Is a Chlamydia vaccine a reality?

Best Practice & Research Clinical Obstetrics and Gynaecology Vol. 16, No. 6, pp. 889±900, 2002

doi:10.1053/beog.2002.0324, available online at http://www.idealibrary.com on

10 Is a Chlamydia vaccine a reality? Gunna Christiansen*

MD, DMSc

Professor, Chairman

Svend Birkelund

MD, PhD, DMSc

Associate Professor Department of Medical Microbiology and Immunology, The Bartholin Building, University of Aarhus, DK-8000 Aarhus C, Denmark

Chlamydia trachomatis is a leading cause of sexually transmitted bacterial infections with severe sequelae such as tubal factor infertility and ectopic pregnancy; infections can also be asymptomatic. So far no vaccine has been developed but studies that may lead to the development of a highly warranted vaccine have been performed. The ®rst attempt to vaccinate children with a whole-cell vaccine initially resulted in protection but the protection was shortlived. In animal models whole-cell vaccination resulted in hypersensitivity reactions, so that new strategies were devised. The ®rst immunogenic molecule described was the major outer membrane protein (MOMP), and this molecule has therefore been studied in great detail as a candidate vaccine. Even though complete protection was not obtained, reduced shedding was observed and vaccine trials in animal models using naked DNA as a vaccine resulted in stimulation of both the humoral and cellular immune response, indicating progress in the development of a vaccine. Key words: Chlamydia trachomatis; humoral immune response; cell-mediated immune response; vaccine; recombinant proteins; animal models.

Developmental cycle Chlamydia is an obligate intracellular Gram-negative bacterium with a unique biphasic developmental cycle. The small (300 nm in diameter) infectious, but metabolically inactive elementary bodies (EBs) ®rst attach to susceptible host cells where they induce their own phagocytosis. Their uptake is within a phagosome, and during their intracellular life chlamydiae are surrounded by a phagosomal membrane called the inclusion membrane. Shortly after uptake an EB transforms into the larger (1000 nm), metabolically active reticulate body (RB); within the phagosome an RB divides by binary ®ssion, and after 48±72 hours, depending on species and serovar, each RB transforms into an EB, the inclusion bursts and infectious EBs are liberated. In contrast to what is normally seen, the phagosomes do not fuse with lysosomes. In that way chlamydiae are protected from the normal intracellular degradation process. The human humoral immune response may thus recognize Chlamydia only during its *Principal correspondent c 2002 Elsevier Science Ltd. All rights reserved. 1521±6934/02/$ - see front matter *

890 G. Christiansen and S. Birkelund

extracellular phase when surface-exposed molecules are the major targets. In contrast, the cellular immune response may recognize the infected cells if chlamydial proteins are processed in the host cell cytoplasm and presented for the immune system at the surface of infected cells.1 Serovar speci®city and disease development Chlamydia trachomatis causes genital and ocular infections. Serovars A±C cause trachoma, serovars D±K cause sexually transmitted genital infections and neonatal ocular and pulmonary infections, and serovars L1±L3 cause the sexually transmitted systemic infection lymphogranuloma venereum.1 The molecular background for serovar variation is found in variation in the major outer membrane protein (MOMP). In C. trachomatis MOMP is the predominant outer membrane protein covering the surface of both RB and EB. Genetically, MOMP has four variable surface exposed regions (variable sequence VS I±IV) and four constant sequences (CS 1±4)2 (Figure 1). It is highly immunogenic and contains both speciesand serovar-speci®c epitopes.2,3 Acute reactive arthritis4, pelvic in¯ammatory disease (PID)5 and tubal scarring6±8 may develop as a consequence of infection. Sequelae to female genital infections are thus ectopic pregnancy and tubal factor infertility (TFI). Because asymptomatic C. trachomatis infections can also develop into TFI, treatment of the infection may not be instituted. Therefore, it is relevant to seek to develop a vaccine against C. trachomatis infections.

IMMUNE RESPONSE TO C. trachomatis INFECTIONS Humoral immune response Following genital infection with C. trachomatis a humoral immune response develops. Antibodies to C. trachomatis are directed against surface components such as the genus-speci®c lipopolysaccharide (LPS) and MOMP; such antibodies can be detected by microimmuno¯uorescence microscopy (MIF).9 Other major antigens are the cysteinerich outer membrane protein 2 (Omp2) and the chlamydial homologues to heat shock proteins 60 and 70 (chlamydial Hsp60/GroEL and chlamydial Hsp70/DnaK).4 Omp2 is localized in the outer membrane of chlamydiae (Figure 2) but it is debated whether it gets access to the surface.10,11 The chlamydial Hsp60 and Hsp70 are chaperones situated within the chlamydial cytoplasm. Several other proteins have been shown to be immunogenic. In a study by Sanchez-Campillo et al12 identi®cation of immunoreactive proteins from C. trachomatis involved a proteomics approach. They separated proteins from puri®ed C. trachomatis L2 EB by two-dimensional gel electrophoresis, transferred the proteins to nitrocellulose membranes and reacted the membranes with patient sera. Sera from three categories of patients were analysed: cervicitis, PID and sterility. Interestingly, patient blots showed individually di€erent patterns comprising a number of spots, varying from two to 28. There was no speci®c pattern that correlated to a disease. All sera recognized Omp2, most, including the sterility patients, reacted with chlamydial GroEL, and several reacted with MOMP and chlamydial DnaK. Also, interestingly, all the patients su€ering from cervicitis reacted with all these proteins and this was not the case for patients with PID and sterility. In addition, some of the sera reacted with the cytoplasmic proteins EF-Tu and ribosomal protein S1. The rest of the spots were characterized by molecular size and isoelectric

Figure 1. Multiple alignment of C. trachomatis MOMP from the genital serovars. The amino acids are shown in one-letter code. Only the variable sequences are shown. C complex and B complex refer to the division of the serovars. VSI-IV are variable sequence I-IV, CS2±4 are constant sequence 2±4 positioned between the variable sequences.

Is a Chlamydia vaccine a reality? 891

892 G. Christiansen and S. Birkelund

Figure 2. Electron micrograph of sarcosyl extracted C. trachomatis elementary body (EB). The Chlamydia outer membrane complex was prepared for electron microscopy by negative staining with phosphotungstic acid 1%. Inserted is a schematic drawing of the structure of the C. trachomatis outer membrane. LPS, MOMP and probably some polymorphic membrane protein (Pmp) proteins are localized at the surface. Omp2 is primarily localized in the periplasmic space, and the lipoprotein Omp3 is attached to the outer membrane by the lipid anchor and to Omp2 by S±S bridges.

point but could not be correlated to speci®c gene products due to lack of genomic data. During an acute infection antibodies are generated. Because such antibodies can be detected by MIF they are most probably generated against MOMP, but, as shown by Sanchez-Campillo et al12, antibodies are also generated against other proteins. In addition to anti-protein antibodies, antibodies are also generated against the genusspeci®c LPS.

Is a Chlamydia vaccine a reality? 893

PID can lead to tubal factor infertility, and even though C. trachomatis cannot be detected, antibodies to C. trachomatis are present. The antibodies are directed mainly against MOMP, Omp2 and chlamydial GroEL. Because chlamydial GroEL has high homology to human HSP60 it has been speculated that antibody cross-reaction to human Hsp60 proteins could be responsible for the development of TFI.13 Antibodies to chlamydial GroEL are frequently found in serum from women with tubal factor infertility, and Persson et al14 found that antibodies to chlamydial GroEL in women with tubal factor infertility are associated with prior infection with C. trachomatis but not with C. pneumoniae. Cellular immune response During a C. trachomatis infection chlamydiae can be phagocytosed by antigenpresenting cells or macrophages. In such cells the chlamydiae-containing phagosome will fuse with lysosomes, the chlamydial antigens are processed and eventually peptide fragments are presented in the groove of HLA class II at the cellular surface, where they will stimulate a T-helper cell response. Such C. trachomatis-speci®c CD4‡ T cells are activated after recognition of the peptide±MHC-class II complexes, and the activated CD4‡ T cells function both as inhibitors of C. trachomatis replication and as stimulators of protective immunity involving immune and in¯ammatory cells.15 Several C. trachomatis antigenic peptides capable of stimulating CD4‡ T cells have been identi®ed. CD4‡ T cell lines from infected patients have thus been shown to be stimulated by MOMP, Omp2, GroEL and PmpD, a protein belonging to the polymorphic membrane protein family, but Enolase and a predicted YopD ortolog have also been shown to activate CD4‡ T cells.15±19 In infected cells, which, for genital C. trachomatis infections, are epithelial cells, immune surveillance is done by the CD8‡ T cells. These cells recognize C. trachomatis peptides presented in MHC class I of the infected cells.15 In the infected cells, chlamydiae are found within the phagosome. Even though the chlamydiae are thus surrounded by the phagosome membrane, chlamydia-derived proteins are transported to the phagosome membrane in contact with the cytosol of the host cell. The ®rst observation was that a C. trachomatis-encoded protein was found in the phagosome membrane20, and later it was found that another C. trachomatis-encoded protein, CPAF, was secreted to the host cell cytoplasm.21 Genes encoding the type III secretion system were identi®ed in the C. trachomatis genome22,23, but whether any of these proteins are secreted by this system remains to be determined. In infected cells, it is thus possible for chlamydial proteins to be transported into the host cell cytosol. In order for such proteins to get access to the MHC class I the transported proteins must be degraded by proteolytic cleavage in the proteasomes and transported by the TAP protein to the endoplasmic reticulum where they combine with MHC class I molecules; the complex then makes its way through the Golgi stacks to be incorporated into the host cell membrane. The size of such peptides is limited to 8±9 amino acids, and so far MOMP and Cap1 are the only C. trachomatis proteins from which CD8‡ T cell epitopes have been identi®ed.24,25 Upon recognition of the MHC class I-peptide complex the CD8‡ T cells become activated. The activated CD8‡ T cells possess a variety of e€ector functions such as secretion of cytokines and lysis of infected cells, and, in addition, CD4‡ and CD8‡ T cells are interdependent in many of their functions. In order to obtain protective immunity a complex interplay between the innate, the humoral and the cellular immune response must take place. An understanding of this

894 G. Christiansen and S. Birkelund

complex system will help to identify components of importance for the development of a vaccine. APPROACHES TO THE DEVELOPMENT OF A VACCINE Original approach In the ®rst approach to induce protection against C. trachomatis infections, Dhir et al26 conducted vaccine trials in an area of high trachoma prevalence in Northern India. Two formalin-inactivated whole-cell C. trachomatis EB preparations, a gentron-puri®ed (tri¯uorotrichloroethane-treated antigen) vaccine and a gradient-puri®ed vaccine, each containing approximately equal amounts of EB, were used for immunization and compared with placebo immunizations; 451 children were vaccinated. Initially, good protection was found. Only one child receiving the gradient-puri®ed vaccine and four children in the gentron-puri®ed vaccine group contracted trachoma in the period of 3 months between the initial vaccination and the booster vaccination compared to 11 children receiving the placebo vaccine. The protection was, however, short. After 1 year the e€ectiveness of the gradient vaccine was 29% and of the gentron vaccine 38%. In a long-term follow-up study 12 years later, an equal number of children in each of the three groups had signs of mostly minimally active trachoma, and, in addition, 6± 10% in each group had signs of potentially blinding sequelae.27 Thus the initial observation of protection had been reversed and there was no long-term protection. There was, however, no evidence of adverse e€ects from the vaccines. This was in contrast to what was found in monkey vaccine trials where higher attack rates and aggravated eye disease were found in the vaccinated animals challenged with heterologous strains, even though protection against infection with homologous strains was obtained.28 Further evidence for a possible hypersensitivity reaction came with the discovery by Morrison et al13 that the 57 kDa chlamydial hypersensitivity antigen was identi®ed as the homologue of the Escherichia coli GroEL/Hsp60, a protein with homologues in both prokaryotic and eukaryotic cells. A guinea-pig model of inclusion conjunctivitis was used by Morrison et al29 to identify chlamydial antigens that may be involved in the development of deleterious immune response resulting in blinding trachoma. They found that the genus-speci®c 57-kDa protein (chlamydial GroEL) puri®ed from chlamydial elementary bodies elicited an ocular hypersensitivity response. This response was characterized by a predominantly mononuclear macrophage and lymphocyte cellular in®ltrate of the submucosal epithelium. The clinical and histological ®ndings were consistent with those of a delayed hypersensitivity response. The role of this protein was elegantly studied by Patton et al30 using the monkey `pocket' model. Pigtailed monkeys were sensitized by inoculation of live C. trachomatis serovar E EB into subcutaneous pockets containing salpingeal autotransplants. At 21 days, anity-puri®ed recombinant C. trachomatis GroEL was injected into pockets either previously sensitized with C. trachomatis or not sensitized. Delayed-type hypersensitivity reaction was observed, characterized by mononuclear cell in®ltration with peak reaction at 48 hours. In this study it was demonstrated that recombinant chlamydial GroEL was able to induce delayed-type hypersensitivity in animals sensitized with live C. trachomatis organisms. These early trials thus suggested a similar pathogenesis for both salpingitis and trachoma but also pointed out two problems of importance for the development of a vaccine: short-lasting protection and potential development of delayed hypersensitivity when whole-cell vaccines were used.

Is a Chlamydia vaccine a reality? 895

Animal models used in protection experiments Several animal models have been developed for studying genital chlamydial infections.31 In such models it is possible to study the local infection, spreading of the infection, development of hypersensitivity, production of antibodies, secretion of cytokines and protection against (re)infection. Even though there are limitations in terms of how similar such models are compared with human infections they provide valuable models for studying the development of infections and for studying protection. In an animal model it was con®rmed that repeated C. trachomatis infection of monkey fallopian tubes was associated with the development of ®brosis and scarring, and that the monkeys developed a Th-1-like cytokine response.32 Furthermore, monkeys with chronic pelvic in¯ammatory disease developed an antibody response to chlamydial GroEL33 similar to that seen in humans. Also a guinea pig model has been developed to study the immune response to chlamydial genital infections31 and in this model a short-lived resistance to chlamydial re-infections was found34, similar to that found in the early vaccine trials.27 Inbred and knock-out mice are increasingly used as animal models to study speci®c immune reactions. A mouse model was developed to study C. trachomatis-induced infertility in mice.35 Using the mouse model it was found that genetic factors played a role in the development of protection36 and that both the humoral immune response37, the route of infection and T cell response38, NK cells39 and Fc receptor regulation40 played a role in the development of protective immunity. It is on this background that a vaccine mediating sterilizing protective immunity should be developed. Such a vaccine most likely should include several antigenic components capable of stimulating various parts of the immune system. Subcomponent vaccine approach An e€ective vaccine should mediate protection, confer immunity to all serovars and avoid development of immunopathology. Because it is suspected that a wholemicroorganism vaccine confers immunopathology, several subcomponent approaches have been attempted. It is, however, not clear which part of the immune system is the most appropriate to target for immunoprophylaxis, and whether protection can be obtained without development of immunopathology. Little is known concerning the response in humans, and despite the similarity in disease development it is unclear to what extent immunoreactions seen in animal models can be translated to e€ects in humans. MOMP is the best studied C. trachomatis molecule. It is an attractive molecule to select as a vaccine candidate because it is the dominant surface exposed protein on C. trachomatis EBs and RBs. It is immunodominant for antibody responses, and certain monoclonal antibodies to MOMP are neutralizing in a tissue culture assay.41,42 Furthermore, MOMP does not seem to elicit immunopathology. MOMP is, however, variable (Figure 2), and it is to the variable domains of MOMP that antibodies are generated. Therefore, a MOMP-based vaccine probably should contain MOMP sequences from di€erent serovars. MOMP also contains T-cell epitopes43, but from humans who have had C. trachomatis infections most of these epitopes are localized in the constant sequences of the molecule44±46 (Figure 2). MOMP vaccines and di€erent formulations The ®rst MOMP vaccine studies in which oligopeptides or recombinant full-length or fragmentary MOMP sequences were used showed, at best, partial protective immunity

896 G. Christiansen and S. Birkelund

(for a review see Beagley and Timms47). More promising results were recently obtained by Pal et al48 who puri®ed and re-folded MOMP obtained by extraction of the molecule from C. trachomatis MoPn and mixed the preparation with Freund's adjuvant using vortexing or sonication. By SDS-PAGE the two preparations di€ered, the vortexed preparation showing bands of 42, 62 and 132 kDa probably consisting primarily of homopolymers of MOMP while the sonicated preparation showed only a band of 42 kDa containing monomeric MOMP. BALB/c mice were immunized and subsequently challenged with live C. trachomatis MoPn in the upper genital tract. Vaginal cultures were taken weekly and the mice were mated after 6 weeks. Following both immunizations a strong humoral immune response was developed. The cellmediated immune response was strong after immunization with the vortexed preparation but weak following immunization with the sonicated preparation. Better protection in terms of less vaginal shedding and a higher fertility was observed when vortexed MOMP vaccine was used. These results illustrate the importance of correct presentation of the proteins to the immune system to obtain protection based on both the humoral and cellular immune responses. The results also indicate that MOMP may still be the only required component in a protective vaccine when C. trachomatis MoPn is used. Whether this is also the case for other C. trachomatis serovars remains to be determined. The results are in agreement with results obtained by Igietseme and Murdin49 who prepared a MOMP±ISCOM vaccine based on MOMP extracted from C. trachomatis serovar D for immunization of mice. This vaccine was able to produce a Th1 antigen-speci®c immune response, and immunized mice cleared a vaginal challenge within 1 week. Immunity was still present 8 weeks after primary infection. Correct folding of recombinantly produced membrane proteins is dicult and no speci®c method that guarantees correct folding has been developed. This is the drawback of these experiments as it would be impossible to mass-produce enough C. trachomatis EB from di€erent serovars to produce a human vaccine. DNA vaccine It was therefore attractive to develop a di€erent concept for application of a MOMP vaccine. Zhang et al50 developed a DNA vaccine which, unlike protein- or peptidebased subcomponent vaccines, provides protection following endogenous expression of the protein encoded by the DNA by expression in host cells, thus allowing the presentation of the antigen to the host's immune system in a way similar to that which occurs during natural infection. Upon intramuscular DNA immunization both a humoral and a cellular immune response were developed when DNA encoding MOMP but not a cytoplasmic enzyme (cytosine triphosphate synthetase) was used. These studies were further developed to study the immune response following the DNA immunization.51 The results showed that the serum antibody response was mainly of the IgG2a type, and that two-thirds of the mice developed such a response. Furthermore, the vaccinated mice developed an antigen-speci®c cellular immune response indicative of a CD4‡ T-helper cell response. The mice showed lymphocyte proliferation and production of interferon-g, a response similar to that seen during infection. A genomics approach to the development of a vaccine Even though MOMP is still the leading candidate for the development of a C. trachomatis vaccine, other approaches may be attractive. With the publication of

Is a Chlamydia vaccine a reality? 897

the C. trachomatis serovar D genome22 new possibilities have occurred. Translation of the genome sequence gave information on which proteins theoretically could be synthesized. Taking advantage of the genome sequence, Shaw et al52 used proteomics to analyse the actual content of proteins in C. trachomatis serovar D and made a comparison with the content of proteins in serovars A and L2. Such analysis could be extended to determine the proteins synthesized to speci®c time points53 and for proteins present in speci®c compartments in the cell, thereby seeking information on identi®cation of speci®c new vaccine candidates that could be a supplement to MOMP. In order to obtain protective immunity there are several steps in the developmental cycle of chlamydiae where the immune system could attack: after transmission of infectious EBs, or when infectious EBs are released from the infected cells. At this point speci®c antibodies could bind to the EBs and facilitate their uptake by phagocytosis. The presence of local (cervical) anti-chlamydial secretory IgA was found to correlate with decreased shedding in women with cervical infections.54 C. trachomatis-derived peptides capable of binding in the groove of MHC class I and II molecules are responsible for the speci®c T-cell activation, and the identi®cation of such peptides may provide additional components of importance for generation of protective immunity. Peptides to be presented in the groove of MHC class II molecules are peptides processed in antigen-presenting cells such as macrophages or dendritic cells. Such peptides are usually 13±25 amino acids long and are produced by degradation within the phagolysosome. Here, they are associated with the MHC class II molecules and transported to the surface of the cell where they activate CD4‡ T cells. Peptides presented by the MHC class I pathway must be processed through the proteasome pathway to combine with the MHC class I molecules. Because the chlamydiae are localized within a phagosome during their intracellular growth such peptides could be derived from secreted proteins such as CPAF.21 Identi®cation of other secreted proteins could have implications for the selection of vaccine candidates. Also, the chlamydial proteins localized within the phagosome membrane20 could be candidates for processing through the proteasome pathway and thus potential vaccine candidates. Will a Chlamydia vaccine be a reality? There is no doubt that the development of a C. trachomatis vaccine is justi®ed and that a great e€ort has been made to develop the background knowledge for such a development. Animal models of a C. trachomatis vaccine are showing promising results, and the animal C. psittaci vaccine has shown good protection.55 The C. psittaci vaccine protects against acute infections. A human vaccine that could protect against a primary infection will be e€ective, but in case it is not fully protective it should be carefully evaluated to determine whether complications such as tubal factor infertility and ectopic pregnancy may still develop. SUMMARY There is today no human C. trachomatis vaccine. Progress has been made to improve our knowledge of both humoral and cellular immunity. During infection antibodies are generated to a number of proteins, of which MOMP and Omp2 are outer membrane proteins and chlamydial GroEL and DnaK are chaperones. MOMP is the best studied molecule and several epitopes have been identi®ed. MOMP is also

898 G. Christiansen and S. Birkelund

involved in the development of the cellular immune response and several epitopes have been identi®ed. Early human vaccine trials with puri®ed inactivated EB showed serovar-speci®c short-lived protection, but in animal models a hypersensitivity reaction probably caused by chlamydial GroEL was observed. To avoid the development of immunopathology an approach to develop a subcomponent vaccine was made. As MOMP was known to be immunogenic several attempts were made to use this molecule in a vaccine. Protection was observed only when MOMP was puri®ed from chlamydiae ± but not when MOMP was produced in recombinant bacteria or produced as synthetic peptide fragments. A new and promising strategy for making a vaccine is to inject naked DNA that will be translated and processed in order to stimulate development of an immune response. A MOMP DNA vaccine has been used in an animal model and both a humoral and a cellular immune response were detected. It is possible to add other genes or gene fragments to the DNA molecule and, in that way, obtain the desired protection. In order to create an optimal vaccine more knowledge should therefore be obtained about which chlamydial proteins are recognized by the cellular immune response and how such proteins are processed. Practice points . until a vaccine against C. trachomatis infections has been developed, screening for and treatment of chlamydial infections are recommended . for screening urine samples PCR or LCR tests could be used, and in order to make the testing cost-e€ective the screening could be done on pooled urine specimens56 . screening of women should focus on young, sexually active women and their partners . to attract a high number of asymptomatic women for screening, home sampling of either urine or swabs could be used57

REFERENCES 1. Pearce JH & Gaston JSH. Chlamydia. In Susmann M (ed.) Molecular Medical Microbiology, pp 1825±1864. San Diego: Academic Press, 2001. 2. Stephens RS, Sanchez-Pescador R, Wagar EA et al. Diversity of Chlamydia trachomatis major outer membrane protein genes. Journal of Bacteriology 1987; 169: 3879±3885. 3. Wang SP & Grayston JT. Human serology in Chlamydia trachomatis infection with microimmuno¯uorescence. Journal of Infectious Diseases 1974; 130: 388±397. 4. Larsen B, Birkelund S, Mordhorst CH et al. The humoral immune response to Chlamydia trachomatis in patients with acute reactive arthritis. British Journal of Rheumatology 1994; 33: 534±540. 5. MaÊrdh PA, Lind I, Svensson L et al. Antibodies to Chlamydia trachomatis, Mycoplasma hominis, and Neisseria gonorrhoeae in sera from patients with acute salpingitis. British Journal of Venereal Disease 1981; 57: 125±129. 6. Westrom L. E€ect of acute pelvic in¯ammatory disease on fertility. American Journal of Obstetrics and Gynecology 1975; 121: 707±713. 7. Eggert-Kruse W, Rohr G, Demirakca T et al. Chlamydial serology in 1303 asymptomatic subfertile couples. Human Reproduction 1997; 12: 1464±1475. 8. Moller BR & MaÊrdh PA. Experimental salpingitis in grivet monkeys by Chlamydia trachomatis. Modes of spread of infection to the Fallopian tubes. Acta Pathologica Microbiologica Scandinavica 1980; 88: 107±114.

Is a Chlamydia vaccine a reality? 899 9. Wang SP & Grayston JT. Human serology in Chlamydia trachomatis infection with microimmuno¯uorescence. Journal of Infectious Diseases 1974; 130: 388±397. 10. Hatch TP. Disul®de cross-linked envelope proteins: the functional equivalent of peptidoglycan in chlamydiae? Journal of Bacteriology 1996; 178: 1±5. 11. Mygind P, Christiansen G & Birkelund S. Topological analysis of Chlamydia trachomatis L2 outer membrane protein 2. Journal of Bacteriology 1998; 180: 5784±5787. 12. Sanchez-Campillo M, Bini L, Comanducci M et al. Identi®cation of immunoreactive proteins of Chlamydia trachomatis by Western blot analysis of a two-dimensional electrophoresis map with patient sera. Electrophoresis 1999; 20: 2269±2279. *13. Morrison RP, Belland RJ, Lyng K & Caldwell HD. Chlamydial disease pathogenesis. The 57-kD chlamydial hypersensitivity antigen is a stress response protein. Journal of Experimental Medicine 1989; 170: 1271±1283. *14. Persson K, Osser S, Birkelund S et al. Antibodies to Chlamydia trachomatis heat shock proteins in women with tubal factor infertility are associated with prior infection by C. trachomatis but not by C. pneumoniae. Human Reproduction 1999; 14: 1969±1973. *15. Loomis WP & Starnbach MN. T cell responses to Chlamydia trachomatis. Current Opinion in Microbiology 2002; 5: 87±91. *16. Kim SK & DeMars R. Epitope clusters in the major outer membrane protein of Chlamydia trachomatis. Current Opinion in Immunology 2001; 13: 429±436. 17. Goodall JC, Beacock-Sharp H, Deane KH & Gaston JS. Recognition of the 60 kilodalton cysteine-rich outer membrane protein OMP2 by CD4(‡) T cells from humans infected with Chlamydia trachomatis. Clinical and Experimental Immunology 2001; 126: 488±493. 18. Goodall JC, Yeo G, Huang M et al. Identi®cation of Chlamydia trachomatis antigens recognized by human CD4‡ T lymphocytes by screening an expression library. European Journal of Immunology 2001; 31: 1513±1522. 19. Kinnunen A, Paavonen J & Surcel HM. Heat shock protein 60 speci®c T-cell response in chlamydial infections. Scandinavian Journal of Immunology 2001; 54: 76±81. 20. Scidmore-Carlson MA, Shaw EI, Dooley CA et al. Identi®cation and characterization of a Chlamydia trachomatis early operon encoding four novel inclusion membrane proteins. Molecular Microbiology 1999; 33: 753±765. 21. Zhong G, Fan P, Ji H et al. Identi®cation of a chlamydial protease-like activity factor responsible for the degradation of host transcription factors. Journal of Experimental Medicine 2001; 193: 935±942. 22. Stephens RS, Kalman S, Lammel C et al. Genome sequence of an obligate intracellular pathogen of humans: Chlamydia trachomatis. Science 1998; 282: 754±759. 23. Hsia RC, Pannekoek Y, Ingerowski E & Bavoil PM. Type III secretion genes identify a putative virulence locus of Chlamydia. Molecular Microbiology 1997; 25: 351±359. 24. Kim SK, Devine L, Angevine M et al. Direct detection and magnetic isolation of Chlamydia trachomatis major outer membrane protein-speci®c CD8‡ CTLs with HLA class I tetramers. Journal of Immunology 2000; 165: 7285±7292. *25. Fling SP, Sutherland RA, Steele LN et al. CD8‡ T cells recognize an inclusion membrane-associated protein from the vacuolar pathogen Chlamydia trachomatis. Proceedings of the National Academy of Sciences of the USA 2001; 98: 1160±1165. 26. Dhir SP, Agarwal LP, Detels R et al. Field trial of two bivalent trachoma vaccines in children of Punjab Indian villages. American Journal of Ophthalmology 1967; 63: 1639±1644. *27. Clements C, Dhir SP, Grayston JT & Wang SP. Long term follow-up study of a trachoma vaccine trial in villages of Northern India. American Journal of Ophthalmology 1979; 87: 350±353. *28. Wang, SP, Grayston JT & Alexander ER. Trachoma vaccine studies in monkeys. American Journal of Ophthalmology 1967; 63: 1615±1630. 29. Morrison RP, Lyng K & Caldwell HD. Chlamydial disease pathogenesis. Ocular hypersensitivity elicited by a genus-speci®c 57-kD protein. Journal of Experimental Medicine 1989; 169: 663±675. 30. Patton DL, Sweeney YT & Kuo C-C. Demonstration of delayed hypersensitivity in Chlamydia trachomatis in monkeys: a pathogenic mechanism of tubal damage. Journal of Infectious Diseases 1994; 169: 680±683. 31. Rank RG. Animal models of urogenital infections. Methods in Enzymology 1994; 235: 83±93. 32. van Voorhis WC, Barrett LK, Sweeney YT et al. Repeated Chlamydia trachomatis infection of Macaca nemestrina fallopian tubes produces a Th1-like cytokine response associated with ®brosis and scarring. Infection and Immunity 1997; 65: 2175±2182. 33. Peeling RW, Patton DL, Cosgrove SweeneyYT et al. Antibody response to the chlamydial heat-shock protein 60 in an experimental model of chronic pelvic in¯ammatory disease in monkeys (Macaca nemestrina). Journal of Infectious Diseases 1999; 180: 774±779. 34. Rank RG, Batteiger BE & Soderberg LSF. Susceptibility to reinfection after primary chlamydial genital infection. Infection and Immunity 1988; 56: 2243±2249.

900 G. Christiansen and S. Birkelund 35. Tu€rey M, Falder P, Gale J et al. Infertility in mice infected genitally with a human strain of Chlamydia trachomatis. Journal of Reproduction and Fertility 1986; 78: 251±260. *36. Stagg AJ, Tu€rey M, Woods C et al. Protection against ascending infection of the genital tract by Chlamydia trachomatis is associated with recruitment of major histocompatibility complex class II antigenpresenting cells into uterine tissue. Infection and Immunity 1998; 66: 3535±3544. 37. Ramsey KH, Newhall WJ & Rank RG. Humoral immune response to chlamydial genital infection of mice with the agent of mouse pneumonitis. Infection and Immunity 1989; 57: 2441±2446. 38. Igietseme JU, Uriri IM, Kumar SN et al. Route of infection that induces a high intensity of gamma interferon-secreting T cells in the genital tract produces optimal protection against Chlamydia trachomatis infection in mice. Infection and Immunity 1998; 66: 4030±4035. 39. Tseng CT & Rank RG. Role of NK cells in early host response to chlamydial genital infection. Infection and Immunity 1998; 66: 5867±5875. 40. Moore T, Ananaba GA, Bolier J et al. Fc receptor regulation of protective immunity against Chlamydia trachomatis. Immunology 2002; 105: 213±221. 41. Zhang YX, Stewart S, Joseph T et al. Protective monoclonal antibodies recognize epitopes located on the major outer membrane protein of Chlamydia trachomatis. Journal of Immunology 1987; 138: 575±581. 42. Cotter TW, Meng Q, Shen ZL et al. Protective ecacy of major outer membrane protein-speci®c immunoglobulin A (IgA) and IgG monoclonal antibodies in a murine model of Chlamydia trachomatis genital tract infection. Infection and Immunity 1995; 63: 4704±4714. 43. Su H, Morrison RP, Watkins NG & Caldwell HD. Identi®cation and characterization of T helper cell epitopes of the major outer membrane protein of Chlamydia trachomatis. Journal of Experimental Medicine 1990; 172: 203±212. 44. Ortiz L, Demick KP, Petersen JW et al. Chlamydia trachomatis major outer membrane protein (MOMP) epitopes that activate HLA class II-restricted T cells from infected humans. Journal of Immunology 1996; 157: 4554±4567. 45. Kim SK, Angevine M, Demick K et al. Induction of HLA class I-restricted CD8‡ CTLs speci®c for the major outer membrane protein of Chlamydia trachomatis in human genital tract infections. Journal of Immunology 1999; 162: 6855±6866. 46. Ortiz L, Angevine M, Kim SK et al. T-cell epitopes in variable segments of Chlamydia trachomatis major outer membrane protein elicit serovar-speci®c immune responses in infected humans. Infection and Immunity 2000; 68: 1719±1723. 47. Beagley KW & Timms P. Chlamydia trachomatis infection: incidence, health cost and prospects for vaccine development. Journal of Reproduction and Immunology 2000; 48: 47±68. 48. Pal S, Theodor I, Peterson EM et al. Immunization with the Chlamydia trachomatis mouse pneumonitis major outer membrane protein can elicit a protective immune response against a genital challenge. Infection and Immunity 2001; 69: 6240±6247. *49. Igietseme JU & Murdin A. Induction of protective immunity against Chlamydia trachomatis genital infection by a vaccine based on major outer membrane protein-lipophilic immune response-stimulating complexes. Infection and Immunity 2000; 68: 6798±6806. *50. Zhang DJ, Yang X, Berry J et al. DNA vaccination with major outer-membrane protein gene induces aquired immunity to Chlamydia trachomatis (mouse pneumonitis) infection. Journal of Infectious Diseases 1997; 176: 1035±1040. 51. Zhang DJ, Yang X, Shen C & Brunham RC. Characterization of immune responses following intramuscular DNA immunization with the MOMP gene of Chlamydia trachomatis mouse pneumonitis strain. Immunology 1999; 96: 314±321. 52. Shaw AC, Gevaert K, Demol H et al. Comparative proteome analysis of Chlamydia trachomatis serovar A, D and L2. Proteomics 2002; 2: 164±186. 53. Vandahl BB, Gevaert K, Demol H et al. Time-dependent expression and processing of a hypothetical protein of possible importance for regulation of the Chlamydia pneumoniae developmental cycle. Electrophoresis 2001; 22: 1697±1704. 54. Brunham RC, Kuo C-C, Cles L & Holmen KK. Correlation of host immune response with quantitative recovery of Chlamydia trachomatis from human endocervis. Infection and Immunity 1983; 39: 1491±1494. 55. Entrican G, Buxton D, Longbottom D. Chlamydial infection in sheep: Immune control versus fetal pathology. Journal of the Royal Society of Medicine 2001; 94: 273±277. 56. Moore SA, Welte R & Postma MJ. Major improvements in the cost e€ectiveness of screening women for Chlamydia trachomatis using pooled urine specimens and high performance testing. Sexually Transmitted Infections 2002; 78: 74±75. 57. Ostergaard L, Andersen B, Olesen F & Moller JK. Ecacy of home sampling for screening of Chlamydia trachomatis: randomised study. British Medical Journal 1998; 317: 26±27.