Role of circumsporozoite protein-specific T-cells in protective immunity against Plasmodium berghei.

Role of circumsporozoite protein-specific T-cells in protective immunity against Plasmodium berghei U. Krzyc h 12 K. White, 12 H.T. Link,1'2 J. Sadoff,3 D. Lanar,2 W.R. BalIou.4 T. Jareed,2 & A. Agarwal3 The Plasmodium berghei circumsporozoite (CS) protein-specific T-cell repertoire was analysed in C57BI/6 (H-2b), Balb/c (H-2d), and C3H/HeN (H-2k) mice immunized with irradiated sporozoites and the proliferative responses were correlated with the protective status of each strain. Splenic lymphocytes responded to the priming antigen, but the responses varied according to both murine strain and immunization schedule. Analysis of cytolytic T lymphocyte (CTL) responses in mice immunized with irradiated P. berghei sporozo- ites or with a Salmonella-recombinant CS protein construct revealed that each immunization induces CTLs recognizing different epitopes on the target cells. The variations in immune reactivities among different murine strains to the CS protein antigens and the variations in the responses to the authentic versus recombinantly expressed CS protein suggest that distinct immune mechanisms may be involved in rendering immune protection. Furthermore, the molecular context of the immunizing antigen may influence the outcome of the fine specificity of T cells involved in immune protection. Introduction Complete immune protection against infection with Plasmodium berghei sporozoites can be induced by immunization with irradiated sporozoites of the homologous strain (1, 2), and partially by oral inocul- ation of the Salmonella-P. berghei recombinant CS protein construct (3, 4). Studies have shown that the circumsporozoite (CS) protein contains the immuno- dominant B-cell determinant(s) localized within the central repeat region of the CS protein (5, 6). Immune protection is mediated in part by anti-CS antibodies, since anti-CS monoclonal antibodies can transfer pro- tection (7, 8). Cell-mediated immunity has also been shown to confer protection as demonsfrated in vivo by depletion of selected T-cell subpopulations (9, 10). Despite these observations, the protective immune mechanism is incompletely understood, particularly concerning the role of CS protein-specific T-cells arising from immunization with irradiated sporo- zoites. Investigations of cellular immune processes occurring during the protective phase will permit a better understanding of anti-sporozoite immunity ne- cessary for vaccine development. In the present study we examined different mouse strains capable of gener- 1 Department of Biology, Catholic University of America, Washing- ton, DC 20064, USA. 2 Department of Immunology, Walter Reed Army Institute of Re- search, Washington, DC 20307-5100, USA. Correspondence should be sent to Dr U. Krzych at this address. 3 Department of Bacterial Diseases, Walter Reed Army Institute of Research, Washington, DC, USA. 4 Current address: Univax, Rockville, MD, USA. ating protective immunity against live sporozoite challenge and analysed the CS protein-specific T-cell proliferative and cytolytic repertoires induced by i.v. immunization with irradiated sporozoite antigens or by oral administration of live attenuated Salmonella- recombinant vaccine expressing P. berghei circum- sporozoite antigens. Materials and methods Mice. Female Balb/c (H-2d), C57BI/6 (H-2b) and male and female C3H/HeN (H-2k) mice were obtained from the Jackson Laboratory (Bar Harbor, ME) and used at 6-12 weeks of age. Sporozoite preparation. Sporozoites dissected from the salivary glands of infected Anopheles stephensi mos- quitos were provided by Dr Imogene Schneider, Dept. of Entomology, Walter Reed Army Institute of Re- search (WRAIR). Upon dissection, sporozoites were placed in Medium 199 (Gibco) containing 3% normal mouse serum (NMS). For immunizations, aliquots of sporozoites were irradiated at 15 000 rad from a cobalt source and stored on ice until the time of immunization. Sporozoite antigen used to stimulate lymphocyte cultures was irradiated at 30000 rad and stored at - 70°C until use. Mosquito saiivary gland antigen. Non-infected mos- quito salivary glands were sham dissected and the debris treated in an identical manner to sporozoites. Gland debris was used at a concentration equivalent to 104 sporozoites/gland. 88 Bulletin of the World Health Organization, 68 (Suppl.): 88-93 (1990) Role of T cells In antl-sporozolte Immunity Construction of Salmonella-P. berghel CS recombinants. The transformation of full length CS gene of P. berghei strain NK65 into Salmonella typhimurium WR4024 trp R- was described in detail previously (4). Immunizations. All mouse strains were immunized in- travenously (i.v.) with 75000 irradiated sporozoites followed by two i.v. boosts of 20 000 irradiated sporo- zoites one week apart. Representative mice were chal- lenged i.v. with 10 000 live sporozoites. For some experiments, separate groups of Balb/c mice were immunized orally with Salmonella-P. berghei CS recombinant constructs as described previously (4). Parasitaemia screening and protection. Individual thin blood smears were taken daily from day 5 through day 14 post challenge and stained with Giemsa. One hundred oil-immersion fields were viewed for para- sitized erythrocytes. Mice were considered protected if parasites were not detected after 14 days. Complete culture media. Cell cultures were maintained on complete media consisting of RPMI 1640 mixed with EHAA Medium (1:1 v/v) (Microbiological Asso- ciates) and enriched with 100 U/ml penicillin (Gibco), 100 pg/ml streptomycin (Gibco), 4 pmol glutamine (Gibco), 2.5 imol sodium pyruvate (Gibco), 10 umol .non-essential amino acids (Gibco), 50 mmol 2-mer- captoethanol (Biorad), and 0.5% NMS (normal mouse serum) or 5% fetal bovine serum (FBS). Proliferative assay. Unfractionated splenic cell popu- lations were cultured in 96-well microtitration plates (Costar) at a final concentration of 5 x 105 cells/ 0.2 ml in complete medium containing sporozoite antigen ranging from 300 to 3000 frozen/thawed/ 30000 rad-irradiated sporozoites, or equivalent amount of mosquito gland debris, or Salmonella-P. berghei CS protein constructs. Control cultures con- tained medium alone. Cells were cultured for 5 days at 37 °C in a humidified atmosphere containing 5% C02, and during the last 16 hours of culture 1 pCi of tritiated thymidine (3H-TdR, NEN, DuPont) was ad- ded to each well to quantitate proliferative activity. Cultures were harvested using an automated sample harvester (LKB Betaplate) and analysed by scintil- lation spectroscopy (LKB Betaplate). Cytolytic T lymphocyte (CTL) assay. Sporozoite- immune spleen cells from Balb/c and C57BI/6 mice and S. typhimurium (WR4204)-P. berghei construct- immune Balb/c mice were analysed for CTL response directed against the CS protein expressed on NIH- 3T3 cells transfected with the P. berghei CS gene or on P815 cells infected with a pseudorabies viral construct containing the P. berghei CS gene (P815-PRV-CSP) WHO Bulletin OMS: Supplement Vol. 68 1990 (gift from Dr Christina Cheng, Syntro, San Diego, CA). Graded numbers of spleen cells were mixed with the 51Cr (100 pCi/tO7 cells) labelled target cells at effector to target (E/T) ratios of 3: 1 to 30: 1 in 96-well plates in 0.2 ml complete medium and incubated for 4 h at 37°C in a 5% CO2 humidified atmosphere. At the end of the incubation period the supernatants were harvested and assayed for 5tCr activity. Naive spleen cells and Salmonella-immune spleen cells were used as effector controls while NIH-3T3 and P815 cells were used as control target cells. The values represent the mean of triplicate wells and the percent- age specific lysis was calculated by the following: [experimental release (cpm)-spontaneous release (cpm)] x 10 [total (cpm)-spontaneous release (cpm)] Transfection of NIH3T3 cells. A 1.7 Kb fragment con- taining the entire 1.4 Kb coding region of the P. berghei CS gene and a 0.3 Kb 5' noncoding se- quence was cloned into the Hind III site of pCDM8 expression vector containing a CMV/T7 promoter region, SV40 origin, splice region, and polyadenyl- ation signal downstream (11). P3-competent bacteria were transformed with the pCDM8-CS vector and screened by colony hybridization to the 32P-TdR labelled 1.4 Kb CS probe. NIH-3T3-fibroblast cells were co-transfected with 10 ug purified pCDM8-CS and 1 pg purified pSV2Neo by calcium phosphate precipitation. Briefly, 5 x 105 NIH-3T3 cells (per 60- mm petri-dish) were incubated with the calcium phos- phate/DNA precipitate for 6 h after which the DNA precipitate was removed and the cells treated with 10% glycerol for 1 min followed by extensive washing with PBS. Cells were incubated in complete medium for 2 days. Transfected cells were selected by culturing in 2 mg/ml G418 (geneticin, Gibco) for 10 days and stable transfectants were maintained in 0.2 mg/ml G418. Transfected cells expressing the CS gene were screened by total transfected cellular RNA northern blot analysis using the 32P-TdR labelled 1.4 Kb CS probe. Transfected clones were obtained by limiting dilution of northern blot-positive cell lines, and were screened for the expression of CS mRNA by northern blot analysis. Pseudorables virusP. berghel CS construct infection of P815 cell. P815 cells were grown in complete medium as described and supplemented with 10% FBS. Twenty-four hours before the CTL assay, 10 x 106 P815 cells/ml were infected with 107 viral plaque- forming units (PFU). The extent of infection was verified microscopically and expression of the CS protein was ascertained by western blot analysis of homogenized cell preparations. 89 U. Krzych et al. CD4+and CD8+ T-cell depletions. Splenic cell popula- tions (20 x 106 cells/ml) were depleted of CD4+ T cells or CD8 + T cells using a 1:3 dilution of culture supernatants of the hybridoma 2RL (MAb 2RL) spe- cific for CD4, or hybridoma 83-12-5 (MAb 83-12-5) specific for CD8 (both gifted by Dr Richard Hodes, NIH) for 1 hour at 4°C. Cells were pelleted and resuspended in Low Tox Rabbit Complement (Cedar Lane Laboratories, Westbury, N.Y.) at a 1:10 final dilution in HBSS containing 1% FBS, and incubated at 37°C for 1 hour. Cells were centrifuged at 400 x g for 10 minutes and washed three times in HBSS. CD4+ and CD8+ enriched populations were resus- pended at 5 x 106 viable cells/ml in complete medium. Results Sporoolt-Indued spleen cells prolifrate to sporozoite antigens in culture. Spleen cells from Balb/c (H-2d), C57B1/6 (H-2b) and C3H/HeN (H-2k) mice, previ- ously immunized i.v. with 75 000 irradiated sporozoi- tes followed by two i.v. boosts of 20000 irradiated sporozoites one week apart, were cultured with irra- diated, frozen/thawed sporozoite antigens. This par- ticular immunization regimen was chosen for analysis since it usually induces protection against a challenge with 10000 live sporozoites (1, 2). Balb/c mice always showed a stronger anti-sporozoite proliferative re- sponse than C57BI/6 strain, whereas C3H/HeN did not respond to sporozoite antigen (Table 1). The vari- ability in the proliferative responses suggested a strain-dependent sensitivity to sporozoite antigens. As shown in the Table, the proliferative response was sporozoite-specific, as graded amounts of mosquito salivary gland debris, obtained from a sham sporo- zoite dissection of uninfected mosquitos, induced only baseline levels of proliferative activity. In an earlier study, lymph node cells from Balb/c and C57Bl/6 mice primed with complete Freund's adjuvant (CFA) did not respond to sporozoite antigen in vitro, demon- strating that the irradiated sporozoites were not mitogenic (12). Moreover, spleen cells from naive animals did not respond to sporozoite antigens in culture (data not shown). DIssocatIon of protection and proliferative T-cell actIvty. To investigate a possible relationship be- tween the proliferative response and protection, Balb/c, C57BI/6, and C3H/HeN mice were immunized with either a single dose of 75 000 irradiated sporozoi- tes or with one or two boosting doses of 20000 irradiated sporozoite following priming. Pooled splenic lymphocytes from each group (3-5 mice/ group) were assayed in vitro for sporozoite-specific proliferative responses in the presence of graded amounts of sporozoite antigen. Additionally, groups of 5 mice from each strain, were challenged i.v. with 10000 live sporozoites and subsequently screened for parasitaemia. Sporozoite-specific responses were consistently observed in cultures of splenic lymphocytes from Balb/c and C57BI/6 strains, although the magnitude of reactivities differed in each strain. A low level of proliferative reactivity to sporozoite antigens was seen in the Balb/c spleens after the priming dose that induced full protection in these mice. In contrast, C57BI/6 mice mounted a high proliferative activity after a priming dose of 75 000 irradiated sporozoites. However, they were not protected until a third dose of 20000 sporozoites, at which time the proliferative activity was diminished to a 3-fold increase over the background responses. Generally, proliferation of sporozoite-immune C3H/HeN splenic lymphocytes was not observed, but protection was induced in most mice in 3 out of 4 experiments using the full 3-dose immunization schedule (Table 2). Interestingly, splenic populations obtained from mice immunized with the Salmonella-P. berghei CS construct failed to proliferate to all malarial antigens tested, including sporozoites, CS protein synthetic peptides, and re- combinantly expressed CS protein (data not shown). Clearly, these data show variations in the sens- itivities of different mouse strains to the immunizing doses of sporozoites eliciting immune protection and proliferative activity. Furthermore, the observed re- verse correlation between immune protection and proliferative responses is not well understood, but it is possible that functionally unique T-cell types may be generated in each strain following a particular immunization regimen, causing the observed fluctuations. Table 1: SporozoHe (SPZ) antigens Induce specifc responses In sporozolte-minune spleen cells No. of sporozoites for In vitro splenic proliferative responses (delta cpm x 103) immunization Mouse strain ( x 103) SPZ antigens Mosquito debris Balb/c 75. 20, 20 17.8 1.9 C57BLJ6 75, 20, 20 9.0 0.7 C3H/HeN 75, 20, 20 1.0 0.5 WHO Bulletin OMS: Supplement Vol. 6819090 Role of T cells In anti-sorozolte Immunity Sporozolte and Salmonella-P. berghel Immunizations In- duce CS protein specific C7Ls. As shown in Table 3, Balb/c mice immunized with irradiated sporozoites generated CTLs specific for 3T3 transfected CS pro- tein targets. CTL activity was detected in both the unfractionated splenic population and to a lesser degree (10%) in the CD8+ enriched population, but not in the CD4+ enriched population; 15% lysis was detected in the CD8 + enriched population using P815-PRV-CSP targets. Interestingly, immunization with Salmonella-P. berghei CS construct only primed for CTLs able to recognize and lyse P815-PRV-CSP and not 3T3-CS protein targets. Once again, the CTL activity was apparent only in the whole spleen and CD8+ enriched populations. In parallel experiments with C57BI/6 mice immunized with irradiated sporo- zoites CTL activity was not detectable (data not shown). Discussion Analysis of the P. berghei sporozoite-specific T-cell repertoire was initiated by investigating whether im- munization with sporozoites induces T cells respon- sive to sporozoite antigen in culture. An earlier report (13) suggested that P. berghei sporozoites are mito- genic for murine T cells. In contrast, in this and other reported studies (12) we found that sporozoite- immunized splenic lymphocytes responded in vitro specifically and in a dose-dependent fashion to anti- gens from sporozoites treated by freeze/thaw and irradiation with 30 000 rad. Upon boosting immunizations with one or two doses of 20000 irradiated sporozoite, Balb/c mice became high proliferative responders, maintaining their protective status, while proliferative responses of C57BI/6 mice diminished to lower levels although they developed protective immunity. It is not entirely clear why the onset of protection is accompanied by depressed lymphoblastic responses, although a de- crease in the IL-2R density on lymphocytes was partly responsible for similar results in other systems (14, 15). In contrast to results obtained with Balb/c and C57BI/6 mice, the nonresponsiveness of C3H/HeN spleen cells to sporozoite antigen was persistent and reproducible in all experiments. The nonreactivity to sporozoite antigens was also observed in A/J mice, Table 2: Strain-dependent requirements for Induction of anti-sporozolte (SPZ) protection and antlgen-speclfic activation of spienic lymphocytes No. of sporozoites for immunization In vitro anti-SPZ proliferative Strain ( x 103) % protected responses (delta CPM x 103) Balbic 75 100 3 ± 0.5 75, 20 100 17 + 2.1 75, 20, 20 100 16 + 2.5 C57B116 75 0 29 ± 4.0 75, 20 40 16 ± 2.2 75, 20, 20 100 9 + 3.5 C3H/HeN 75 0 0.8O 0.3 75, 20 60 0.8 ± 0.2 75, 20, 20 100 1.1+ 0.3 Table 3: Cytolytic activity of sporozolte Immune splenic populations % specific lysis of: In vivo Effector" priming cells P815-PRV-CSP P815 3T3-CS 3T3 SPZ WSPLC 5 1 25 5 SPZ CD4+ 1 1 5 5 SPZ CD8+ 15 1 10 1 WR2024 WSPCL 60 1 10 10 WR2024 CD4+ 5 1 10 10 WR2024 CD8+ 20 1 1 1 C-V WSPCL 0 0 0 0 NAIVE WSPCL 0 0 0 0 8 E/T = 30:1. WHO Bulletin OMS: Supplement Vol. 68 1990 91 U. Krzych et al. and both strains are not identical at all, except for one locus (D) of the major histocompatibility complex. Although some important information is already available about the T-cell involvement in anti-sporo- zoite protective immunity, we still do not have a full understanding about the fine cellular mechanisms involved in immune protection to sporozoites. The observations reported here strongly suggest the in- volvement of multifactorial mechanisms, whereby not only unique T-cell specificities, but also diverse T-cell functions may be induced. For example, CD8 + T cells may suppress proliferative CD4+ T cells in the C3H/HeN and A/J strains, yet they may function as cytolytic T cells in Balb/c mice. The induction of presumably diverse functional T cells in different mouse strains may be triggered by minor antigenic variations among P. berghei sporozoites from the uncloned NK65 P. berghei strain that we used. An important degree of variability in the gene coding for the CS protein has been noted within different strains of many Plasmodia species (16, 17). The molecular context of the CS protein antigen and mouse strain also appeared to affect detectable CTL activity. Whereas the authentic sporozoite-asso- ciated CS protein primed for CTLs predominately capable of recognizing CS protein expressed on PbCS gene transfected 3T3 targets, immunization with Sal- monella CS construct induced cells directed against P815-P. berghei CS protein targets. Protein and total RNA analysis revealed transcription of the CS gene by the target cells, but the portion of the gene product being expressed is not known. The CTL activity was directed against some CS epitope as evidenced by the inability of either primed populations to lyse normal 3T3 or P815 targets. Also naive or Salmonella control-vector primed mice were unable to lyse any of the targets. The CTL activity was detected immedi- ately after dissection of spleens, without any further in vitro stimulation. This supports the idea that these CTLs are not artifacts of culturing techniques but rather are actually present in vivo. It is of interest that in multiple experiments CTL activity was only detec- ted in Balb/c mice. It is conceivable that sporozoite priming of other mouse strains may not induce specific CTLs or that they may require additional in vitro stimulation with the appropriate antigen. Supported by the apparent differences in the sporozoite-specific T-cell sets, these findings under- score the differences in P. berghei sporozoite-specific immune responses generated in different murine strains. Furthermore, these data demonstrate that identical antigens, like CS protein, presented in an authentic form or as a recombinant protein delivered by either bacteria or virus may induce T cells with diverse specificities and functions. We hypothesize that these functionally diverse T-cell specificities arise owing to different processing and presentation path- ways used by each form of the antigen. Acknowledgements We thank Dr Imogene Schneider for providing us with the dissected sporozoites and Dr Jeffrey D. Chulay for many helpful discussions and support. References 1. Nussenzwelg, R. et al. Protective immunity produced by the injection of X-irradiated sporozoites of Plasmo- dium berghei. Nature, 216: 160 (1967). 2. Nussnzwelg, R. et al. Protective immunity produced by the injection of X-irradiated sporozoites of Plasmo- dium berghei. IV. Dose response, specificity and hum- oral immunity. Mil. med., 134 (Suppl.): 1176 (1969). 3. Sadoff, J.C. et al. Salmonella typhimurium vaccine expressing circumsporozoite protein protects against malaria. Science, 240: 336-338 (1988). 4. Aggarwal, A. et al. Salmonella-malaria circumsporo- zoite recombinant vaccines induce specific CD8+ cytotoxic T-cells. J. exp. med., 172: 1083-1090 (1990). 5. Elchinger, D.J. et al. Circumsporozoite protein of Plas- modium berghei: gene cloning and identification of the immunodominant epitopes. Mol. cell biol., 6: 3965-3972 (1986). 6. Weber, J.L. et al. Plasmodium berghei: cloning of the circumsporozoite protein gene. Exp. parasitol., 63: 295 (1987). 7. Egan, J.E. et al. Efficacy of murine malaria sporozoite vaccines: implications for human vaccine develop- ment. Science, 336: 435 (1987). 8. Potocniak, P. et al. Monovalent fragments (Fab) of monoclonal antibodies to a sporozoite surface anti- gen (Pb44) protect mice against malarial infection. J. exp. med., 151: 1504 (1980). 9. Schofield, L. et al. Gamma-interferon, CD8+ T cells and antibodies required for immunity to malaria sporozoites. Nature, 330: 664 (1987). 10. Weiss, W.R. et al. CD8+ T cells (cytotoxic/suppres- sors) are required for protection in mice immunized with malaria sporozoites. Proc. Natl Acad. Sc., USA, 85: 573 (1988). 11. Seed, B. An LFA-3 cDNA encodes a phospholipid- linked membrane protein homologous to Tc receptor CD2. Nature, 329: 840 (1987). 12. Link, H.T. et al. Immune response to Plasmodium bergheisporozoite antigen(s). I. Evaluation of murine T cell repertoire following immunization with irradi- ated sporozoites. Am. j. trop. med, hyg., 43:452 (1990). 13. Ojo-Amalze, E.A. et al. Plasmodium berghei sporozoi- tes are mitogenic for murine T cells, induce interferon, and activate natural killer cells. J. immunol., 133: 1005 (1984). 14. Beltz, L.A. et al. Novel mechanism for Trypansoma cruzi-induced suppression of human lymphocytes: in- 92 WHO Bulletin OMS: Supplement Vol. 68 1990 Role of T cells In antl-sporozolte Immunity hibition of IL-2 receptor expression. J. immunol., 141: 289 (1988). 15. Klerazenbaum, F. et al. Decreased human IL-2 re- ceptor expression due to a protozoan pathogen. Im- munol. today, 10: 129 (1989). 16. Sharma, S. et al. Diversity of circumsporozoite anti- gen genes from two strains of the malarial parasite Plasmodium knowlesi. Science, 229: 779-782 (1985). 17. de Ia Cruz, V.F. et al. Variation among circumsporo- zoite protein genes from rodent malaria. Mol. bio- chem. parasitol., 28: 31 (1988). WHO Bulletin OMS: Supplement Vol. 68 1990 93