Abstract
Vaccination of malaria-naive humans with recombinant RTS,S/AS02, which includes the C-terminus of the circumsporozoite protein (CS), has been shown to induce strong T cell responses to both the whole protein antigen and to peptides from CS. Here we show that strong T cell responses were also observed in a semi-immune population in The Gambia, West Africa. In a Phase I study, 20 adult male volunteers, lifelong residents in a malaria-endemic region, were given three doses of RTS,S/AS02 at 0, 1 and 6 months. Responses to RTS,S, hepatitis B surface antigen and peptides from CS were tested using lymphocyte proliferation, interferon (IFN)-γ production in microcultures, and IFN-γ ex vivo and cultured ELISPOT, before and after vaccination. Cytotoxic responses were tested only after vaccination and none were detected. Before vaccination, the majority of the volunteers (15/20) had detectable responses in at least one of the tests. After vaccination, responses increased in all assays except cytotoxicity. The increase was most marked for proliferation; all donors responded to RTS,S after the third dose and all except one donor responded to at least one peptide after the second or third dose. There was a lack of close association of peptide responses detected by the different assays, although in microcultures IFN-γ responses were found only when proliferative responses were high, and responses by cultured ELISPOT and proliferation were found together more frequently after vaccination. We have therefore identified several peptide-specific T cell responses induced by RTS,S/AS02 which provides a mechanism to investigate potentially protective immune responses in the field.
Keywords: cell-mediated, mmunity, circumsporozoite, ELISPOT, Plasmodium falciparum, vaccine
INTRODUCTION
Malaria is the most important parasitic disease in humans with 1–2 million deaths and up to 500 million cases every year [1]. There is currently no effective vaccine and the development of one is increasingly important in the face of the growing worldwide threat of the disease [2].
Plasmodium falciparum enters the human host as sporozoites from the bite of an infected Anopheles mosquito. Immunization with irradiated sporozoites can protect humans reliably [3] and although such a vaccine strategy is impractical, this led to the identification of the circumsporozoite protein (CS) as a candidate vaccine antigen. A recombinantly produced fusion of CS and hepatitis B surface antigen (HBsAg), called RTS,S, has been developed which is significantly more potent than previous CS formulations [4]. When three doses were administered with the adjuvant system AS02 to malaria-naive volunteers, it protected 6/7 from parasitaemia following homologous challenge [5] and subsequent trials with varying doses gave a mean protection of 41% (95% CI: 22–56%) [6], although the protection was short-lived [7]. It also provided significant protection against natural exposure to naturally circulating strains of P. falciparum[8].
The immune mechanisms underlying protection by RTS,S are unclear. Although high titre antibody responses to the CS tandem-repeat epitopes were found in protected subjects, not every subject with such antibodies was protected [5]. Cellular responses measured by lymphoproliferation or interferon (IFN)-γ production in response to antigenic stimulation were induced, but these did not correlate with protection [5]. Improvement of vaccine efficacy requires the identification of those human responses that correlate with protection, as these could then be used to select and improve vaccination regimes. One important approach is to identify T cell responses that are induced by vaccination.
We report here the immunogenicity results from an open-label Phase 1 safety, tolerability and immunogenicity trial with RTS,S/AS02 in a semi-immune population. A cohort of 20 semi-immune adults received the same formulation as in the previous study of Stoute et al. [5], but 0·5 ml doses were used throughout. The vaccine was found to be safe and induced humoral immunity [9]. Using peripheral blood mononuclear cells (PBMC) we examined T cell responses by proliferation, IFN-γ secretion into culture supernatants, IFN-γ ELISPOT conducted both after 18 h (ex vivo) and after 2 weeks of antigen stimulation in vitro (cultured) and cytotoxicity (CTL) assays. RTS,S/AS02 was found to induce many of these responses above the levels induced by natural exposure to P. falciparum.
METHODS
Subjects
Twenty male volunteers, aged 18–45 years living in the urban village of Bakau, Kombo St Mary, The Gambia, were recruited under a protocol approved by the joint Gambian Government/Medical Research Council (MRC) Ethics Committee. Bakau is an area of low malaria challenge, having about 0–5 infectious bites per person per year [10], most of which occur during or immediately after the rainy season (June–October). Most Gambians had been exposed to hepatitis B virus at an early age [11] until the introduction of effective vaccination against this virus in the mid-1980s. Exclusion criteria included: evidence of chronic disease, a history of splenectomy, a haematocrit of <0·30, blood transfusions or administration of any vaccine 2 weeks prior to or after vaccination. The mean age was 28·4 years, s.d. 8·34. Two subjects were chronic carriers of HBsAg. One subject was withdrawn from the study after the second vaccination because of a previously unrecognized pre-existing clinical condition. One subject withdrew consent before the third vaccination. Thus 20 subjects received the first two vaccinations and 18 the third, but one subject received a tetanus toxoid vaccination 6 days prior to the second vaccination and was thus excluded from the analysis as a protocol violation. There were no significant local or systemic adverse reactions after each vaccine, as has been reported previously [9].
Study design
The trial was an open-label Phase I study of the RTS,S/AS02 vaccine (GlaxoSmithKline Biologicals, Rixensart, Belgium) with no placebo control, conducted from November 1997 to June 1998. A vaccine dose of 0·5 ml, 50 µg RTS,S antigen was administered intramuscularly at 0, 1 and 6 months; 20 ml of venous blood was taken at the time of the first vaccination (day 0) and 2 weeks after the second (V2 + 14) and third (V3 + 14) doses. The same vaccine regimen was used in a study on malaria-naive donors [12] and in an efficacy study in Gambian adults [8]; however, the earlier study of Stoute et al. [5] used an 0·5 ml, 0·5 ml, 0·1 ml dosing regimen. Blood samples were also taken at 0·5, 5·5 and 6·0 months for haematology, parasitaemia, clinical chemistry and humoral antibody measurements. All clinical and cellular immunology assays were performed in our laboratories at MRC, Fajara. Immunological results are presented from the 17 volunteers who received all three vaccinations and did not receive any other vaccine during the trial.
Parasitaemia
Giemsa-stained thick blood smears were examined for 200 high-powered fields at 1000 magnification by two experienced slide readers in all subjects at least six times during the study. One volunteer, 17307, experienced a clinical attack of P. falciparum before dose 1. No other episodes of clinical malaria occurred during the study but three subjects, 17321, 17235 and 17337, showed low-level asymptomatic parasitaemia during the first month of the trial. This pattern of parasitaemia would be expected in the dry, low-transmission season.
Stimulants
A set of eight 15-mer, semi-overlapping peptides corresponding to residues 321–390 of the carboxy terminus of CS in the vaccine strain (3D7) were used (Table 1). These were designed to include previously identified CD4 and CD8 T cell epitopes and covered the Th2R/region II/Th3R/CS.T3 T cell regions of the molecule (reviewed in [13]). In a trial of non-malaria-exposed volunteers vaccinated with RTS,S/AS02, the majority of responses were to these peptides ([12] and A. Lalvani, personal communication). An additional peptide from the TH3R region of the 7G8 variant was used to test for cross-reactive responses. All peptides were purchased from Research Genetics (Birmingham, AL, USA). Following preliminary titrations, peptides were used at a final concentration of 25 µg/ml in all tests. The toxicity of peptides was examined by their effect on purified protein derivative (PPD) responses in ELISPOT assays, but none was found (data not shown).
Table 1.
Sequence of the peptides from CS (strain 3D7, except peptide 20) that were used in all the assays
| Peptide | Amino acids | Sequence | Region of CS |
|---|---|---|---|
| 35 | 328–342 | DKHIKEYLNK IQNSL | TH2R |
| 36 | 333–347 | EYLNKIQNSLSTEWS | |
| 37 | 338–352 | IQNSLSTEWSPCSVT | |
| 16 | 348–362 | PCSVTCGNGIQVRIK | Region II |
| 38 | 363–377 | PGSANKPKDELDYAN | TH3R |
| 39 | 368–382 | KPKDELDYANDIEKK | |
| 20 | 368–382(7G8 strain) | KPKDELDYENDIEKK | |
| 22 | 378–392 | DIEKKICKMEKCSSV | CS.T3 |
| 23 | 383–397 | ICKMEKCSSVFNVVN |
Recombinant RTS,S and HBsAg whole antigens were purified from yeast cultures and used at 10 µg/ml final concentration, provided by GlaxoSmithKline Biologicals. The PPD of Mycobacterium tuberculosis (Staten Seruminstitut, Copenhagen, Denmark) was used as a control antigen at 10 µg/ml final concentration. In cultured ELISPOTS purified phytohaemagglutinin (PHA) (Sigma, Poole, UK) was used as a control stimulant at 10 µg/ml final concentration.
Preparation of PBMC
PBMC were prepared from heparinized whole blood by density gradient separation on Ficoll (Nycomed Pharma AS, Norway). PBMC were then suspended at 4 × 106/ml in RPMI-1640 (Sigma) with 100 U/ml penicillin/streptomycin and 5% heat-inactivated non-immune human AB serum (NHS) (National Blood Bank, UK) (RN5). This suspension was used for the ex vivo ELISPOT assays.
Proliferation assays
Cells were added at 105/well in U-bottomed 96-well tissue culture plates (Falcon B. D., Meylan, France) in 200 µl proliferation medium (αMEM + 10% NHS + 2 mm l-glutamine + 100 U/ml penicillin/streptomycin) and incubated at 37°C in 5% CO2/95% air. After 6 days 100 µl supernatant was removed and replaced with proliferation medium containing 10 µCi/ml [3H]-thymidine. Supernatants were assayed for IFN-γ and interleukin (IL)-5 by enzyme-linked immunosorbent assay (ELISA). Significant concentrations were defined as 3 s.d. above controls and these were >100 pg/ml for IFN-γ and >13·6 pg/ml for IL-5. The cells were harvested 16 h later onto glass fibre filters and [3H]-thymidine incorporation was measured by liquid scintillation in a β-counter. The stimulation index (SI) was defined as the geometric mean cpm of triplicate experimental wells divided by the geometric mean of the triplicate control wells. We defined responders as individuals whose mean log stimulated value was more than 1·96 s.e. higher than their mean log control value. The pooled estimate of the s.d., which did not vary significantly with time-point, obtained from a two-way analysis of variance was 0·58957. Thus an SI of more than 1·95 (95% CI) was treated as a positive response and 3·065 (99·9% CI) was treated as a substantial positive response. χ2 tests were used to compare the proportion of responders with Yates's correction when appropriate.
Ex vivo IFN-γ ELISPOT
Ex vivo IFN-γ ELISPOT assays were performed to detect antigen-specific effector T cells from freshly isolated PBMC, as described previously [12,14]. Briefly, an IFN-γ ELISPOT kit (Mabtech, Nacka, Sweden) was used according to the manufacturer's recommendations using nitrocellulose-backed 96-well plates (MAIP S45, Millipore, Molsheim, France) coated with 15 µg/ml of capture antibody (MoAb 1-DIK). PBMC in RN5, 4 × 105, were added to each well and the plates were incubated for 18 h at 37°C in 5% CO2 in air. They were then washed and a secondary antibody added (1 µg/ml), followed by streptavidin alkaline phosphatase (1 µg/ml) and developing with the ALP substrate (Bio-Rad, CA, USA). After drying, spot-forming units (SFU) were counted by two independent observers by naked eye. Individuals were defined as positive if the test ELISPOT count was significantly greater than the background count, by assuming that under the null hypothesis the test count is a bionomial (n = 0·5) variable, where n is the sum of the test and background counts. To facilitate comparison with other studies the total SFU/106 PBMC were calculated for peptides 35, 37, 16, 38, 20, 22 and 23 (Table 1).
Cultured IFN-γ ELISPOT and cytotoxic lymphocyte assays
Four wells of 106 cells were incubated in 1 ml RPMI-1640 supplemented with 10% human AB serum + 20 mm l-glutamine + 100 iU/ml penicillin/streptomycin (RN10) in the presence of peptides (25 µg/ml) or RTS,S (10 µg/ml). On days 5 and 10, 10 U/ml Lymphocult-T (Biotest, Dreieich, Germany) was added to each culture as a source of IL-2 and on day 10 cells were subcultured if necessary. On day 14, the cultures from each donor were pooled, washed three times with RPMI-1640 without serum and resuspended to 106/ml in RN5. Each ELISPOT well received 105 cells and stimulants at 25 µg/ml for peptides and 10 µg/ml for PHA or proteins. Results were not included in the analysis if the background was above 5000 spots per million.
Cytotoxic lymphocytes were measured after the third vaccination. PBMC were cultured for 14 days in the presence of peptide and recombinant-antigens as for the cultured ELISPOT. Autologous B cells were Epstein-Barr virus (EBV)-transformed and these cell lines were used as targets in a standard 51Cr-release assay, as described previously [15]. Only donors giving a positive cultured ELISPOT response to at least one peptide were tested for CTL using cells from the 14-day cultures as effectors against peptide-pulsed autologous, or HLA-matched, B cells. All donors were HLA typed by molecular sequence specific primer (SSP)-polymerase chain reaction (PCR) method [16].
RESULTS
Proliferation
Before vaccination (Table 2), 5/18 volunteers had positive responses to RTS,S and two of these also responded to HBsAg. Certain peptides were recognized in a few individuals: peptide 37 in one, peptide 38 in one and peptide 20, an allelic variant of the vaccine peptide 39, in three donors. One individual gave strong responses (SI > 3·054) to RTS,S and to both peptides 37 and 20. There were no positive responses to peptides 16, 23, 35 or 36 before vaccination. This limited number of positive responses might be expected in malaria-exposed adults from an area of low malaria transmission.
Table 2.
Proliferation responses induced by vaccination with RTS,S/AS02.
| Day | PPD | RTS,S | HBS | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Day 0 | Responders/total | 16/18 | 5/18 | 2/18 | ||||||
| Gm SI | 26·41 | 1·24 | 0·96 | |||||||
| V2 + 14 | Responders/total | 16/16 | 14/16 | 12/16† | ||||||
| Gm SI | 51·70 | 32·28** | 4·81** | |||||||
| V3 + 14 | Responders/total | 17/17 | 17/17 | 17/17 | ||||||
| Gm SI | 116·28 | 82·72 | 15·03** | |||||||
| 35 ○ | 36 ○ | 37 | 16 | 38 ○ | 39 • | 20 | 22 s | 23 | ||
| Day 0 | Responders/total | 0/18 | 0/18 | 1/18 | 0/18 | 1/18 | ND | 3/18 | 2/18 | 0/18 |
| Gm SI | 0·68 | 0·85 | 1·16 | 0·96 | 1·10 | ND | 1·32 | 1·14 | 0·75 | |
| V2 + 14 | Responders/total | 4/16 | 7/16† | 5/16 | 4/16 | 4/16 | 4/16 | 13/16† | 8/16 | 9/16† |
| Gm SI | 1·21 | 2·37** | 1·82 | 1·31* | 1·40 | 1·50 | 5·43** | 2·68* | 2·05** | |
| V3 + 14 | Responders/total | 2/17 | 10/17 | 4/17 | 3/17 | 2/17 | 3/17 | 9/17 | 10/17 | 9/17 |
| Gm SI | 1·15 | 2·87 | 1·40 | 1·26 | 1·68 | 1·61 | 3·58 | 4·43 | 3·82 |
○ Indicates that adjacent peptides overlap by 10 amino acids; • these peptides are identical except for one amino acid. Geometric mean SI were compared between day 0 and V2 + 14 and V2 + 14 and V3 + 14.
,
indicate P < 0·05, < 0·001. The numbers of responders (responders/total) were compared between day 0 and V2 + 14 and V2 + 14 and V3 + 14 using a χ2 test.
P < 0·05
After vaccination, responses to all the stimulants increased in frequency and magnitude. This was most marked comparing prevaccination levels to those found after the second vaccination (V2 + 14); the mean SI increased significantly for both vaccine antigens and for five peptides (Table 2). Compared to V2 + 14, the responses after the third vaccination (V3 + 14) were higher to HBsAg (P = 0·028, paired t-test) while those to peptides changed little (P = 0·75, paired t-test), indicating limited boosting by the third vaccination. Responses to HBsAg were weaker than those to the whole RTS,S, suggesting that CS epitopes were being recognized in this assay and emphasizing the need to measure specific peptide responses (Table 2).
After vaccination new peptide-specific responses were found in all donors except one. Vaccination also increased the magnitude of the responses; only 2/7 positive responses present before vaccination were substantial (SI > 3·054), while after the second and third vaccinations 36/58 (62%) and 40/52 (77%) reached this level.
IFN-γ responses after vaccination
Cytokine production in proliferative cultures
IFN-γ was me-asured in day 6 supernatants from proliferative cultures before vaccination and after the second and third doses. No stimulant induced significant levels (>100 pg/ml) of IFN-γ before vaccination. After the third dose, 4/17 donors responded to RTS,S and three of these also responded to HBsAg (Fig. 1). Similarly 5/17 donors responded to peptides at this time-point. In two individuals the responses were >1000 pg/ml for RTS,S and certain peptides.
Fig. 1.
Summary of responses detected before and after the third dose of vaccine. Positive responses are indicated: shading and p, proliferation; E, ex vivo ELISPOT; C, cultured ELISPOT; 5, IL-5 in the proliferation supernatant; I, IFN-γ in the supernatant; peptide 39 (Th3R) was not used for day 0 responses. Donor identification numbers are listed in the first column. Antibodies against the tetrapeptide repeat of CSP and HBsAg were measured as detailed previously [9].
In contrast, IL-5 induction to antigens and peptides was less marked (Fig. 1). Before vaccination no significant levels (>13·6 pg/ml) of IL-5 were found and after the third vaccination only 3/16 donors responded, one to RTS,S alone, two to HBsAg and RTS,S and one to several peptides.
Ex vivo ELISPOT responses
The number of responders in the ex vivo ELISPOT assay increased in frequency with vaccination (Fig. 1 and Table 3). This was most marked for the antigens RTS,S and HBS, with almost 50% of the subjects responding after the third dose. However, the size of the positive responses was small (Table 3), averaging 34 specific spots per million PBMC for the peptides and 33 per million for the whole antigens, and not significantly greater than the size of the positive responses prevaccination (P = 0·3358, t-test).
Table 3.
Frequency and size of the ex vivo ELISPOT and cultured ELISPOT responses to vaccination. The results are presented as the number of responders with the percentage of total tested in brackets and the number of spot-forming units/106 cells plated ± s.e.m. To aid comparison with other studies the total mean SFU and responders for peptides 35, 37, 16, 38, 20, 22 and 23 are given in the last two lines
| Ex vivo ELISPOTS | Cultured ELISPOTS | |||||
|---|---|---|---|---|---|---|
| Day | d0 | V2 + 14 | V3 + 14 | d0 | V2 + 14 | V3 + 14 |
| n | 16 | 17 | 17 | 17 | 14 | 17 |
| RTS,S | 2 (13%) | 3 (18%) | 8 (47%) | 3 (18%) | 5 (34%) | 1 (6%) |
| 13·8 ± 1·3 | 21·7 ± 5·2 | 29 ± 5 | 380 ± 148 | 140 ± 28 | 270 | |
| HBS | 0 | 3 (18%) | 8 (47%) | 2 (12%) | 4 (29%) | 0 |
| 23·3 ± 5·2 | 37 ± 8·2 | 145 ± 15 | 130 ± 19 | |||
| 35 | 1 (6%) | 2 (12%) | 3 (18%) | 5 (29%) | 2 (14%) | 7 (41%) |
| 15 | 264 ± 226 | 15 ± 2·5 | 438 ± 198 | 375 ± 165 | 265 ± 46 | |
| 36 | 1 (6%) | 3 (18%) | 2 (12%) | 5 (29%) | 5 (36%) | 9 (53%) |
| 25 | 32·5 ± 14·2 | 56 ± 38 | 826 ± 420 | 262 ± 35 | 266 ± 43 | |
| 37 | 2 (13%) | 3 (18%) | 1 (6%) | 2 (12%) | 2 (14%) | 3 (18%) |
| 36 ± 24 | 24·2 ± 9·3 | 17·5 | 955 ± 795 | 190 ± 60 | 227 ± 63 | |
| 16 | 0 | 0 | 5 (29%) | 4 (24%) | 0 | 0 |
| 40 ± 9·3 | 305 ± 67 | |||||
| 38 | 1 (6%) | 1 (6%) | 2 (12%) | 3 (18%) | 0 | 0 |
| 13 | 18 | 17·5 ± 5 | 360 ± 164 | |||
| 39 | n.d. | n.d. | 2 (12%) | n.d. | 1 (7%) | 1 (6%) |
| 43 ± 27 | 320 | 210 | ||||
| 20 | 2 (13%) | 3 (18%) | 7 (41%) | 6 (35%) | 1 (7%) | 1 (6%) |
| 18·8 ± 1·3 | 48 ± 16 | 40 ± 16 | 288 ± 80 | 370 | 390 | |
| 22 | 0 | 2 (12%) | 2 (12%) | 3 (18%) | 4 (28%) | 7 (41%) |
| 33 ± 20 | 24 ± 11 | 216 ± 26 | 277 ± 82 | 220 ± 48 | ||
| 23 | 1 (6%) | 2 (12%) | 1 (6%) | 2 (11%) | 0 | 0 |
| 23 | 251 ± 240 | 25 | 130 ± 10 | |||
| Total SFU | 105·8 | 638·2 | 179 | 2692 | 1212 | 1102 |
| Total responders | 7 (6%) | 16 (13%) | 21 (18%) | 25 (21%) | 9 (8%) | 18 (16%) |
Excepting peptides 37 and 38, the number of responders increased to all peptides after vaccination (Table 3). Increases were most marked for peptides 16 and 20, with responses being absent or rare (0/16 and 2/16, respectively) before vaccination and present in 5/17 (29%) or 7/17 (41%) after the third vaccination.
Ex vivo ELISPOT response to peptides, expressed as total SFU, were maximal after the second vaccination (Table 3), particularly for peptides 35 and 23. Although the increase in cell numbers was not sustained after the third dose, the frequency of responders remained significantly elevated above background (7/112 versus 21/119, χ2 Yates's correction, P = 0·031). There was, however, a lack of stability of responses measured by this assay at the different times; of the 10 responses that were detected before vaccination, only three were detected later. Similarly, for all 17 positive responses found after the second dose that were tested after the third vaccination, only three remained positive.
Cultured IFN-γ ELISPOT responses
This technique was used to detect antigen/peptide-specific interferon-γ-secreting memory T cells. PBMC were stimulated with pools of peptides for 14 days and then assayed against each of the individual peptides, RTS,S and HBsAg in an ELISPOT assay for IFN-γ.
Before vaccination cultured ELISPOT responses were detected, in at least one donor, to all the peptides and antigens tested (Table 3). The numbers of positive responses to peptides 22, 35, 36 and 37 increased with vaccination, but not those to peptides 16, 20, 23 and 38 or the whole vaccine antigens, so that the total numbers of responders to selected peptides actually decreased between day 0 and after the second vaccination (25/119 versus 9/119, χ2 Yates's correction, P = 0·017). These responses, however, were consistent between the second and third boosts, with 10/14 responses present at V2 + 14 also present at V3 + 14, and with a similar total SFU at both these time-points (Table 3).
There were often responses to peptides when there was no response to RTS,S. This suggests that testing with a large antigen, which includes the peptide sequences, does not reveal all the responses that are present and may be due to inefficient antigen processing by these cells after 14 days of culture. It is likely that these peptides represent epitopes that can be generated in vivo, as not only were they selected as being those recognized most frequently after RTS,S vaccination: they are from CS regions recognized frequently after natural infection. Although many previous authors have used 20-mers (e.g. [17,18]), responses to these 15-mers are common in malaria-exposed Gambian adults (W. W. H. Reece, unpublished results).
Cultured cells from seven donors were examined in CTL assays. In no case was the specific CTL response above 10%, although for one donor a level of 9·7% was found with peptide 37. These results suggest that cytotoxic cells are not induced readily by this vaccine schedule to the peptides tested, a finding in keeping with previous studies [7,12].
HLA typing
All volunteers were HLA typed (results not shown). The most common HLA class I alleles were A30, A2301, A2, A68, B35 and A alleles, while common class II alleles were DR13, DR11, DQ6 and DQ7. This is typical of Gambian populations (A. Jaye, personal communication).
Correlation between assays
The responses after the third vaccination were studied for their correlation with one another, both with respect to correlation between overlapping peptides in the same assays and with respect to correlation between different assays on the same peptides.
The responses to two pairs of overlapping peptides were correlated highly in the proliferation assay, 22 and 23 (R = 0·68, P = 0·002), and 36 and 37 (R = 0·76, P = 0·0003), while peptides 35 and 36 were correlated highly in the cultured ELISPOT assay (R = 0·69, P = 0·002). This indicates that there may be frequently recognized epitopes in the overlapping regions of these peptides. Five of nine donors who responded to peptide 20 after the third dose did not respond to the variant peptide 39 (Fig. 1), indicating that the amino acid substitution at position 368 is frequently important in recognition. Proliferative responses to the whole protein antigens RTS,S and HBS were also correlated strongly (R = 0·75, P = 0·0003). Similar results were found using a Spearman's rank correlation.
There was little correlation between the ex vivo ELISPOT and either proliferation or the cultured ELISPOT (Fig. 2). In contrast, in the cultured ELISPOT assay and proliferation assay, positive responses to individual peptides occurred together frequently (P < 0·0001, χ2 test). However, there was no strong correlation between the magnitude of the responses in these two assays (Fig. 2), and often responses were present in only one assay, e.g. peptide 20 (proliferation > cultured ELISPOT) and peptide 35 (cultured ELISPOT > proliferation), suggesting that these assays measure separate subsets of cells, as has been suggested previously [14]. There was no strong correlation between IFN-γ in the proliferation culture supernatant and the proliferative response, although the former was present only when proliferation was high.
Fig. 2.
A scatter-plot matrix showing that the cultured ELISPOT, ex vivo ELISPOT and proliferation assays are not well associated with each other for any of the antigens tested: RTS,S, HBS or individual peptides.
DISCUSSION
RTS,S/AS02 induces T cell responses against both CS and HBS that are higher than those induced by natural exposure to CS or hepatitis B virus in this area of low endemicity for P. falciparum. T cell-mediated immune responses measured by proliferation, cultured IFN-γ ELISPOT and ex vivo IFN-γ ELISPOT responses were all increased by vaccination, indicating that this vaccine is immunogenic and induces a wide range of immune responses.
The magnitude of the proliferative responses to RTS,S recombinant antigen were extremely high, and comparable to those induced in malaria-naive volunteers [7]. The responses induced by peptides also gave SI > 10 frequently and, in two cases, SI > 100. In contrast, natural exposure rarely induces SI > 5 [14,17]. The responses to the whole vaccine antigens were stable over time, providing little evidence of boosting by natural exposure in this environment. However, the strong association between proliferation to RTS,S and HBsAg suggests that the proliferative response to the whole RTS,S antigen could be primarily measuring responses to the HBsAg portion of the protein, emphasizing the importance of the need to measure peptide specific responses.
Responses were found to peptides spanning variant and invariant regions of the CS and these were more frequent after vaccination. After vaccination, peptide-specific responses were more common in these malaria-exposed donors than were previously found in malaria-naive donors [12]. In 19 malaria-naive donors tested in Oxford there was no detectable reactivity by the ex vivo IFN-γ ELISPOT assays to any of the known variants of the Th2R or Th3R regions of CS (data not shown). It is thus possible that in our malaria-exposed volunteers, RTS,S/AS02 is boosting infrequent, naturally induced memory cells to detectable levels after vaccination. The observation that responses were observed more frequently to the variant peptides, as has been described previously following natural exposure [17], supports this possibility. Responses to the invariant peptides were also increased after vaccination, as were responses to the heterologous peptide 20, and these may contribute the cross-strain immunity induced by RTS,S/AS02 [8].
The relative balance between immediate effector cells and long-lived memory cells [19] is a central issue in vaccine immunology. The14-day culture period, which is too long to measure accurately precursor frequency, was chosen for sensitivity and the results should reflect accurately the frequency of responding volunteers. It is noticeable that the percentage of responders with long-lived memory cells decreased dramatically after the second vaccination. In contrast, immediate effector cells increased both in magnitude of response and frequency of responders. For certain peptides, e.g. 16 and 20, the increase in immediate effector cells occurred at the same time as a fall in memory cells. These results may indicate a change in the balance between these two cell types. It would seem unlikely that a limited dose vaccine, even with a powerful Th1 adjuvant, would drive memory clones to exhaustion, when this does not occur after natural infection. It is possible that the memory clones are absent from peripheral blood after vaccination, or at least below the level of detection, as T cells are known to relocate during and after natural malaria infection [20,21]. Longer-term follow-up of vaccines may distinguish between these possibilities.
IFN-γ measured by ELISA in proliferation culture supernatant was found only when the proliferation was high, and this may indicate a lack of sensitivity in this method. This could also be a contributing factor to the observed poor correlation between ELISPOT and ELISA results. Because it is possible to predict a high IFN-γ response from the proliferative response, ELISA is unlikely to be of extra value in future studies of this vaccine. The responses in the two ELISPOT assays were not associated closely with proliferation. Flanagan and colleagues also found a lack of correlation between these assays in malaria-exposed Gambian donors [14]. A combination of assays thus provides complementary and non-redundant information on the nature of the induced immune response and it is worthwhile to include all three assays for future vaccine evaluation. These results show that T cell-mediated immune responses are induced by RTS,S/AS02 in naturally exposed donors and provides a mechanism to investigate potentially protective responses in the field [8].
Acknowledgments
The authors thank the volunteers of Bakau, The Gambia, for their willing participation in this study. We thank our field assistant, Sambah Baldeh, for his dedication, the laboratory assistants, Lang Bayo, Simon Correa, Kebba Jammeh and Idrissa Sambou, for serology and parasite detection. We thank Dr H. Whittle in The Gambia and Professor B. M. Greenwood of the London School for Hygiene and Tropical Medicine for advice and encouragement. SmithKline Biologicals (Rixensart, Belgium) and the Medical Research Council in London supported these studies. M. Plebanski is a Howard Hughes International Scholar and NH–MRC Senior Fellow.
REFERENCES
- 1.World Health Organization (WHO). State of the world's vaccines and immunisation. Geneva: WHO; 1996. Who report. [Google Scholar]
- 2.Kwiatkowski D, Marsh K. Development of a malaria vaccine. Lancet. 1997;350:1696–701. doi: 10.1016/S0140-6736(97)03256-X. [DOI] [PubMed] [Google Scholar]
- 3.Sinnis P, Nussenweig V. Preventing sporozoite invasion of hepatocytes. In: Hoffman SL, editor. Malaria vaccine development: a multi-immune response approach. Washington: ASM Press; 1996. pp. 15–34. [Google Scholar]
- 4.Gordon DM, McGovern TW, Krzych U, et al. Safety, immunogenicity and efficacy of a recombinantly produced Plasmodium falciparum circumsporozoite protein–hepatitis B surface antigen subunit vaccine. J Infect Dis. 1995;171:1576–85. doi: 10.1093/infdis/171.6.1576. [DOI] [PubMed] [Google Scholar]
- 5.Stoute JA, Slaoui M, Heppner DG, et al. A preliminary evaluation of a recombinant circumsporozoite protein vaccine against Plasmodium falciparum malaria. N Engl J Med. 1997;336:86–91. doi: 10.1056/NEJM199701093360202. [DOI] [PubMed] [Google Scholar]
- 6.Kester KE, McKinney DA, Tornieporth N, et al. Efficacy of recombinant circumsporozoite protein vaccine regimens against experimental Plasmodium falciparum malaria. J Infect Dis. 2001;183:640–7. doi: 10.1086/318534. [DOI] [PubMed] [Google Scholar]
- 7.Stoute JA, Kester KE, Krzych U, et al. Long-term efficacy and immune responses following immunisation with RTS,S malaria vaccine. J Infect Dis. 1998;178:1139–44. doi: 10.1086/515657. [DOI] [PubMed] [Google Scholar]
- 8.Bojang KA, Milligan PJM, Pinder M, et al. Efficacy of RTS,S/AS02 malaria vaccine against Plasmodium falciparum infection in semi-immune adult men in The Gambia: a randomised trial. Lancet. 2001;358:1927–34. doi: 10.1016/S0140-6736(01)06957-4. [DOI] [PubMed] [Google Scholar]
- 9.Doherty JF, Pinder M, Tornieporth N, et al. A phase 1 safety and immunogenicity trial with the candidate malaria vaccine RTS,S/SBAS2 in semi-immune adults in The Gambia. J Am J Trop Med Hyg. 1999;61:865–8. doi: 10.4269/ajtmh.1999.61.865. [DOI] [PubMed] [Google Scholar]
- 10.Lindsay SW, Campbell H, Adiamah JH, Greenwood AM, Bangali JE, Greenwood BM. Malaria in a peri-urban area of The Gambia. Ann Trop Med Parasitol. 1990;84:553–62. doi: 10.1080/00034983.1990.11812510. [DOI] [PubMed] [Google Scholar]
- 11.Whittle HC, Bradley AK, McLauchlan K, et al. Hepatitis B virus infection in two Gambian villages. Lancet. 1983;1:1203–6. doi: 10.1016/s0140-6736(83)92477-7. [DOI] [PubMed] [Google Scholar]
- 12.Lalvani A, Moris P, Voss G, et al. Potent induction of focused Th1-type cellular and humoral immune responses by RTS,S/SBAS2, a recombinant Plasmodium falciparum vaccine. J Infect Dis. 1999;180:1656–64. doi: 10.1086/315074. [DOI] [PubMed] [Google Scholar]
- 13.Hoffman SL, Franke ED, Hollingdale MR, Druilhe P. Attacking the infected hepatocyte. In: Hoffman SL, editor. Malaria vaccine development: a multi-immune response approach. Washington: ASM Press; 1996. pp. 35–76. [Google Scholar]
- 14.Flanagan KL, Lee EA, Gravenor MB, et al. Unique T cell effector functions elicited by Plasmodium falciparum epitopes in malaria-exposed Africans tested by three T cell assays. J Immunol. 2001;167:4729–37. doi: 10.4049/jimmunol.167.8.4729. [DOI] [PubMed] [Google Scholar]
- 15.Jaye A, Magnusen AF, Whittle HC. Human leucocyte antigen class I- and class II-restricted cytotoxic T lymphocyte responses to measles antigens in immune adults. J Infect Dis. 1998;177:1282–9. doi: 10.1086/515271. [DOI] [PubMed] [Google Scholar]
- 16.Bunce M, O'Neill CM, Barnardo MC, et al. Phototyping: comprehensive DNA typing for HLA-A, B, C, DRB1, DRB3, DBR4, DRB5 & DQB1 by PCR with 144 primer mixes utilizing sequence-specific primers (PCR-SSP) Tissue Antigens. 1995;46:355–67. doi: 10.1111/j.1399-0039.1995.tb03127.x. [DOI] [PubMed] [Google Scholar]
- 17.Good MF, Pombo D, Quakyi IA, et al. Human T-cell recognition of the circumsporozoite protein of Plasmodium falciparum: immunodominant T-cell domains map to the polymorphic regions of the molecule. Proc Natl Acad Sci USA. 1988;85:1199–203. doi: 10.1073/pnas.85.4.1199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Riley EM, Allen SJ, Bennett S, et al. Recognition of dominant T cell-stimulating epitopes from the circumsporozoite protein of Plasmodium falciparum and relationship to malaria morbidity in Gambian children. Trans Roy Soc Trop Med Hyg. 1990;84:648–57. doi: 10.1016/0035-9203(90)90133-y. [DOI] [PubMed] [Google Scholar]
- 19.Geginat J, Sallusto F, Lanzavecchia A. Cytokine-driven proliferation and differentiation of human naïve, central memory and effector memory CD4 (+) T cells. J Exp Med. 2001;194:1711–9. doi: 10.1084/jem.194.12.1711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Elhassan IM, Hviid L, Satti G, et al. Evidence of endothelial inflammation, T cell activation and T cell relocation in uncomplicated Plasmodium falciparum malaria. Am J Trop Med Hyg. 1994;51:372–9. doi: 10.4269/ajtmh.1994.51.372. [DOI] [PubMed] [Google Scholar]
- 21.Helmby H, Johnson G, Troye-Blomberg M. Cellular changes and apoptosis in the spleens and peripheral blood of mice infected with blood-stage Plasmodium chabaudi chabaudi AS. Infect Immun. 2000;68:1485–90. doi: 10.1128/iai.68.3.1485-1490.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]