Mapping of Specific and Promiscuous HLA-DR-Restricted T-Cell Epitopes on the Plasmodium falciparum 27-Kilodalton Sexual Stage-Specific Antigen

Abstract

We have characterized HLA-DR-restricted T-cell epitopes on the 27-kDa protein (Pfg27), a sexual stage-specific antigen, of the human malaria parasite Plasmodium falciparum in subjects with a history of malaria. Pfg27, expressed early in the sexual stages, is recognized by monoclonal antibodies capable of reducing the infectivity of gametocytes in mosquitoes. By using 16 Pfg27-specific CD4⁺-T-cell clones derived from three donors, seven different T-cell epitopes were identified. Among them, P11 (amino acids 191 to 210 of the Pfg27 sequence, IDVVDSYIIKPIPALPVTPD) was found to contain a previously described binding motif for multiple HLA-DR allotypes. Indeed, P11 was found to be promiscuous in that it could be recognized by T cells in the context of at least five different HLA-DR molecules. The cytokine profile of the clones was mixed. Seven of nine T-cell clones exhibited a Th0-like cytokine profile, producing high levels of gamma interferon (IFN-γ) and interleukin-4 (IL-4) upon stimulation with specific peptides and mitogens. The other two clones had a Th1-like cytokine profile with high expression of IFN-γ and no IL-4. Identification of a promiscuous epitope in Pfg27 could play a significant role in the design of a subunit vaccine for suppressing malaria transmission.

Malaria parasites have a complex life cycle involving asexual and sexual development in the vertebrate host and the mosquito vector, respectively. The existence of multiple distinct life cycle stages, each with a spectrum of stage-specific antigens (Ags) and unique ways of interaction with the host immune system, has complicated the development of vaccines against human malaria parasites.

In general, the immunological mechanisms involved in protection against malaria infection are still poorly understood. Adults living in areas of endemic infection develop a nonsterilizing form of short-lived immunity to malaria (2, 14, 17). Studies with humans naturally exposed to malaria parasites and those with experimental animal models have shown that both humoral and cellular immune responses are important (5, 52, 53, 55). Target Ags defined in various life cycle stages of malaria parasites have been the focus of intense immunological investigation (36). In many of these studies, CD4⁺ T lymphocytes have been identified as immune effector cells occupying a central role. CD4⁺ T lymphocytes recognize Ags as peptides associated with major histocompatibility complex (MHC) class II molecules and promote both antibody and cell-mediated immune responses. Incorporation into vaccines of parasite-derived T-helper epitopes that bind to multiple class II MHC molecules would be expected to elicit humoral and cellular immunity in individuals from diverse genetic backgrounds.

Studies in our laboratory have focused on immunity against Plasmodium falciparum sexual stages, also known as malaria transmission-blocking immunity (TBI). Erythrocytic female and male sexual stages (gametocytes) are responsible for transmission of malaria parasite from the vertebrate host to mosquitoes. Gametocytes ingested by the mosquito during the blood meal develop within the midgut into extracellular male and female gametes, which rapidly undergo fertilization followed by further development. Several Ags produced in the sexual stages (gametocytes and gametes: Pfs230 and Pfs48/45) (42, 44, 57) and others expressed only in mosquito midgut stages (zygotes and ookinetes: Pfs25 and Pfs28) (19, 24, 57) have been identified as potential targets of antibody (Ab)-mediated TBI. While Abs elicited against identified targets of TBI are capable of preventing gamete fertilization and oocyst development in the gut of the mosquito vector, thus suppressing malaria transmission (8), the specific role of T cells in the development of what are presumably T-cell-dependent Ab has not been characterized. It is well established that T cells of the helper phenotype are required for growth, differentiation, and immunoglobulin G isotype switching and secretion by B cells.

The studies described in this paper were aimed at gaining a better understanding of human T-cell responses against epitopes in the 27-kDa gametocyte protein (Pfg27), which has been identified in more recent studies from our laboratory as yet another potential target of TBI (41, 59). Pfg27 is a highly abundant protein which is expressed from the onset of sexual differentiation of P. falciparum and throughout gametocyte maturation (1, 9, 29). Murine monoclonal Abs (MAbs) recognizing a reduction-insensitive (continuous) epitope in Pfg27 are effective blockers of P. falciparum development in mosquitoes. These transmission-blocking MAbs recognize a 15-amino-acid region in the recombinant Pfg27 (rPfg27) that has been shown to be conserved in several P. falciparum isolates and also cross-react with the reduced and denatured forms of two other targets of TBI, Pfs230 and Pfs48/45, in immunoprecipitation (41, 59).

We derived human CD4⁺-T-cell clones specific to Pfg27 from three subjects previously exposed to P. falciparum. Characterization of Pfg27-specific T-cell responses included epitope mapping and analysis of MHC restriction and cytokine production by the T-cell clones. A putative promiscuous T-cell epitope in Pfg27, P11, which is recognized by T cells in the context of multiple HLA-DR alleles was identified. In recent years, DR-restricted peptides that exhibit promiscuity at the level of binding to MHC class II molecules have been defined in proteins of diverse infectious agents as well as in self proteins (7, 18, 30, 32, 39, 50). This is the first report, however, on identification of such a promiscuous epitope in an Ag which is specifically expressed in the sexual stages of P. falciparum and is recognized by MAb capable of suppressing the infectivity of parasites in mosquitoes.

MATERIALS AND METHODS

Blood donors.

Healthy volunteers who had lived in countries with endemic malaria infection and had been exposed to malaria parasites several times in the past, as well as nonexposed volunteers, were included in this study. The blood samples were collected after informed consent was obtained in accordance with the Johns Hopkins School of Hygiene and Public Health Committee on Human Research. Blood was collected in heparin and immediately processed for isolation of peripheral blood mononuclear cells (PBMC) by centrifugation through Ficoll-Hypaque (Pharmacia Fine Chemicals, Uppsala, Sweden). HLA typing of blood donors was performed at the Institute of Immunogenetics (Johns Hopkins University). Donor cells were HLA class I typed serologically by the immunomagnetic method (56) and HLA class II typed genomically by hybridization of sequence-specific oligonucleotide probes to PCR-amplified DNA (25).

Parasites.

P. falciparum non-gametocyte-producing (clone HB2; Honduras) and gametocyte-producing (isolate NF54; Amsterdam airport, probably from Africa) strains were cultured in vitro with O⁺ erythrocytes (RBC) (54). Asexual P. falciparum (HB2) synchronous cultures were produced by treatment with sorbitol (28), and concentrated schizont-infected RBC were purified by gelatin flotation (22). Mature gametocytes were obtained by in vitro culture as previously described (21). Female gametes and zygotes were purified from cultured mature gametocytes after induction of gametogenesis in vitro (35) with Nycodenz (Nycomed UK Ltd., Birmingham, England) discontinuous gradients (6, 11, and 16%) (42, 57).

Expression and purification of recombinant proteins.

The gene encoding Pfg27 (651 bp) and various overlapping fragments were obtained by PCR. The full-length Pfg27 gene and fragments F2a, F2b, and F2c were expressed as fusion proteins containing an 81-amino-acid sequence derived from the nonstructural protein (NS-1) of influenza A virus in Escherichia coli AR58 by using the pSK F301 (pMG1) expression vector (41). Fragments F1, F1a, and F2 cloned in the pRSET expression vector (Invitrogen) were expressed as His₆-tagged fusion proteins in E. coli BL21 (DE3) (41) and purified by nickel column affinity chromatography. rPfg27 was also expressed, without any fusion sequence, by deletion of the sequence encoding the fusion partner NS-1 (41) and purified by immunoaffinity chromatography with a Pfg27-reactive MAb (6B6) covalently linked to protein A-Sepharose 4B beads (27, 48). Induced bacterial lysate expressing rPfg27 was centrifuged at 16,000 × g for 30 min at 4°C, and the supernatant was applied to the 6B6 affinity column. The column was washed in 50 bed volumes of NETT (0.15 M NaCl, 0.005 M EDTA, 0.05 M Tris, 0.5% Triton X-100 [pH 7.4]) followed by 50 bed volumes of NET (NETT without Triton X-100). Bound protein was eluted with 0.1 M triethylamine (pH 11.5). Fractions (1 ml) were collected and neutralized by adding 50 μl of 1 M NaH₂PO₄ (pH 2.5). The absorbance at 280 nm was measured, and the fractions containing the bulk of the eluted protein were pooled, dialyzed overnight at room temperature against phosphate-buffered saline (pH 7.4), and sterilized by passage through a 0.22-μm-pore-size filter, and protein concentration was determined by Bradford’s method (4). The purity of the protein was assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (15% polyacrylamide) and Western blotting.

Synthetic peptides.

Peptides spanning the Pfg27 amino acid sequence were synthesized on a Milligen/Biosearch 9050 automated peptide synthesizer by the solid-phase peptide technique with Fmoc (9-fluorenylmethoxycarbonyl) chemistry (31). The peptides were generally 20 amino acids long with a 10-amino-acid overlap. They were shown to be 85 to 95% pure by high-pressure liquid chromatography analysis.

Establishment of T-cell lines.

PBMC were washed three times in Hanks balanced salt solution (GIBCO-BRL) and resuspended in complete medium (CM) consisting of RPMI 1640 medium (GIBCO) supplemented with 4 mM l-glutamine, 50 U of penicillin per ml, 50 μg of streptomycin per ml, 5 mM HEPES, and 10% heat-inactivated pooled AB human serum (Interstate Blood Bank, Inc., Memphis, Tenn.). Adherent and nonadherent cells were isolated as previously described (51). Briefly, PBMC were plated at 2 × 10⁶ cells/ml in 25 ml of CM in 100-mm glass petri dishes for 4 to 12 h at 37°C (95% humidity and 5% CO₂). After incubation, the nonadherent cell fraction was removed and the plate was gently rinsed three times with warm CM to remove additional nonadherent cells. The adherent-cell fraction containing monocytes was then harvested by scraping with cold Ca²⁺- and Mg²⁺-free Hanks balanced salt solution supplemented with 2% heat-inactivated fetal bovine serum (Sigma Chemical Co, St. Louis, Mo.), centrifuged at 400 × g for 15 min, and adjusted to 10⁷ cells/ml in CM. The adherent cells were pulsed overnight with 10 μg of rPfg27 per ml at 37°C, washed once to remove unbound Ag, incubated with the autologous nonadherent cells at a ratio of 1:10 in 24-well plates at 4 × 10⁶ cells/well in 2 ml of CM, and incubated for 7 days at 37°C under 5% CO₂. On day 7, an aliquot of culture was assayed for Ag-specific proliferative responses.

T-cell cloning.

rPfg27-specific T-cell clones were isolated by limiting-dilution cloning as previously described (37). Briefly, after two rounds of in vitro stimulation with autologous monocytes pulsed with 10 μg of rPfg27 per ml, responder T cells were cloned by limiting dilution by plating (five 96-well plates for each dilution) cells at a density of 1, 3, or 10 cell(s)/well with irradiated (5,000 R) allogeneic PBMC (50,000/well) in 96-well U-bottom microtiter plates. In some cases, the cloning was repeated by plating cells at 0.3, 1, and 3 cells/well. The medium (T-cell medium [TM]) consisted of CM supplemented with 50 U of recombinant human interleukin-2 (rh1L-2) (Perkin-Elmer, Norwalk, Conn.) and 0.25 μg of phytohemagglutinin (PHA) (GIBCO) per ml. The clones were fed with fresh TM on day 5, and every 3 to 4 days thereafter. Growing clones were isolated from plates seeded with cell doses that produced growth in fewer than 10 to 15% of the wells. Positive clones were transferred to single wells of a 24-well plate and restimulated in 2 ml of TM containing 0.25 μg of PHA per ml and 10⁶ irradiated feeder cells (allogeneic PBMC). After 2 to 4 days in culture, most of the medium was removed from the wells and replaced with TM without PHA. At 15 to 20 days after this restimulation, the clones were tested for proliferation in the presence of rPfg27 as described below. T-cell clones were maintained in TM by restimulation every 7 to 15 days. T-cell clones were analyzed for cell surface phenotype by the direct immunofluorescence microtechnique (11) on a fluorescence-activated cell sorter (FACS II; Becton Dickinson & Co., Sunnyvale, Calif.) with fluorescein isothiocyanate-conjugated mouse MAb to human CD4 or phycoerythrin-conjugated mouse MAb to human CD8 and fluorescein isothiocyanate- and phycoerythrin-conjugated isotype-matched negative control antibodies (CALTAG Lab.).

Proliferation assays.

Ag-specific proliferative responses were measured by culturing PBMC (2 × 10⁵/well) at 37°C in triplicate wells in 96-well flat-bottom microtiter plates in a final volume of 200 μl of CM alone or CM containing any of the following: rPfg27 (5 and 10 μg/ml), a sonicated lysate of purified gametes (4 × 10⁶/ml), a sonicated lysate of asexual stages (4 × 10⁶/ml), tetanus toxoid (10 μg/ml) (Wyeth Lab. Inc., Marietta, Pa.), PHA (1 μg/ml), or nonparasitized RBC (4 × 10⁶/ml). On day 5, 1 μCi of [³H]thymidine (Amersham Corp., Arlington Heights, Ill.) was added to each well in 50 μl of CM and the cells were harvested 16 h later onto fiberglass filters with a cell harvester (Packard, Filtermate 196); incorporation of [³H]thymidine was determined with a Packard Matrix 96 direct β counter. T-cell clones were assayed for Ag-specific proliferation by using either irradiated adherent cells (2 × 10⁴/well) or irradiated autologous PBMC (1 × 10⁵ to 2 × 10⁵/well) as Ag-presenting cells (APCs). T-cell clones that had last been restimulated 15 to 20 days earlier were washed three times with Hanks balanced salt solution and plated at 5 × 10⁴ cells/well in 100 μl of CM. Various Ags were used at the following concentrations: rPfg27, 10 μg/ml; purified recombinant fragments (F1, F1a, and F2), 10 μg/ml; bacterial lysates of fragments (F2a, F2b, and F2c), 10⁷ bacteria/ml, incubated for 30 min in a boiling-water bath and sterilized by gamma irradiation at 40,000 R (27); and synthetic peptides, 10 μg/ml. T-cell proliferation was checked microscopically, 1 μCi of [³H]thymidine was added to each well after incubation for 24 h, and the cells were processed as above.

EBV transformation of B cells.

Epstein-Barr virus (EBV)-transformed B-lymphoblastoid cell lines (EBV-B-cell lines) were established as previously described (40). Briefly, 5 × 10⁶ PBMC were resuspended in 2 ml of CM supplemented with 100 U of rhIL-2, 20 μg of PHA, and 2 ml of filtered supernatant from B95-8 marmoset cells infected with EBV (ATCC CRL 1612; a kind gift of M. Pedrotti-Krueger, Genetic Resources Core Facility Cell Culture Laboratory, Johns Hopkins University). The cells were maintained at 37°C in 95% humidity and 5% CO₂ and fed every week for 3 weeks by removing half of the supernatant and replenishing with an equal volume of CM. After 20 to 30 days, the cells were transferred to 25-cm² flasks containing 3 ml of fresh CM and maintained at 10⁶ cells/ml. EBV-B-cell lines were expanded for at least 2 months before being tested as APCs.

Determination of the MHC restriction patterns of the T-cell clones.

MHC class II molecules involved in Ag presentation to T-cell clones were identified by three different approaches: inhibition assays with a blocking Ab (anti-HLA-DR monomorphic, ATCC L243; a kind gift of Mary Leffel, Institute of Immunogenetics, Johns Hopkins University), and stimulation in the presence of HLA-typed reference EBV-B-cell lines (ASHI Repository, Johns Hopkins University) or transfected murine L cells (a kind gift of Robert W. Karr, G. D. Searle Co., St. Louis, Mo.) expressing the desired human class II molecules (26) as APCs. Briefly, T-cell clones (2 × 10⁴ cells) were cocultured (0.2 ml) with irradiated (7,000 R) autologous EBV-B cells (2 × 10⁴ cells) as APC in the presence or absence of the specific peptide (10 μg/ml). The inhibition assays were performed by preincubation of irradiated autologous EBV-B-cell lines in 96-well U-bottom microtiter plates with the anti-DR MAb for 2 h. When transfected murine L cells were used as APCs, cells (10⁴ cells) were irradiated (4,000 R) and preincubated for 24 h in the presence (10 μg/ml) or absence of the specific peptide, T-cell clones (2 × 10⁴ cells) were added, and proliferative responses were measured as described above.

Analysis of cytokines.

T-cell clones (10⁶ cells/ml in a 24-well plate) were stimulated with mitogens (1 μg of phorbol myristate acetate per ml and 1 μg of ionomycin per ml) without APCs or with specific peptides (10 μg/ml) or an unrelated nonspecific peptide (10 μg/ml) in the presence of irradiated APCs for 18 h. Secreted cytokines were analyzed in culture supernatants by an enzyme-linked immunosorbent assay with kits supplied by Endogen (gamma interferon [IFN-γ], IL-4, and IL-2) or Pharmingen (IL-10).

Cytokine transcript levels in T-cell clones were measured by minor modifications of a semiquantitative technique described previously (23). Briefly, RNA was isolated from pelleted cells by using RNAzol (Tel-Test "B) as described previously (10) and stored at −75°C until use. A 1-μg portion of total RNA was reverse transcribed with Moloney murine leukemia virus reverse transcriptase (GIBCO) in a total volume of 25 μl. The reaction mixture was then diluted 1:8, and 10 μl was used for specific amplification of the reverse-transcribed cytokine mRNA with Taq DNA polymerase (Boehringer Mannheim). The PCR product was separated on 1.2% agarose gels and transferred to Hybond N⁺ membranes (Amersham International) by standard blotting techniques. Southern transfers were subsequently probed with internal cytokine-specific oligonucleotides and visualized with the enhanced chemiluminescence detection system (Amersham International). Autoradiographs were analyzed by laser scanning densitometry (Molecular Dynamics, Sunnyvale, Calif.).

The PCR conditions were strictly defined for each cytokine such that a log-linear relationship was obtained between the amount of specific cytokine mRNA and the signal intensity of the probed PCR product in the detection system throughout the range of specific cytokine mRNA levels in the samples evaluated. For quantitation and corroboration that sample amplification was in the log-linear range for all samples, diluted standard samples (PMA- and ionomycin-stimulated PBMC) were amplified in each PCR run. To control for the relative amounts of total mRNA transcribed in each reverse transcription (RT) reaction, the RT products used for the analysis of specific cytokine mRNA levels were also analyzed in parallel to assess the amount of mRNA in each sample for the constitutively expressed housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH). For reliable comparison of cytokine mRNA amounts, all samples were tested simultaneously for a given cytokine.

PCR primers pairs were chosen to span at least one intron. The nucleotide sequences for sense and antisense primers and probes, respectively, were as follows: IFN-γ, GGACAATCTCTTCCCACCC, AGCTCTGAGACAATGAACGC, and GATTTTCATGTCACCATCCTTTTGCCAG; IL-2, GAGCCTTATGTGTTGTAAGC, CTTGCCCAAGCAGGCCACAG, and GCTTTGAGTCAAATCCAGAACATGCCCAG; IL-4, CATGGTGGCTCAGTACTACG, TCTTTCTCGAATGTACCAGG, and GACCTCGTTCAAAATGCCGATGATCTCTCT; IL-5, ACTCTTGCAGGTAGTCTAGG, CCACAGAAATTCCCACAAGTG, and CACCAACTGTGCACTGAA; IL-10, GAGTCCAGCAGACTCAATAC, CAAACAAAGGACCAGCTGGAC, and ATCACTCTTCACCTGCTCCACTGCCTTGCTC; and GAPDH, CTCAGTGTAGCCAGGATGC, ACCACCATGGAGAAGGTGG, and GTGGAAGGACTCATGACCACAGTCCATGCC. The numbers of PCR cycles used were 23 for GAPDH, 22 for IFN-γ, 28 for IL-2, 35 for IL-4 and IL-5, and 40 for IL-10.

RESULTS

Selection of donors for T-cell cloning.

PBMC from 17 individuals with and 8 individuals without a history of malaria were screened for proliferation against several malaria Ag including immunoaffinity-purified rPfg27 (Table 1). The rPfg27 induced strong proliferation (stimulation index [SI] > 4) in PBMC from 83% of P. falciparum-exposed donors (SI range, 1.3 to 33), while PBMC from only one of eight individuals with no history of malaria showed a proliferative response (SI = 5.3) to rPfg27 (SI range, 0.8 to 5.3). Interestingly, PBMC from four of five P. vivax-exposed donors also showed a strong proliferative response to rPfg27 (SI range, 2 to 12.6).

TABLE 1.

PBMC lymphoproliferation in response to rPfg27 (5 μg/ml)

Donora	Sex	Malaria statusb	SIc
1	M	P. falciparum (>30 times)	33
2	M	P. falciparum (last, 1986)	8
3	F	P. falciparum (>10 times)	16
4	M	P. falciparum (>10 times)	15
5	M	P. falciparum (>5 times)	15
6	M	P. falciparum (last, 1985)	5.5
7	F	P. falciparum (last, 1991)	6.6
8	M	P. falciparum (last, 1992)	4
9	M	P. falciparum/P. vivax	1.3
10	M	P. vivax (current)	4.9
11	M	P. vivax (current)	4
12	M	P. vivax (current)	2
13	M	P. vivax (current)	12.6
14	M	P. vivax (last, 1978)	7.3
15	F	P. falciparum (1989)	5.8
16	F	P. falciparum (1991)	2
17	F	P. falciparum (1994)	6.4
18	M	?	1.2
19	M	?	2.2
20	M	?	3.3
21	M	Nonexposed	5.3
22	M	Nonexposed	1.5
23	F	Nonexposed	0.8
24	F	Nonexposed	1.2
25	F	Nonexposed	1.3

^aT-cell clone donors (1, K; 2, CH; 3, A). 

^bThe blood donors were adult residents of Africa (donors 1 to 9), Venezuela (donors 10 to 13; thick smear positive at the time of blood collection and treated afterward), and India (donor 14). Donors 15 to 17 were Caucasians who had lived in Africa for 2 to 3 years and had been back in the United States for more than 2 years. Donors 18 to 20 were North Americans who travelled to areas of endemic malaria but did not report any malaria symptoms. Donors 21 to 25 were North Americans who had never travelled to any areas of endemic malaria infection. Malaria status recorded for various donors was based on personal recollection of any malaria infections in the past. 

^cSI, stimulation index, expressed as counts in antigen-stimulated cultures/counts in unstimulated cultures. 

Generation of Pfg27-specific T-cell clones.

Pfg27-specific T cells were produced with PBMC from three volunteers chosen based on their proliferative response to rPfg27 and the generous willingness of the volunteers to repeatedly donate the blood required to produce and maintain the clones. Figure 1 shows proliferative responses of PBMC of the three donors (donors 1 [K], 2 [CH], and 3 [A] in Table 1) to the rPfg27 (5 to 10 μg/ml), gamete Ag from P. falciparum NF54, and schizont Ag from P. falciparum HB2 tested before the initiation of T-cell cloning.

FIG. 1.

Proliferative responses of PBMC from donors K, CH, and A to P. falciparum Ag. PBMC were cultured for 5 days in the presence of either rPfg27 (5 and 10 μg/ml) or different preparations of Ag. The cells were then pulsed with [³H]thymidine, and the radioactivity (mean total counts ± standard deviations [SD]) was measured. Med, medium; RBC, erythrocytes; Gam, gametes; Asex, asexual stages.

Nonadherent cells from the donors were stimulated twice in vitro with rPfg27-pulsed autologous adherent cells before being cloned by limiting dilution in the presence of rhIL-2, PHA, and irradiated allogeneic PBMC as feeder cells. Enrichment of rPfg27-specific cells was demonstrated by the strong lymphoproliferation at the time of cloning (SI > 49 [donors K and A tested]). Clones obtained from these cultures (90 to 95 clones from each donor) were tested for Ag-specific proliferation by using rPfg27-pulsed and unpulsed irradiated autologous adherent cells as APCs. Of all tested clones, 22, 19, and 9 clones from donors K, CH, and A, respectively, showed high proliferative response to rPfg27. The phenotype of all Pfg27-specific T-cell clones was uniformly CD4⁺ CD8⁻ as determined by flow cytometric analysis.

T-cell epitope mapping of rPfg27.

Epitopes recognized by the various T-cell clones were mapped in a two-step process. Initially, overlapping Pfg27 recombinant fragments expressed in E. coli (Fig. 2A) were used in lymphoproliferation assays. The pattern of positive responses to these fragments guided the synthesis of peptides (Fig. 2B) used to map the epitopes to specific regions in Pfg27.

FIG. 2.

Schematic representation of rPfg27 and location of recombinant overlapping fragments (A) and synthetic peptides P1 to P20 (B). The numbers above each bar for the fragments represent amino acid residues.

(i) K clones.

Initial screening of seven Pfg27-specific T-cell clones from donor K with the rPfg27 fragments indicated the presence of at least five distinct T-cell epitopes. Three of the clones (K-31, K-46, and K-89) responded to fragment F1 (amino acid residues 1 to 102 in the Pfg27 sequence). Fine mapping of the epitope recognized by clones K-31 and K-46 (nonreactive with F1a), using synthetic peptides, showed a strong response to P8 (residues 71 to 90) (Fig. 3A). The epitope recognized by clone K-89 (reactive to F1a) was mapped to P4 (residues 31 to 50) (Fig. 3A). The other four clones (K-4, K-9, K-12, and K-64) responded to fragment F2 (residues 74 to 217). Two of these (K-4 and K-9) responded to F2b (residues 125 to 180), and further analysis localized the epitope recognized by K-4 to an overlapping region of P17 (residues 141 to 160) and P18 (residues 151 to 170) and that recognized by K-9 to P18 (Fig. 3C). The response of clone K-12 to fragments F2a and F2b indicated that the epitope was within their overlapping sequence. This was confirmed by reactivity of the clone with P9 and weakly with P16 (residues 125 to 140) (Fig. 3B). Clone K-64 responded to P13 (residues 91 to 110), which is within fragment F2a (Fig. 3B).

FIG. 3.

Epitope mapping of Pfg27-specific T-cell clones from donor K (A to C), donor CH (D), and donor A (C and D) by using rPfg27 fragments and synthetic peptides based on the Pfg27 sequence. T cells were stimulated with rPfg27, various fragments, and synthetic peptides, and proliferation was measured by [³H]thymidine incorporation. Results are expressed as mean total counts of duplicate wells ± SDs. The SDs in each case were always <10% of the means.

(ii) CH clones.

The response of clones CH-3 and CH-95 to the full-length rPfg27, fragment F2, and subfragment F2c (residues 165 to 217) suggested the presence of epitopes in the C-terminal region of Pfg27. Fine epitope mapping of these clones with synthetic peptides spanning fragment F2c localized the epitope recognized by both clones to P11 (residues 191 to 210) (Fig. 3D).

(iii) A clones.

Two reactivity patterns indicating the presence of at least three T-cell epitopes were observed when rPfg27-specific T-cell clones from donor A were screened. All seven A clones responded to fragment F2. In addition, four of the clones, A-25, A-28, A-47, and A-81, responded to fragment F2b and the other three clones responded to fragment F2c. Further studies mapped the epitope recognized by clones A-25, A-28, and A-81 to the P17 and P18 overlapping region, the epitope recognized by clone A-47 to P18 (Fig. 3C), and the epitope recognized by clones, A-41, A-43, and A-48 to P11 (Fig. 3D).

As described above, using 16 Pfg27-specific T-cell clones derived from three donors, we were able to identify six or possibly seven distinct T-cell epitopes scattered throughout the Pfg27 sequence. It appears, however, that such peptide-specific T cells may be present in small numbers. When PBMC from these three donors were tested with the panel of synthetic peptides corresponding to the whole sequence of Pfg27, they showed no significant proliferative response to any of the peptides. PBMC from three other donors (donors 4, 5, and 13 [Table 1]) which initially showed a response to rPfg27, lysates of gamete and schizont, did display a positive response to some of the synthetic peptides. PBMC from donors 4 and 5 showed a strong proliferative response to P10 (SI = 8) and to P15 (SI = 7). In a similar analysis, PBMC from donor 13 proliferated in the presence of P8 (SI = 6) and P10 (SI = 4) (data not shown). It is noteworthy that these additional peptides overlap some of the epitopes mapped with T-cell clones derived from donors K, CH, and A. These results stress the importance of the labor-intensive cloning of T cells for epitope analysis. Most of the epitopes mapped in these studies with T-cell clones could not have been mapped by looking at peptide-driven proliferative responses in the PBMC from which these clones were derived.

Next, we investigated if the peptide-specific T-cell clones would recognize epitopes in the context of the native parasite protein. All but one of the 16 rPfg27-specific T-cell clones showed a strong proliferative response to lysates of gametes (Table 2). As expected, none of the clones reacted with lysates of asexual stage parasites which do not express Pfg27 (29).

TABLE 2.

Proliferative responses of the rPfg27-specific T-cell clones to P. falciparum crude Ag^a

Clone (peptide)	Proliferative response (mean total counts ± SD) to:
	APC	APC + peptide	Gametes	Schizonts
K-4 (P17)	31 ± 35	3,954 ± 410	682 ± 253	17 ± 5
K-9 (P18)	48 ± 11	17,735 ± 599	778 ± 226	67 ± 8
K-12 (P9)	47 ± 5	3,818 ± 275	1,679 ± 574	9 ± 12
K-31 (P8)	76 ± 2	4,068 ± 504	1,820 ± 183	10 ± 11
K-46 (P8)	60 ± 7	2,046 ± 10	459 ± 107	37 ± 30
K-64 (P13)	53 ± 5	3,822 ± 120	271 ± 24	38 ± 29
K-89 (P4)	55 ± 33	861 ± 290	280 ± 104	54 ± 4
CH-3 (P11)	12 ± 16	3,650 ± 179	17 ± 17	5 ± 1
CH-95 (P11)	5 ± 1	1,410 ± 18	1,109 ± 173	11 ± 8
A-35 (P17/18)	18 ± 24	2,958 ± 490	640 ± 93	40 ± 7
A-28 (P11)	23 ± 32	7,319 ± 1,296	1,849 ± 321	7 ± 10
A-41 (P11)	28 ± 25	9,690 ± 2,013	7,063 ± 1,646	15 ± 21
A-43 (P11)	58 ± 12	7,483 ± 69	5,351 ± 445	31 ± 42
A-47 (P18)	17 ± 5	15,585 ± 755	2,174 ± 232	19 ± 7
A-48 (P11)	13 ± 6	2,867 ± 274	1,770 ± 110	9 ± 13
A-81 (P17/18)	35 ± 15	3,977 ± 309	1,137 ± 498	11 ± 16

^arPfg27-specific T-cell clones (2 × 10⁴ cells) were cocultured with APC (10⁵ irradiated autologous PBMC) in duplicate wells (0.2-ml U-bottom 96-well plates) for 24 h in the absence or presence of the specific peptide (P, 10 μg/ml) and different preparations of P. falciparum Ag (gametes or schizonts; 4 × 10⁶/ml). The wells were then pulsed with 1 μCi of [³H]thymidine, and the radioactivity incorporated was measured after an additional 24-h incubation. 

HLA restriction analysis and identification of promiscuous T-cell epitopes.

Details on the HLA-DR typing of the donors used for T-cell cloning and the other reagents used to establish the MHC restriction elements of the various T-cell clones are shown in Table 3.

TABLE 3.

MHC class II DR typing of PBMC from T-cell clone donors and EBV-B-cell lines or transfectant L cells used to define MHC restriction elements

Donor	Serological specificity	DR allelesa	EBV-B cellsb	Transfectantsc
K	DR8	NDf	DR11,8,52	ND
	DR13(6)	ND	DR13,14,52	DRB1*1301
	DR52	ND	DR11,52	DRB3*0101
				DRB3*0202
				DRB3*0301
CH	DR15(2)	DRB1*1503	DR15,11,51,52	DRB1*1501d
			DR15,15	DRB1*1502d
	DR18(3)	DRB1*03	ND	DRB1*0301e
	DR51	DRB5*0101	DR15,11,51,52	DRB5*0101
			DR16,14,51,52
	DR52	DRB3*0101	DR15,11,51,52	DRB3*0101
			DR11,52
			DR16,14,51,52
A	DR15(2)	DRB1*1503	DR15,15	DRB1*1501d
			DR15,6,52	DRB1*1502d
	DR17(3)	DRB1*03011	DR7,17,53,52	DRB1*0301e
	DR51	DRB5*0101	DR15,11,51,52	DRB5*0101
			DR16,14,51,52
	DR52	DRB3*02AB	DR15,11,51,52	DRB3*0101

^aHLA-DRB1 products define serological specificities DR1 to DR18; HLA-DRB3 products define serological specificity DR52; HLA-DRB5 products define serological specificity DR51 (3). 

^bHLA-typed reference EBV-B cell lines. 

^cTransfected L cells expressing the indicated human class II molecules. 

^dDRB1*1501 and DRB*1502-L cells were used to define the participation of DRB1*1503 (DR15) allele as restriction element (all three alleles have structural homologies). 

^eDRB1*0301-L cells were used to define the participation of DRB1*03 (DR18) and DRB1*03011 (DR17) alleles as restriction elements (all three alleles have structural homologies). 

^fND, not done. 

In the first approach, a MAb recognizing the HLA-DR class II Ag was used to investigate if these alleles were involved in the presentation of peptides to various T-cell clones. Peptide-specific proliferation of several clones (K-4, K-46, and CH-95) was inhibited by an anti-DR MAb in a dose-dependent manner (1:2 to 1:64) (Fig. 4A). In contrast, no inhibition of clone K-89 was observed, suggesting that a different allele (DQ or DP) may be working as the restriction element. Figure 4B to D shows inhibition of peptide-specific proliferation of T cells by anti-DR MAb in other T-cell clones. The extent of inhibition was between 49 and 85% in all the tested clones except K-12 and K-89. The specificity of anti-DR MAb inhibition was established by using an unrelated hybridoma supernatant. These studies provided the first evidence that a majority of these clones are restricted by DR alleles.

FIG. 4.

Inhibition of peptide-specific proliferative responses of Pfg27-specific T-cell clones by anti-DR MAb. (A) Autologous irradiated EBV-B cells were preincubated with various concentrations of anti-DR MAb for 2 h before being added to T-cell cultures and specific peptides. (B to D) Inhibition of peptide-specific proliferation of clones from donors K (B), CH (C), and A (D) by anti-DR MAb (1:4) or an unrelated hybridoma supernatant (SN). After addition of T-cell clones and the specific peptide, cells were incubated for 24 h and pulsed with [³H]thymidine and the radioactivity incorporated was measured after an additional 24-h incubation. Results are expressed as mean total counts of duplicate wells ± SDs. The SDs in each case were always <10% of the means.

Alternatively, HLA-typed homozygous EBV-B-cell lines and a panel of murine L-cell transfectants expressing different human class II molecules (26) were used as APCs to identify the MHC class II restriction element. The HLA phenotype of donor K is A68,33; B51,53; Bw4; DR8,13[6],52; DQ6,7 (DP alleles are not routinely defined with the use of serologic reagents). Clone K-4 showed peptide-specific proliferation in the presence of autologous EBV-B cells as well as allogeneic cells expressing DR8, DR13, and DR52 (Fig. 5A). Further analysis with fibroblast transfectants identified DR13 as the restriction element for this clone (Fig. 5B). Clones K-31 and K-46, both recognizing P8, showed proliferative responses in the presence of EBV-B-cell lines expressing DR8, DR13, and DR52 but not to transfectant fibroblasts (DR13-L, DR52-L), suggesting that DR8 is the likely element involved in peptide presentation (data not shown). No DR alleles could be identified either by EBV-B-cell lines or by transfectant fibroblasts for the other two clones K-9 and K-64 (data not shown). A variant DR allele, not included in our panel of APCs, probably represents the restriction element involved in the Ag presentation to these clones. The HLA phenotype of donor CH is A74,34; B17,35; Bw4,6; C4,7; DR15[2],18[3],51,52; DQ6,4. A panel of EBV-B-cell lines indicated DR15 and DR51 as possible restriction elements for CH-3 (Fig. 5C) and DR15, DR51, and DR52 as possible restriction elements for CH-95 (Fig. 5G). The use of class II transfectant fibroblasts further confirmed that P11 is presented to CH-95 by DR15, DR18, DR51, and DR52 (Fig. 5H; Table 3). A similar approach for clones from donor A (HLA phenotype, A74,36; B53; Bw4; DR15[2],17[3],51,52; DQ2,6) identified DR17 as the restriction element for clones A-25, A-28, A-47, and A-81 (Fig. 5E and F) and DR15, DR17, DR51, and DR52 for A-41 (Fig. 5I and J), A-43, and A-48 (Fig. 5K and L).

FIG. 5.

MHC restriction specificity of representative clones. APCs included irradiated autologous or allogeneic EBV-B cells matched at the indicated HLA-DR class II alleles (A, C, E, G, I, and K) or irradiated autologous PBMC or transfected murine L cells expressing the indicated human class II molecules (B, D, F, H, J, and L). Irradiated transfected class II L cells (10⁴ cells) were plated in flat-bottom 96-well plates in the presence of medium alone or the specific peptide and incubated for 24 h. T-cell clones were added to the wells and pulsed 24 h later with [³H]thymidine. The radioactivity incorporated was measured after an additional 24-h incubation. The results are expressed as mean total counts of duplicate wells ± SDs. The SDs were always <10% of the mean.

In summary, four different patterns of MHC responses were observed among Pfg27-specific T-cell clones: (i) clones from different donors that recognized the same peptide through different restriction elements (K-4: P17-18, DR13; A-25, A-28, A-47, and A-81: P17-18, DR17); (ii) clones from different donors (CH-3, CH-95, A-41, A-43, and A-48) that recognized the same peptide (P11) in association with different DR molecules (DR15, DR17, DR51, and DR52), suggesting that it is a promiscuous peptide, i.e., capable of being presented in the context of multiple class II molecules; (iii) clones in which the restriction element could not be identified even though peptide-proliferative responses were inhibited by anti-DR MAb (K-9 and K-64); and (iv) clones in which proliferative responses to specific peptides were not inhibited by anti-DR MAb (K-12 and K-89).

Reciprocal presentation of specific peptides to T-cell clones by APC from donor CH and donor A.

Clone CH-95 from donor CH and clones A-41, A-43, and A-48 from donor A recognize P11 in the context of DR15, DR17, DR18, DR51, and DR52 (Fig. 5G to L). Because of the similarities in the HLA phenotypes of the two donors, we tested if irradiated APC from one donor could present peptides to T-cell clones from the other donor. CH clones displayed proliferative responses to P11 when PBMC from donor A were used as APC. Likewise, A clones showed proliferative responses to P11 and P17-18 when PBMC from donor CH were used as the APC (Table 4).

TABLE 4.

Reciprocal presentation of peptides by allogeneic APCs^a

Clone (peptide)	APC (donor)	Proliferation (cpm)b in response to:
		Medium	Peptide
CH-3 (P11)	A	288	3,822
CH-95 (P11)	A	307	7,838
A-25 (P17–18)	CH	84	1,062
A-28 (P17–18)	CH	152	2,397
A-81 (P17–18)	CH	80	201
A-47 (P18)	CH	50	7,544
A-41 (P11)	CH	59	3,606
A-43 (P11)	CH	34	4,446
A-48 (P11)	CH	23	169

^aT-cell clones (10⁴ cells) from donors CH and A were cocultured with irradiated PBMC from A and CH, respectively, in duplicate wells (0.2-ml U-bottom 96-well plates) for 24 h in the absence or presence of the specific peptides (10 μg/ml). 

^bCell proliferation was measured by thymidine incorporation and is expressed as mean total counts. Standard deviations were always <10%. Proliferation of CH and A clones in the presence of autologous APCs is shown in Fig. 3C and D.  

Cytokine profiles of Pfg27-specific T-cell clones.

T-cell clones obtained from donors K and CH were tested for the analysis of cytokine profiles at the levels of both protein and mRNA after stimulation with specific peptides and mitogens. Large amounts of IFN-γ were secreted by all nine T-cell clones when stimulated with either specific peptides or mitogen (Table 5). Seven of these nine clones also secreted significant amounts of IL-4 in the presence of mitogens and thus were considered to be Th0 like. The other two clones did not produce any detectable IL-4 and thus were Th1-like. Enzyme-linked immunosorbent assay analysis also revealed IL-10 production in all but one (CH-3) clone when stimulated by specific peptides (data not shown). This phenotyping of T-cell clones was further confirmed by RT-PCR. IFN-γ mRNA was detected in all nine clones when stimulated with mitogen (not shown) and in all but one T-cell clone when stimulated with specific peptide (Fig. 6). The RT-PCR results for IL-2, IL-4, and IL-5 in the cells stimulated with specific peptides are also shown in Fig. 6. The expression of IL-2 in all the clones was quite low. The patterns of expression of additional cytokines measured by RT-PCR were generally consistent with the above designations.

TABLE 5.

Production of IFN-γ and IL-4 by Pfg27-specific human T-cell clones

Clone	Cytokine production (pg/ml) after:				Th type
	Peptide stimulation		Mitogen stimulation
	IFN-γ	IL-4	IFN-γ	IL-4
K-4	1,201	20	7,145	37	Th0
K-9	1,469	271	17,140	1,170	Th0
K-12	1,740	0	17,140	4	Th1
K-31	1,740	48	17,140	55	Th0
K-46	1,030	314	4,880	302	Th0
K-64	3,480	0	17,140	0	Th1
K-89	2,795	37	17,140	105	Th0
CH-3	8,990	0	17,140	77	Th0
CH-95	17,140	331	17,140	600	Th0

FIG. 6.

Cytokine analysis by RT-PCR. Various T-cell clones were stimulated with specific peptides, and expression of cytokines was analyzed by RT-PCR. The results are expressed as the ratio of cytokine to GAPDH. The range of IL-2 ratios in T cells stimulated with mitogens was 1.4 to 3.8.

DISCUSSION

Previous studies have shown that MAbs recognizing a linear epitope in Pfg27 are effective blockers of the infectivity of P. falciparum gametocytes in mosquitoes (59). T-helper cell responses to Pfg27 in humans, however, have not been characterized previously. These studies were undertaken to investigate and characterize epitopes recognized by Pfg27-specific T-cell clones from three individuals. It is generally believed that identification of T-cell epitopes capable of eliciting immune responses in individuals of different genetic backgrounds would enhance the efficacy of any subunit vaccine.

Our results indicated that PBMC from 83% of donors with a history of P. falciparum malaria gave strong proliferative responses to rPfg27 while those from only 13% of donors with no history of malaria infection responded. This is in marked contrast to the previously reported high percentage of T cells responding to a crude mixture of gamete Ag (13) or to asexual and gametocyte parasite extracts (15) in individuals with no history of malaria. The results described in this paper also demonstrate the existence of strong T-cell proliferative responses in the three subjects (residents of Nigeria and Ghana) even though these subjects were exposed to malaria parasites 2 to 8 years before the start of this study and none was positive for anti-Pfg27 antibodies by immunoprecipitation analysis (data not shown). Phenotypic characterization of the clones by flow cytometric analysis indicated that all 16 Pfg27-specific T-cell clones belong to the CD4⁺ subset. Although our cloning conditions generated several CD8⁺ clones, none was specific for Pfg27.

Figure 7 summarizes results on mapping of T-cell epitopes for all the T-cell clones produced. One of the significant observations made in this study is that of recognition of certain epitopes (P11, P17, and P18) by several T-cell clones from two different donors. Epitopes within P17 and P18 were shown to be presented in the context of DR17 (DRB1*0301) and DR13 (DRB1*1301). These studies also led to the finding of the recognition of P11 by six clones from two different donors (CH and A) in the context of DR15 (DRB1*1503), DR17 (DRB1*03011), DR18 (DRB1*03), DR51 (DRB5*0101), and DR52 (DRB3*0101/DRB3*02) molecules. Interestingly, two of the peptides, P9 and P11, identified in this study as human T-cell epitopes had also been identified as T-cell epitopes among five previously mapped epitopes recognized by T-cell clones from two different strains of mice (27). We also observed an apparent concentration of T-cell epitopes (five of seven epitopes) in the C-terminal half of the molecule. In contrast, the B-cell epitope recognized by transmission-blocking MAb had earlier been mapped to amino acid residues 10 to 25 in the N-terminal region of Pfg27 (41). Earlier studies have indicated the significance of the polarity of epitopes in immunogens, and thus the relative positions of T-cell and B-cell epitopes within colinear peptide constructs could be of great relevance in the design of synthetic peptide vaccines (12).

FIG. 7.

Summary of epitopes recognized by CD4⁺ Pfg27-specific T-cell clones and HLA restriction elements involved in Ag presentation.

The ability of a peptide epitope to bind to and to be presented in the context of several DR molecules reflects a broad MHC promiscuous binding pattern. A universal T-cell epitope, contained within amino acid residues 378 to 398 of the sequence of the circumsporozoite protein (CS.T3), which binds to most human and murine class II molecules (50), and an epitope within amino acid residues 326 to 345 of the P. falciparum circumsporozoite protein, which binds to several human class II molecules (32), have been previously described. Similarly, bacterial (tetanous toxoid residues, 830 to 843) and viral (hemagglutinin residues, 307 to 319) peptides that are capable of binding in an apparently indiscriminate manner to most DR alleles have been found (6, 38, 39). Inclusion of such promiscuous or universal T-cell epitopes in an engineered subunit vaccine could help overcome genetic restriction and greatly increase the immunogenicity of vaccine constructs. P11 identified in our studies represents first such epitope in an antigen expressed in the sexual stages of P. falciparum.

Studies involving truncation and single-amino-acid substitution, as well as evaluation of direct binding of peptides to soluble MHC molecules or competitive binding analysis, have provided useful information in defining DR-binding motifs in peptides. The specificity of peptide binding in the groove of the MHC class II molecules is determined by the interaction of peptide side chains and the hydrophobic pocket in the floor of the groove. Such analyses have revealed that the N-terminal anchor position (p-1) in a peptide frequently contains an aromatic or hydrophobic residue (W, F, Y, V, L, or I). Other amino acid positions affecting the binding of peptides in the DR groove include a noncharged small amino acid residue at p-6 and another hydrophobic residue at p-9 (6, 32, 38). More recent results, however, obtained from peptides eluted from DR class II molecules have revealed that the aromatic or hydrophobic residue at position 1 represents the motif required to bind most DR molecules (DR1, DR2, DR4, DR5, and DR7) (20, 49). Less prominent roles are also played by other residues (positions 4, 6, 7, and 9) depending on the particular DR allele. The Pfg27 peptide P11 sequence (residues 197 to 205) shares the immunological and structural features of DR-binding motifs previously described in such promiscuous T-cell epitopes: an aromatic residue (Y₁₉₇) at position 1, a small noncharged residue (I₂₀₂) at position 6, and an aliphatic residue (L₂₀₅) at position 9 (Table 6). Our analysis of the presumed promiscuous binding of P11 is based of the actual presentation of epitopes to specific T cells and thus takes into consideration both the binding of peptides to various MHC molecules and the recognition of this MHC-peptide complex by the T-cell receptor. This is important because it represents the first demonstration of an HLA-permissive association of epitopes in malaria sexual stage Ag.

TABLE 6.

T-cell epitopes of viral, bacterial, and parasite proteins with DR-binding motifs

Peptide	Motif at residue:a
	1	2	3	4	5	6	7	8	9
TT (830–843)b	Y	I	K	A	N	S	K	F	I
HA (307–319)b	Y	V	K	Q	N	T	L	K	L
CS.T3 (378–398)b	I	A	K	M	E	K	A		S	S
Pfg27 (197–205)c	Y	I	I	K	P	I	P	A	L

^aDR peptide-binding residues are underlined. 

^bUniversal T-cell epitopes. 

^cPfg27/P11 sequence showing DR-binding motif (residues 197 to 205). 

Another potentially significant finding in these studies was that 15 of 16 clones were shown to proliferate vigorously in response to extracts of P. falciparum sexual stage parasites, suggesting a similar pattern of processing and immune recognition between native parasite proteins and the rPfg27. It is also important to note that Pfg27 sequence is highly conserved in several isolates of P. falciparum examined (41). These observations could prove to have an immunologically significant implication if vaccination with mapped T-cell epitopes to prime memory T-helper cells could be boosted by recall Ag during natural infection. In other studies, very few T-cell clones have been shown to respond to both a synthetic malaria peptide and crude parasite Ag (34, 43).

Cytokine analysis of seven of the nine studied T-cell clones suggested a predominantly Th0 phenotype for Pfg27-specific T-cell clones. The other two T-cell clones displayed a pure Th1-type cytokine pattern, producing only IFN-γ and no IL-4 or IL-5. These findings are consistent with those for the majority of human T-cell clones in the literature, either alloreactive or specific for many other antigens, which mostly display a Th0 phenotype (45, 47). Whether it reflects a natural tendency of human CD4⁺ T cells to undergo less of a polarization in cytokine responses than murine T cells in vivo or is a consequence of the peculiarities of the culture condition used for human T-cell cloning and expansion in vitro remains unclear. For example, the addition of IL-4 during culture favors the clones exhibiting Th0- and Th2-like cytokine profiles. On the other hand, Th0 and Th1 cytokine-producing clones are favored when cultured in the presence of IL-2, a condition used in the present study. In these studies, T cells were stimulated with various agents for 18 h. While a detailed time-kinetic evaluation would have provided a more accurate assessment of Th subtypes, it was not attempted because of difficulties in obtaining sufficiently large numbers of the cloned T cells needed for such analysis. Nonetheless, the finding of prominent production of IFN-γ by Pfg27-specific T-cell clones may be of biological significance. IFN-γ has been shown to play a critical role in both protective and harmful responses to malaria in animal models (46, 58). The role of IFN-γ in immunity to sexual stages of the human malaria parasite is less clear; however, elevated levels of IFN-γ have been reported during infection with P. falciparum and P. vivax (16). Other studies have also suggested inactivation of Plasmodium gametocytes (33) in the presence of IFN-γ and tumor necrosis factor alpha. Since Pfg27 is expressed in the sexual stages persisting during infection for several days, it might provide one possible mechanism leading to increased levels of IFN-γ. Additionally, elevated levels of IFN-γ may contribute to TBI by favoring the production of complement-fixing antibody isotypes.

Further characterization of the various T-cell epitopes identified in this study for their ability to prime B lymphocytes for the synthesis of Ab capable of blocking gametocyte infectivity remains to be done. The identification of T-cell epitopes recognized in the context of multiple HLA-DR (present study) as well as murine strains of different haplotypes (27) represents a significant finding. Such promiscuous T-cell epitopes may play a major role in the design of a synthetic malaria vaccine capable of reducing the infectivity of gametocytes in the mosquitoes and thus interrupting malaria transmission.