulatory capacity and safety issues.
2.8. Human studies
Even before the discovery of MHC class I and class II binding motifs, it was shown that peptides could be used for primary in vitro induction of cytotoxic T lymphocytes (Carbone et al., 1988). After learning more about the molecular basis of CTL recognition through the identification and characterization of viral encoded peptides eluted from MHC class I molecules, it became possible to test the repertoire of naive sero-negative persons for the presence of HBV-specific cytolytic T cell precursors, which could readily be detected using repetitive in vitro restimulation ( Cerny et al., 1995). This was possible even in the absence of specific T helper cell stimulation by using a relatively high concentration of IL-2. This approach was taken from the laboratory bench to clinical application using a lipopeptide construct containing the immunodominant epitope HBV core 18-27 which binds to HLA-A2.1 with high affinity linked to a HTL epitope. The HTL epitope was covalently linked to the CTL epitope and contained a palmitic acid moiety at the N terminus. The TT peptide 830-843 was selected as the helper T cell epitope. The administration of four injections of up to 500 g per dose was found to yield CTL precursor frequencies similar to those previously observed in patients who successfully cleared HBV. The responses lasted for more than 1 month after the last injection and the presence of a TT helper T cell response was significantly associated with the development of memory CTL activity (Livingston and Vitiello). In an attempt to expand the limitations set by a particular HLA allele, Sette and co-workers demonstrated that the HBV core 18-27 peptide was capable of binding to a series of HLA alleles grouped together as supertypes. The broad binding activity of that particular epitope (HBV core 18-27) to different HLA class I was correlated to its in vivo immunogenicity in persons with different HLA class I alleles ( Livingston et al., 1999). This promising approach was taken even further to a pilot study lead by Heathcote et al. (1999), in which patients with chronic hepatitis B received up to four doses of the CY-1899 lipopeptide vaccine (ranging from 0.05 to 15 mg) within 6 weeks intervals. This vaccine consists of a palmitic acid lipid tail linked to the TT 830-843 helper T cell epitope which is linked to the cytotoxic T cell epitope core 18-27 with a spacer of 3-alanines. A total of 19 patients with chronic hepatitis B infection were immunized and cytotoxic T cell responses induced were low and dose dependent. Cytotoxic T cell responses remained lower than those observed following resolution of acute hepatitis B infection or those that could be induced by the vaccine to HBV negative individuals. Elevation of transaminases or e-antigen negativization was not increased over controls. This trial may serve as a proof of principle for a peptide-based immunotherapeutic approach demonstrating a dose-dependant augmentation of the cytotoxic T cell response to the HBV. The failure to reach viral elimination may be related to CD4 T cell dysfunction in chronic hepatitis B or to the fact that an insufficiently intense and polyspecific cytotoxic T cell response was induced. Future efforts will, thus, aim to induce polyspecific CTL responses respecting the need for more potent helper T cell induction.
Such an approach may be combined with antiviral treatment, thus, reducing the viral load.
Concay

3. Conclusions
The discovery of MHC-restriction and the rules governing the recognition of viruses by the cellular immune system has given us new opportunities to selectively induce specific effector mechanisms of acquired immunity. This article reviews progress made in the field of HBV vaccine development using immunogenic peptides in animal and human studies both in vitro and in vivo. The approach has so far taken us to a first large scale multicenter trial in patients with chronic hepatitis B infection. While the failure to reach viral clearance in immunized patients may seem disappointing it is nonetheless important proof of the principle that peptide-based vaccines can raise a significant cellular immune responses in humans. Improving our knowledge about disease pathogenesis combined with more potent vaccine approaches eventually combined with other antiviral strategies are the next steps.

Acknowledgements
The authors thank Lulu Cerny for carefully reviewing the manuscript for the proper use of the English language. This work was supported by grants from the European Commission, Brussels, Belgium (PL95-1064 and QLRT-PL1999-00356), the Swiss National Science Foundation, Berne, Switzerland (SNF 32-52915.97, 32-59564.99, and (31-52284.97), the Department of Clinical Research of the University of Berne, Switzerland, the Helmut Horten Foundation, Lugano, Switzerland, the Désirée and Niels Yde Foundation, Zurich, Switzerland, and the Central Laboratory Blood Transfusion Service of the Swiss Red Cross, Berne, Switzerland.

Concay

3. Conclusions
The discovery of MHC-restriction and the rules governing the recognition of viruses by the cellular immune system has given us new opportunities to selectively induce specific effector mechanisms of acquired immunity. This article reviews progress made in the field of HBV vaccine development using immunogenic peptides in animal and human studies both in vitro and in vivo. The approach has so far taken us to a first large scale multicenter trial in patients with chronic hepatitis B infection. While the failure to reach viral clearance in immunized patients may seem disappointing it is nonetheless important proof of the principle that peptide-based vaccines can raise a significant cellular immune responses in humans. Improving our knowledge about disease pathogenesis combined with more potent vaccine approaches eventually combined with other antiviral strategies are the next steps.

Acknowledgements
The authors thank Lulu Cerny for carefully reviewing the manuscript for the proper use of the English language. This work was supported by grants from the European Commission, Brussels, Belgium (PL95-1064 and QLRT-PL1999-00356), the Swiss National Science Foundation, Berne, Switzerland (SNF 32-52915.97, 32-59564.99, and (31-52284.97), the Department of Clinical Research of the University of Berne, Switzerland, the Helmut Horten Foundation, Lugano, Switzerland, the Désirée and Niels Yde Foundation, Zurich, Switzerland, and the Central Laboratory Blood Transfusion Service of the Swiss Red Cross, Berne, Switzerland.

Concay