Early Gamma Interferon and Interleukin-2 Responses to Vaccination Pred
http://www.100md.com
《感染与免疫杂志》
Kenya Medical Research Institute, Centre for Geographical Medical Research (Coast), P.O. Box 230, Kilifi, Kenya
Centre for Clinical Vaccinology and Tropical Medicine, University of Oxford, Oxford, OX3 7LJ United Kingdom
Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford, OX3 7BN United Kingdom
Nuffield Department of Clinical Medicine, Oxford University, John Radcliffe Hospital, Oxford, United Kingdom
Naval Medical Research Center, 503 Robert Grant Avenue, Silver Spring, Maryland
GlaxoSmithKline Biologicals, Rixensart, Belgium
ABSTRACT
Two different cell populations respond to potent T-cell-inducing vaccinations. The induction and loss of effector cells can be seen using an ex vivo enzyme-linked immunospot (ELISPOT) assay, but the more durable resting memory response is demonstrable by a cultured ELISPOT assay. The relationship of the early effector response to durable resting memory is incompletely understood. Effector phenotype is usually identified by gamma interferon (IFN-) production, but interleukin-2 (IL-2) has been specifically linked to the differentiation of memory cells. Here, IFN-- and IL-2-secreting effector cells were identified by an ex vivo ELISPOT assay 1 week after vaccination and compared with the resting memory responses detected by a cultured ELISPOT assay 3 months later. The different kinetics and induction of IL-2 by different vaccines and natural exposure are described. Furthermore, both early IFN- and IL-2 production independently predicted subsequent memory responses at 3 months in malaria-nave volunteers, but only IFN- predicted memory in malaria-exposed volunteers. However, dual ELISPOT assays were also performed on malaria-exposed volunteers to identify cells producing both cytokines simultaneously. This demonstrated that double-cytokine-producing cells were highly predictive of memory. This assay may be useful in predicting vaccinations most likely to generate stable, long-term memory responses.
INTRODUCTION
Experimental studies with mice (7), field studies with humans (16), and irradiated sporozoite immunization (15) suggest that T-cell responses to pre-erythrocytic malaria antigens are protective. A heterologous prime-boost strategy that induces T cells has been developed (23), using sequential immunization with different vectors delivering a common pre-erythrocytic malaria antigen. Two antigens have been used, i.e., ME-TRAP (multiple pre-erythrocytic stage epitopes joined with the whole pre-erythrocytic stage antigen TRAP) and CS (the circumsporozoite antigen with one additional P. falciparum epitope). Each antigen can be delivered by two different viral vectors, either the attenuated fowlpox strain FP9 or modified virus Ankara (MVA). In previous studies of ME-TRAP-encoding vaccinations in nonimmune volunteers, some individuals were fully protected, and many immunized groups showed mean delays in time to parasitemia after controlled bites from Plasmodium falciparum-infected mosquitoes (31).
An alternative approach to vaccination against malaria uses a particulate protein vaccine with adjuvant (RTS,S, expressing the CS antigen). This vaccination is protective in sporozoite challenge (27) and also potently induces T-cell responses (20). In further studies, it has been given in combination with a viral vector encoding CS (9).
Vaccination appears to induce two different populations of T cells. There is a primary expansion and contraction of effector T cells over several weeks and a subsequent differentiation of memory T cells (18). The initial expansion and contraction of effector T cells following prime-boost vaccination can be followed using the ex vivo enzyme-linked immunospot (ELISPOT) assay (23). This assay detects peptide-specific gamma interferon (IFN-)-producing T cells immediately following isolation during 18 h of incubation. However, a resting memory population that acquires the effector phenotype (i.e., IFN- production) after a second exposure to antigen can be identified by a cultured ELISPOT assay (19). This assay employs a 10-day culture period, initially with antigen and then with exogenous interleukin-2 (IL-2).
The cell populations identified by these two assays are different. A study of naturally acquired immunity to the CS protein demonstrated no correlation between cultured ELISPOT and ex vivo ELISPOT assay results (11). Furthermore, the cells identified by the cultured ELISPOT assay persist for at least 6 months after vaccination of malaria-nave subjects, despite waning of cells detected by the ex vivo ELISPOT assay (19). Any reactive T cell persisting beyond the initial expansion and contraction of effector cells could be described as a memory cell, but those cells requiring 10 days of culture to acquire antigen-specific IFN- secretion are referred to here as resting memory cells. This description does not directly relate to the division between central and effector memory, which refers to cell surface markers and homing patterns (26). However, experiments that selectively depleted CCR7-positive T cells before culture suggest that the cultured ELISPOT assay does detect central memory cells (14). In previous studies, we have shown that resting memory cells are expanded following vaccination in malaria-na ve subjects (19) and malaria-exposed subjects (2) and are not simply due to prior exposure.
Mean effector responses among vaccination groups predicted protection from sporozoite challenge, determined by time to parasitemia (31), but did not predict the variability in protection at an individual level. However, cultured ELISPOT assays to detect resting memory responses were recently shown to predict the individual variability in protection (19), and following vaccination with RTS,S/AS02, the cultured ELISPOT assay correlated with protection (25, 28).
Experimental data are contradictory on the relationship of initial effector cell expansion with memory populations. It is not clear whether memory cells differentiate immediately on stimulation with antigen or whether they differentiate from a subset of effector cells (18). Experiments on adoptive transfer in mice suggest that effector differentiation is not required (21), whereas other work suggests these cell populations are not distinct subsets (32). Several lines of investigation link IL-2 to memory cell differentiation. In vitro work suggests that memory cells require IL-2 for proliferation (5), but IL-2 signaling can lead to either proliferation or apoptosis (6), and the effects of IL-2 may vary according to the presence or absence of antigen (12). Exogenous IL-2 promoted resting memory in mice but accelerated cell death when given during expansion (3). In human immunodeficiency virus (HIV)-infected humans, exogenous IL-2 produces a sustained rise in CD4 count (1). Furthermore, although memory is critical to the long-term efficacy of vaccinations in the field, little is known about IL-2 production following immunization.
This study aimed to characterize the early effector response to vaccination in terms of both IL-2 and IFN- production (measured at 1 week) and to compare this with durable resting memory responses (measured at 3 months) in order to determine whether either IL-2 or IFN- production (or both) would predict memory. Phase 1 studies of candidate malaria vaccines were conducted in an area where malaria is not endemic and then in an area where it is endemic. Immunogenicity is described in detail elsewhere (2, 9). Resting memory cell responses were significantly greater after vaccination in malaria-nave (19) and malaria-exposed (2) subjects. The current manuscript describes the additional studies of IL-2 production and memory responses.
MATERIALS AND METHODS
Vaccination was not performed in a blind manner. Subjects were recruited in two locations, from the Kenya Medical Research Institute (KEMRI), Centre for Geographical Medicine Research (Coast), located in Kilifi, Kenya, an area where malaria is endemic, and from Oxford, United Kingdom, an area where malaria is not endemic. Data from 43 subjects are presented here (details of vaccination regimens given in Table 1). Ethical approval was obtained from the Kenyan Medical Research Council (KEMRI) National Ethics Committee, and OXTREC, the Oxford University Tropical Ethics Committee for studies in Kenya. The studies in Oxford were approved by both the Oxfordshire Research Ethics Committee and the Naval Medical Research Center Institutional Review Board and the U.S. Navy Surgeon General (in accordance with U.S. Navy regulations, SECNAVINST 3900.39B). Exclusion criteria included significant medical illness and HIV seropositivity.
Vaccines. Two antigen inserts were used. These were ME-TRAP, encoding the pre-erythrocytic antigen TRAP (thrombospondin-related adhesion protein) and CS (the circumsporozoite protein). TRAP is coupled to a multiple-epitope (ME) string (13). The ME string contains 14 major histocompatibility complex class I preerythrocytic epitopes, three class II epitopes (from Mycobacterium bovis BCG, tetanus toxin, and the circumsporozoite antigen), two preerythrocytic B-cell epitopes, and pb9 (a Plasmodium berghei T-cell epitope that allows preclinical potency and stability testing). CS is coupled to the P. falciparum class I epitope ls6 (from liver-stage antigen 1) and pb9.
DNA CS was manufactured and supplied by Vical Inc., San Diego, CA, and has been called the P. falciparum circumsporozoite protein (PfCSP) DNA vaccine and VCL-2510 elsewhere (10). It is full length and not coupled to ls6 or pb9. DNA ME-TRAP was manufactured by the contract manufacturer QIAGEN, Germany.
Each insert was also used in one of two viral vectors, either in the attenuated fowlpox virus (FP9) or in modified vaccinia Ankara (30, 31). Recombinant vaccine stock was supplied to the contract manufacturer IDT (Rosslau, Germany), who produced clinical lots under good manufacturing practice conditions.
The adjuvant particulate protein vaccine RTS,S/AS02 (GlaxoSmithKline Biologicals, Rixensart, Belgium) comprises the carboxyl-terminal (amino acids 207 to 395) of the 3D7 circumsporozoite protein fused to the hepatitis B surface antigen, coexpressed in yeast with the nonfused hepatitis B surface antigen. The proprietary adjuvant AS02A is composed of an oil-in-water emulsion and the two immunostimulants QS21 and monophosphoryl lipid A (27).
The different vaccine regimens used are shown in Table 1. The medical supervision of vaccination, follow-up, and adverse events is described elsewhere (2). DNA was given as a 4-mg dose (intramuscular). FP9 ME-TRAP, FP9 CS, and MVA ME-TRAP were given as 1 x 108 PFU. MVA CS was given as 1.5 x 108 PFU, and RTS,S/AS02 was given as 0.5 ml (intramuscular).
Malaria-nave volunteers. One group of malaria-nave volunteers received a priming vaccination with DNA, followed by boosting with modified vaccinia Ankara, encoding either one of two pre-erythrocytic malaria antigens (CS or ME-TRAP). Vaccinations were given 4 weeks apart.
The second group received regimens using RTS,S/AS02. This vaccine was used in two similar regimens for malaria-nave volunteers. RTS,S/AS02 was given twice, with MVA encoding CS antigen either before or after the two RTS,S/AS02 immunizations (9). Vaccinations were given 4 weeks apart.
Malaria-exposed volunteers. The malaria-exposed volunteers studied here had received two regimens using viral vectors. These were a prime-boost CS-delivering regimen, using the FP9 virus to prime and MVA to boost, or alternating-vector immunizations (i.e., sequential vaccinations with FP9, MVA, and then FP9 again) delivering ME-TRAP. Vaccinations were given 3 weeks apart for ME-TRAP and 4 weeks apart for CS.
ELISPOT assays. Peripheral blood mononuclear cells (PBMCs) were isolated from blood taken 7 days after each immunization to study the acute response (by IL-2 and ex vivo IFN-), with the exception of volunteers immunized with RTS,S/AS02-containing regimens for which PBMCs were available only 21 days after immunization. Cultured ELISPOT assays were conducted on PBMCs taken 3 months and 9 months after the final immunization. Previous studies indicate that CD8 responses are infrequent and that most responses are CD4 T cells (20, 24), and cell separations were not performed for the current study.
IFN- ELISPOT assays. IFN- ELISPOT assays used Millipore MAIP S45 plates and MabTech antibodies according to the manufacturer's instructions. A total of 4 x 105 of freshly isolated PBMCs per well were incubated in 100 μl RPMI (Sigma-Aldrich) with 10 μg ml–1 peptides for 18 to 20 h before developing. 20-mers overlapping by 10 were used for TRAP, and 15-mers overlapping by 10 were used for CS. Both TRAP and CS peptides were divided into three pools. For TRAP peptides, this was a pool of peptides from the N-terminal region, a pool from the C-terminal region, and a central pool. CS peptides were grouped in N-terminal, C-terminal, and TH2R/TH3R regions. Phytohemagglutinin (20 μg/ml) (Sigma-Aldrich) was used as a positive control, and PBMCs cultured in medium alone was used as negative control. Spot-forming cell numbers were counted with an ELISPOT plate reader (Autoimmun Diagnostika; version 3.0).
Cultured ELISPOT assays. For cultured ELISPOT assays, 1 x 106 PBMCs in 0.5 ml were incubated with 10 μg/ml/peptide of all T9/96 strain TRAP peptides or all 3D7 strain CS peptides in a 24-well plate. On day 3 and day 7, 250 μl of culture supernatant was replaced with 250 μl of 10 IU/ml recombinant IL-2. On day 9, the cells were washed three times and left overnight before the standard ELISPOT assay. Responses were presented by giving the starting cell numbers, rather than the postculture cell numbers, as for previous studies (19, 25).
PBMCs were incubated in RPMI (Sigma-Aldrich) with 10% human AB serum. All ex vivo assays were performed on fresh cells. For malaria-nave volunteers, cultured assays were performed on cells cryopreserved in 10% dimethyl sulfoxide-fetal calf serum and thawed with 25 U/ml benzonase nuclease (Novagen, Darmstadt, Germany) and RPMI with 10% fetal calf serum (Sigma-Aldrich).
The cells identified by cultured ELISPOT assay are referred to as "resting memory" cells throughout the Results and Discussion sections. These cells were identified by IFN- production after stimulation. (They did not produce detectable IL-2, even in response to phytohemagglutinin.)
IL-2 ELISPOT assays were conducted as for IFN-, but using a biotinylated polyclonal and monoclonal mouse anti-human IL-2 antibody pair (BD Pharmingen).
In dual-color ELISPOT assays, each well was coated with IL-2 and IFN- capture antibodies, incubated, and developed as for IL-2 ELISPOT assay using alkaline phosphatase-conjugated streptavidin. The wells were then incubated with RPMI to block biotin binding sites, and developed for IFN- using horseradish peroxidase-conjugated streptavidin.
Positive controls were developed from wells coated with IL-2 capture antibody only, IFN- capture antibody only, or both. This allowed comparison of spots in sample wells with spots of known color. Cells producing IL-2 only were light blue. Cells producing IFN- only were light red, and dual-cytokine-producing cells were dark purple. Developing plates briefly with the alkaline phosphatase substrate (4 min) but longer with the horseradish peroxidase substrate (10 to 15 min) produced the greatest discrimination between the dark purple and light blue spots. Wells were read in a blind manner (the reader did not know the results of cultured ELISPOT assays) and confirmed by two separate observers. Results varying by more than 10% were reread a third time, and the two closer readings were accepted.
Analysis. ELISPOT assay wells were assayed in duplicate. The mean was taken, and the negative-control well result was subtracted from each peptide well. Assays were considered failures if the positive-control well gave less than 150 spots per well or if the negative-control well gave more than 20 spots per well for ex vivo ELISPOT assays or more than 40 spots per well for cultured ELISPOT assays. Pools were summed to calculate total responses. Spot numbers given are per million PBMCs throughout.
Data were log transformed to normalize distributions. In examining for either cross-reactivity or relating resting memory and effector responses, results were paired according to the region of the CS/TRAP tested and volunteer. Each individual may contribute two or three points. This can lead to an overestimate of significance due to the linked nature of observations. To account for this, the cluster subcommand of Stata 8 (Stata Corp.) was used to adjust P values upwards.
Grouping of vaccination regimens for analysis. When the immunological responses and vaccination regimens were very similar, samples were pooled for analysis. DDM regimens had used two different antigen inserts (ME-TRAP and CS). However, immunogenicities were similar (S. J. Dunachie et al., unpublished data), and both regimens were given to malaria-nave volunteers. Samples from patients given these two regimens were therefore grouped for analysis. The regimens that combined RTS,S/AS02 and MVA CS also gave very similar immunogenicities (9), and both were given to malaria-nave volunteers, so samples were pooled for analysis.
In malaria-exposed volunteers, alternating-vector regimens (i.e., FMF and MFM) and prime-boost regimens (FM and FFM) were used. These two groups used different inserts and had previously been shown to induce quite different memory and immediate effector responses (2) so they were analyzed separately.
RESULTS
Ex vivo IL-2 production was analyzed by malaria exposure, and the magnitude and kinetics of vaccine-induced responses were compared (Fig. 1). IFN- responses are shown at key time points for comparison with IL-2 responses, but the full kinetics of IFN- production are described elsewhere (2, 9, 31). PBMCs were not available for IL-2 studies 7 days after RRM/MRR vaccinations. The samples from four subjects at baseline, two subjects at 14 days after DDM, and four subjects at 90 days after RRM/MRR had assay results with unsatisfactory positive or negative-control results.
IL-2 production at baseline differs with malaria exposure. PBMCs from malaria-nave, unimmunized subjects showed very little ex vivo IFN- or IL-2 production in response to malaria antigen (CS/TRAP). Malaria-exposed subjects had low ex vivo IFN- production and despite a few individuals with strong responses, the overall geometric mean was similar to that from malaria-nave subjects (P = 0.24). However, ex vivo IL-2 responses were significantly higher among the malaria-exposed population (P = 0.007).
IL-2 responses after viral vector immunizations are short-lived. In malaria-nave volunteers, prime-boost vaccination regimes induced both ex vivo IFN- (as previously described) and ex vivo IL-2 production (P = 0.0047). IL-2 production at 3 months was similar to baseline levels (P = 0.1).
RTS,S/AS02A induces strong, long-lived IL-2 responses. Viral vectors induced equal numbers of ex vivo IL-2- and IFN--producing cells, but the regimens using RTS,S/AS02A induced more ex vivo IL-2 than IFN- (P = 0.02). For RTS,S/AS02A regimens, PBMCs were not available at 7 days postimmunization. The earliest IL-2 response measured after RTS,S/AS02A is 21 days postvaccination. At 14 days, IL-2 production induced by DNA (prime)-MVA (boost) regimens (DDM) in malaria-nave volunteers had fallen to 30 spots per million PBMCs, whereas IL-2 production 21 days postvaccination with RTS,S/AS02A-containing regimens was 147 spots per million (P = 0.03). Furthermore, the IL-2 responses induced by RTS,S/AS02A were sustained 3 months later, when DDM-induced responses had fallen to baseline levels.
IL-2 responses were not significantly induced after a single RTS,S/AS02A immunization, or after MR (i.e., MVA CS followed by RTS,S/AS02A). PBMCs following two RTS,S/AS02A immunizations were not available, but IL-2 production after MRR (i.e., MVA CS followed by two sequential RTS,S/AS02A immunizations) was significantly higher than after MR (P = 0.02).
Responses induced among malaria-exposed volunteers. In malaria-exposed volunteers, the alternating-vector vaccinations with the ME-TRAP insert provoked a rise in ex vivo IFN- production and resting memory cells (2). The induction of IL-2 following alternating-vector vaccinations (FMF/MFM) was marginal (from 50 spots per million at baseline [95% confidence interval {95% CI }, 30 to 90 spots per million] to 102 spots per million after vaccination [95% CI, 40 to 320 spots per million]; P = 0.07). However, although FP9/MVA regimens encoding CS generated a rise in IFN- (2), there was no evidence of a rise in IL-2 production (P = 0.6).
After IL-2 responses were examined by vaccination group, the correlation of IL-2 production with IFN- and resting memory was then examined by individual responses. This aimed to determine whether the acute response, in terms of IL-2 production or IFN- production, predicted the durable resting memory response seen at 3 months.
Early IFN- and IL-2 responses are both independent predictors of durable resting memory after vaccination of malaria-nave subjects (Fig. 2). The IL-2 and IFN- effector responses at 1 week postvaccination were then compared with durable resting memory detected by cultured ELISPOT assays 3 months after the final vaccination. In nave volunteers, the effector responses at 1 week, measured either by IFN- or IL-2 production, predicted the resting memory population seen 3 months later by cultured ELISPOT assays. Since IFN- and IL-2 correlate, a multiple-regression model was examined, including data from all nave volunteers. This was to determine whether IL-2 or IFN- were independent predictors of resting memory, since the correlation between IL-2 and resting memory might have been spurious, seen only as a result of IFN- as a confounding factor. In fact, in a multiple-regression analysis, resting memory at 3 months was predicted by both IFN- production (coefficient of 0.24, P = 0.001) and by IL-2 production (coefficient of 0.21, P = 0.028), demonstrating that IL-2 production is associated with memory at 3 months, even after adjusting for the effect of IFN- production.
Naturally acquired immunity (Fig. 2). The impact of natural exposure could be assessed only at a single time point (i.e., prevaccination), and so effector IFN- and IL-2 production was compared to concurrent resting memory (rather than resting memory at 3 months as above). There was no correlation between resting memory and ex vivo IFN- production or IL-2 production prior to vaccination, but ex vivo IFN- and IL-2 production correlated highly.
Early IFN-, but not IL-2, predicted resting memory after some vaccinations in malaria-exposed subjects (Fig. 2). For both alternating-vector and prime-boost vaccinations in malaria-exposed subjects, the effector IFN- response measured at 1 week correlated with subsequent resting memory populations at 3 months, but IL-2 production 1 week after vaccination did not predict resting memory at 3 months. However, whereas the IFN- effector correlated with IL-2 after alternating-vector immunizations, no such correlation was seen after prime-boost vaccination using FP9/MVA encoding CS.
IL-2 inversely correlated with resting memory at 9 months in malaria-exposed subjects (Fig. 2). Data on long-lived resting memory (from samples taken 9 months after the final vaccination) were available for the volunteers given alternating-vector vaccinations. The correlation between IFN- production 7 days after vaccination and resting memory at 3 months was no longer present for resting memory at 9 months. Rather, there was an inverse correlation of marginal significance (P = 0.027) between IL-2 effector response at 7 days and resting memory at 9 months.
In dual-cytokine ELISPOT assays, IFN- and IL-2 coproduction by individual cells predicts memory (Fig. 3). The dual-cytokine ELISPOT assay distinguished single-cytokine-producing cells from cells producing both cytokines simultaneously by a double-color ELISPOT assay. Responses were studied from 11 volunteers, all from the area where malaria is endemic. Six of these volunteers had received alternating-vector immunizations, and five had received prime-boost CS encoding immunizations. Volunteers were chosen according to the availability of PBMCs for this additional assay. In samples from the 11 volunteers, 33% of effector cells produced IFN- only, 35% produced IL-2 only, and 32% produced both cytokines.
These early effector cells were then compared with durable resting memory, assayed on samples taken 3 months after vaccination. The double-cytokine-producing cells yielded the strongest correlations with subsequent resting memory responses (top three panels of Fig. 3). When all three populations were examined in a multiple-regression model, it was the double-cytokine-producing cells that produced the higher coefficient. The coefficients were 0.94 (95% CI, 0.2 to 2) for double-cytokine-producing cells, 0.21 (95% CI, 0.9 to 1.4) for IFN--producing cells, and 0.01 (–1.2 to 1.2) for IL-2 alone. A simple regression model for the effect of double-cytokine-producing cells alone gave a coefficient of 1.1 (95% CI, 0.4 to 1.9). Double-positive IL-2 and IFN--secreting cells at 1 week also significantly predicted the magnitude of the effector response at 3 months (bottom panels of Fig. 3), although this analysis was dominated by two outlying points.
DISCUSSION
Effector T-cell responses to immunization have previously been characterized by IFN- production. This correlates with effector function in general and is probably the key mediator of T-cell inhibition of liver-stage malaria parasites (8). However, effector T cells have been shown to produce a variety of cytokines, and this has been incompletely characterized after immunization. Furthermore, several lines of evidence implicate a role for IL-2 in the differentiation of memory cells, a key component of vaccine-induced immunity. We have previously demonstrated that resting memory cells are significantly expanded by vaccination of malaria-nave (19) and malaria-exposed (2) volunteers.
Endogenous IL-2 production was induced by vaccination in this study. Production was short-lived after DNA-primed, MVA-boosted regimens in malaria-nave volunteers but sustained over 3 months after regimens using the particulate protein vaccine RTS,S/AS02A in malaria-nave volunteers. Limited IL-2 production was induced either by a single RTS,S/AS02A immunization or by MVA followed by RTS,S/AS02A. A significant rise in IL-2 production was seen after the second RTS,S/AS02A vaccination. This temporal association and the less sustained IL-2 production seen after other viral vector regimens using MVA CS indicate that either RTS,S/AS02A itself or the combination of RTS,S/AS02A with the viral vector caused the more persistent IL-2 production.
Malaria exposure alone induced proportionally more IL-2 production than IFN-. This was probably a memory response, since the entomological inoculation rate in the area is one infective bite per year (22), and it would be unlikely that a significant proportion of the volunteers had recently been exposed to sporozoites. IL-2 and IFN- effector responses closely correlated, but resting memory responses did not correlate with either of these. This is in agreement with a previous study of IFN--producing effector and resting memory cells in an area where malaria is endemic (11).
IL-2-producing CD4-positive cells have also been identified by intracellular cytokine staining 7 days after vaccination with RTS,S/MVA CS regimens (9) and 7 days after vaccination with viral vector regimens (29). However, intracellular cytokine staining was less sensitive and less reproducible due to the low numbers of cytokine-producing cells. Although bead depletions were not repeated in this study, in previous studies CD4 cells have been shown to be the main producers of IL-2 secretion (9, 29), IFN- secretion (23, 31), and resting memory responses (19).
In malaria-nave volunteers, IFN- and IL-2 production at 1 week after vaccination predicted the resting memory population present 3 months after vaccination, analyzed alone or simultaneously in multiple regression. This relates the size of the early effector response to resting memory per se and suggests either that cells producing IL-2 are more likely to differentiate into memory cells or that the secreted IL-2 promotes memory differentiation or that common factors influence the magnitude of both the effector and memory responses that may be triggered in parallel.
Ex vivo IFN-, but not IL-2, production predicted resting memory responses 3 months after vaccination in the malaria-exposed vaccinees. However, the FP9-primed/MVA-boosted CS-encoding regimen studied in the malaria-exposed vaccinees gave lower overall immunogenicity and only weakly induced resting memory cells. Resting memory cell populations were seen before vaccination in volunteers in areas where malaria is endemic, and in other studies using a single viral vector predicted postvaccination effector responses (Bejon et al., unpublished). The limited expansion of resting memory induced by vaccination would have greatly reduced the statistical power to detect any correlation with vaccine-induced effector responses.
Results from dual-cytokine ELISPOT assays suggest that cells producing IFN- and IL-2 together are more closely related to the differentiation of memory cells than cells producing IFN- only. This assay was performed on cells from malaria-exposed donors, and the results differs from the results described using the IL-2 ELISPOT assay alone. This may be because the cells producing IL-2 only correlate weakly with memory. However, the IL-2 ELISPOT assay is unable to distinguish between these and coproducing cells and so fails to detect the apparent synergy of IFN- and IL-2 coproduction in determining durable resting memory. There was a tendency for cells producing IL-2 only to correlate with resting memory in the dual-cytokine ELISPOT assays (Fig. 3) that was not apparent in the single-cytokine ELISPOT assays (Fig. 2). However, the cell populations are slightly different (since IL-2-producing cells identified in the single-color ELISPOT assay include double-cytokine and single-cytokine-producing cells), and the correlation seen in the dual-cytokine ELISPOT assay with IL-2 production was marginally significant (P = 0.04). The correlation with dual-cytokine-producing cells was much more robust (P = 0.005) than for IL-2 production alone. Eleven of 19 volunteers for whom the dual-cytokine ELISPOT assay was conducted were selected when sufficient PBMCs had been separated, and the assays were conducted concurrently with the IL-2 ELISPOT assay. Sample bias seems an unlikely explanation for this result.
In the small number of volunteers assessed at 9 months, resting memory at 9 months inversely correlated with IL-2 production, in contrast to the positive correlation seen earlier. However, the correlation was marginal (P = 0.027), and the opposite result had been seen at 3 months.
The double-color ELISPOT assay allowed differentiation between cells producing single cytokines and those producing both IL-2 and IFN-. All three populations significantly predicted subsequent resting memory, but IFN- IL-2 double-positive cells most strongly predicted both resting and effector memory. The correlation with memory and double-positive cells had the highest independent r value, and when all three populations were included in multiple regression, the coefficient relating IFN- IL-2 double positives at 1 week to resting memory at 3 months was 5 times that for cells producing IFN- or IL-2 only. This suggests that the association between IL-2 and memory at 3 months reflects IL-2 coproducing cells themselves persisting, rather than local IL-2 secretion promoting the persistence of other T cells. IL-2 and IFN- coproduction also supports antigen-specific proliferation in chronic HIV infection (33) and is associated with nonprogression to AIDS (4).
It has previously been demonstrated in murine studies that cells expressing the IL-7 receptor are more likely to develop memory phenotype (17). However, this cannot be identified readily by the ELISPOT assay, and in the authors' hands, the lower frequencies of T cells induced by vaccination of humans have been more difficult to identify by intracellular cytokine staining using four-color FACS (S. Keating, unpublished data). Here, coproduction of IFN- and IL-2 appeared to mark potential memory cells.
The factors influencing the differentiation of memory cells are incompletely understood, and the role of IL-2 in generated memory cells is unclear. The data on vaccinated human volunteers here strongly support a role for IL-2 in memory cell differentiation and should inform the design of further vaccines capable of raising longer-lasting memory.
ACKNOWLEDGMENTS
The volunteers are thanked for their participation. Marcelle Van Mechelen (GlaxoSmithKline Biologicals, Rixensart, Belgium) is thanked for her presubmission review of the manuscript.
P. Bejon holds a Wellcome Trust training fellowship (073597), K. Marsh holds a Wellcome Trust senior fellowship (061702), and A. V. S. Hill is a Wellcome Trust Principal Research Fellow.
The study was performed with the permission of KEMRI National Ethics Committees, and OXTREC, the Oxford University Tropical Research Ethics Committee, and published with the permission of the Director of KEMRI.
A.V.S.H. is cofounder of and an equity holder in Oxxon Therapeutics, a company developing prime-boost therapeutic vaccines. J.C. and P.M. are employees of GlaxoSmithKline Biologicals. No other authors have a conflict of interest.
FOOTNOTES
Corresponding author. Mailing address: Kenya Medical Research Institute, Centre for Geographical Medical Research (Coast), P.O. Box 230, Kilifi, Kenya. Phone: 254 415 22063. Fax: 254 415 22396. E-mail: pbejon@kilifi.mimcom.net.
Published ahead of print on 11 September 2006.
REFERENCES
1. Arduino, R. C., E. C. Nannini, M. Rodriguez-Barradas, S. Schrader, M. Losso, K. Ruxrungtham, M. C. Allende, S. Emery, L. Fosdick, J. Neaton, J. A. Tavel, R. T. Davey, and H. C. Lane. 2004. CD4 cell response to 3 doses of subcutaneous interleukin 2: meta-analysis of 3 vanguard studies. Clin. Infect. Dis. 39:115-122.
2. Bejon, P., O. K. Kai, J. Mwacharo, S. Keating, T. Lang, S. C. Gilbert, N. Peshu, K. Marsh, and A. V. Hill. 2006. Alternating-vector immunizations encoding pre-erythrocytic malaria antigens enhance memory responses in a malaria endemic area. Eur. J. Immunol. 36:2264-2272.
3. Blattman, J. N., J. M. Grayson, E. J. Wherry, S. M. Kaech, K. A. Smith, and R. Ahmed. 2003. Therapeutic use of IL-2 to enhance antiviral T-cell responses in vivo. Nat. Med. 9:540-547.
4. Boaz, M. J., A. Waters, S. Murad, P. J. Easterbrook, and A. Vyakarnam. 2002. Presence of HIV-1 Gag-specific IFN-gamma+IL-2+ and CD28+IL-2+ CD4 T-cell responses is associated with nonprogression in HIV-1 infection. J. Immunol. 169:6376-6385.
5. Brenchley, J. M., D. C. Douek, D. R. Ambrozak, M. Chatterji, M. R. Betts, L. S. Davis, and R. A. Koup. 2002. Expansion of activated human naive T-cells precedes effector function. Clin. Exp. Immunol. 130:432-440.
6. Cheng, L. E., C. Ohlen, B. H. Nelson, and P. D. Greenberg. 2002. Enhanced signaling through the IL-2 receptor in CD8+ T cells regulated by antigen recognition results in preferential proliferation and expansion of responding CD8+ T cells rather than promotion of cell death. Proc. Natl. Acad. Sci. USA 99:3001-3006.
7. Doolan, D. L., and S. L. Hoffman. 2000. The complexity of protective immunity against liver-stage malaria. J. Immunol. 165:1453-1462.
8. Doolan, D. L., and S. L. Hoffman. 1997. Pre-erythrocytic-stage immune effector mechanisms in Plasmodium spp. infections. Philos. Trans. R. Soc. Lond. B 352:1361-1367.
9. Dunachie, S. J., M. Walther, J. M. Vuola, D. P. Webster, S. M. Keating, T. Berthoud, L. Andrews, P. Bejon, I. Poulton, G. Butcher, K. Watkins, R. E. Sinden, A. Leach, P. Moris, N. Tornieporth, J. Schneider, F. Dubovsky, E. Tierney, J. Williams, D. Gray Heppner, Jr., S. C. Gilbert, J. Cohen, and A. V. Hill. 2006. A clinical trial of prime-boost immunisation with the candidate malaria vaccines RTS,S/AS02A and MVA-CS. Vaccine 24:2850-2859.
10. Epstein, J. E., Y. Charoenvit, K. E. Kester, R. Wang, R. Newcomer, S. Fitzpatrick, T. L. Richie, N. Tornieporth, D. G. Heppner, C. Ockenhouse, V. Majam, C. Holland, E. Abot, H. Ganeshan, M. Berzins, T. Jones, C. N. Freydberg, J. Ng, J. Norman, D. J. Carucci, J. Cohen, and S. L. Hoffman. 2004. Safety, tolerability, and antibody responses in humans after sequential immunization with a PfCSP DNA vaccine followed by the recombinant protein vaccine RTS,S/AS02A. Vaccine 22:1592-1603.
11. Flanagan, K. L., E. A. Lee, M. B. Gravenor, W. H. Reece, B. C. Urban, T. Doherty, K. A. Bojang, M. Pinder, A. V. Hill, and M. Plebanski. 2001. Unique T cell effector functions elicited by Plasmodium falciparum epitopes in malaria-exposed Africans tested by three T cell assays. J. Immunol. 167:4729-4737.
12. Geginat, J., A. Lanzavecchia, and F. Sallusto. 2003. Proliferation and differentiation potential of human CD8+ memory T-cell subsets in response to antigen or homeostatic cytokines. Blood 101:4260-4266.
13. Gilbert, S. C., M. Plebanski, S. J. Harris, C. E. Allsopp, R. Thomas, G. T. Layton, and A. V. Hill. 1997. A protein particle vaccine containing multiple malaria epitopes. Nat. Biotechnol. 15:1280-1284.
14. Godkin, A. J., H. C. Thomas, and P. J. Openshaw. 2002. Evolution of epitope-specific memory CD4+ T cells after clearance of hepatitis C virus. J. Immunol. 169:2210-2214.
15. Herrington, D., J. Davis, E. Nardin, M. Beier, J. Cortese, H. Eddy, G. Losonsky, M. Hollingdale, M. Sztein, M. Levine, et al. 1991. Successful immunization of humans with irradiated malaria sporozoites: humoral and cellular responses of the protected individuals. Am. J. Trop. Med. Hyg. 45:539-547.
16. Hill, A. V., C. E. Allsopp, D. Kwiatkowski, N. M. Anstey, P. Twumasi, P. A. Rowe, S. Bennett, D. Brewster, A. J. McMichael, and B. M. Greenwood. 1991. Common west African HLA antigens are associated with protection from severe malaria. Nature 352:595-600.
17. Kaech, S. M., J. T. Tan, E. J. Wherry, B. T. Konieczny, C. D. Surh, and R. Ahmed. 2003. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Nat. Immunol. 4:1191-1198.
18. Kaech, S. M., E. J. Wherry, and R. Ahmed. 2002. Effector and memory T-cell differentiation: implications for vaccine development. Nat. Rev. Immunol. 2:251-262.
19. Keating, S. M., P. Bejon, T. Berthoud, J. M. Vuola, S. Todryk, D. P. Webster, S. J. Dunachie, V. S. Moorthy, S. J. McConkey, S. C. Gilbert, and A. V. Hill. 2005. Durable human memory T cells quantifiable by cultured enzyme-linked immunospot assays are induced by heterologous prime boost immunization and correlate with protection against malaria. J. Immunol. 175:5675-5680.
20. Lalvani, A., P. Moris, G. Voss, A. A. Pathan, K. E. Kester, R. Brookes, E. Lee, M. Koutsoukos, M. Plebanski, M. Delchambre, K. L. Flanagan, C. Carton, M. Slaoui, C. Van Hoecke, W. R. Ballou, A. V. S. Hill, and J. Cohen. 1999. Potent induction of focused Th1-type cellular and humoral immune responses by RTS,S/SBAS2, a recombinant Plasmodium falciparum malaria vaccine. J. Infect. Dis. 180:1656-1664.
21. Manjunath, N., P. Shankar, J. Wan, W. Weninger, M. A. Crowley, K. Hieshima, T. A. Springer, X. Fan, H. Shen, J. Lieberman, and U. H. von Andrian. 2001. Effector differentiation is not prerequisite for generation of memory cytotoxic T lymphocytes. J. Clin. Investig. 108:871-878.
22. Mbogo, C. M., J. M. Mwangangi, J. Nzovu, W. Gu, G. Yan, J. T. Gunter, C. Swalm, J. Keating, J. L. Regens, J. I. Shililu, J. I. Githure, and J. C. Beier. 2003. Spatial and temporal heterogeneity of Anopheles mosquitoes and Plasmodium falciparum transmission along the Kenyan coast. Am. J. Trop. Med. Hyg. 68:734-742.
23. McConkey, S. J., W. H. Reece, V. S. Moorthy, D. Webster, S. Dunachie, G. Butcher, J. M. Vuola, T. J. Blanchard, P. Gothard, K. Watkins, C. M. Hannan, S. Everaere, K. Brown, K. E. Kester, J. Cummings, J. Williams, D. G. Heppner, A. Pathan, K. Flanagan, N. Arulanantham, M. T. Roberts, M. Roy, G. L. Smith, J. Schneider, T. Peto, R. E. Sinden, S. C. Gilbert, and A. V. Hill. 2003. Enhanced T-cell immunogenicity of plasmid DNA vaccines boosted by recombinant modified vaccinia virus Ankara in humans. Nat. Med. 9:729-735.
24. Moorthy, V. S., M. Pinder, W. H. Reece, K. Watkins, S. Atabani, C. Hannan, K. Bojang, K. P. McAdam, J. Schneider, S. Gilbert, and A. V. Hill. 2003. Safety and immunogenicity of DNA/modified vaccinia virus ankara malaria vaccination in African adults. J. Infect. Dis. 188:1239-1244.
25. Reece, W. H., M. Pinder, P. K. Gothard, P. Milligan, K. Bojang, T. Doherty, M. Plebanski, P. Akinwunmi, S. Everaere, K. R. Watkins, G. Voss, N. Tornieporth, A. Alloueche, B. M. Greenwood, K. E. Kester, K. P. McAdam, J. Cohen, and A. V. Hill. 2004. A CD4+ T-cell immune response to a conserved epitope in the circumsporozoite protein correlates with protection from natural Plasmodium falciparum infection and disease. Nat. Med. 10:406-410.
26. Sallusto, F., D. Lenig, R. Forster, M. Lipp, and A. Lanzavecchia. 1999. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401:708-712.
27. Stoute, J. A., M. Slaoui, D. G. Heppner, P. Momin, K. E. Kester, P. Desmons, B. T. Wellde, N. Garcon, U. Krzych, M. Marchand, W. R. Ballou, and J. D. Cohen for The RTS,S Malaria Vaccine Evaluation Group. 1997. A preliminary evaluation of a recombinant circumsporozoite protein vaccine against Plasmodium falciparum malaria. N. Engl. J. Med. 336:86-91.
28. Sun, P., R. Schwenk, K. White, J. A. Stoute, J. Cohen, W. R. Ballou, G. Voss, K. E. Kester, D. G. Heppner, and U. Krzych. 2003. Protective immunity induced with malaria vaccine, RTS,S, is linked to Plasmodium falciparum circumsporozoite protein-specific CD4+ and CD8+ T cells producing IFN-gamma. J. Immunol. 171:6961-6967.
29. Vuola, J. M., S. Keating, D. P. Webster, T. Berthoud, S. Dunachie, S. C. Gilbert, and A. V. Hill. 2005. Differential immunogenicity of various heterologous prime-boost vaccine regimens using DNA and viral vectors in healthy volunteers. J. Immunol. 174:449-455.
30. Walther, M., F. M. Thompson, S. Dunachie, S. Keating, S. Todryk, T. Berthoud, L. Andrews, R. F. Andersen, A. Moore, S. C. Gilbert, I. Poulton, F. Dubovsky, E. Tierney, S. Correa, A. Huntcooke, G. Butcher, J. Williams, R. E. Sinden, and A. V. Hill. 2006. Safety, immunogenicity, and efficacy of prime-boost immunization with recombinant poxvirus FP9 and modified vaccinia virus Ankara encoding the full-length Plasmodium falciparum circumsporozoite protein. Infect. Immun. 74:2706-2716.
31. Webster, D. P., S. Dunachie, J. M. Vuola, T. Berthoud, S. Keating, S. M. Laidlaw, S. J. McConkey, I. Poulton, L. Andrews, R. F. Andersen, P. Bejon, G. Butcher, R. Sinden, M. A. Skinner, S. C. Gilbert, and A. V. Hill. 2005. Enhanced T cell-mediated protection against malaria in human challenges by using the recombinant poxviruses FP9 and modified vaccinia virus Ankara. Proc. Natl. Acad. Sci. USA 102:4836-4841.
32. Wherry, E. J., V. Teichgraber, T. C. Becker, D. Masopust, S. M. Kaech, R. Antia, U. H. von Andrian, and R. Ahmed. 2003. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat. Immunol. 4:225-234.
33. Zimmerli, S. C., A. Harari, C. Cellerai, F. Vallelian, P. A. Bart, and G. Pantaleo. 2005. HIV-1-specific IFN-gamma/IL-2-secreting CD8 T cells support CD4-independent proliferation of HIV-1-specific CD8 T cells. Proc. Natl. Acad. Sci. USA 102:7239-7244.(Philip Bejon, Sheila Keating, Jedidah Mw)
Centre for Clinical Vaccinology and Tropical Medicine, University of Oxford, Oxford, OX3 7LJ United Kingdom
Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford, OX3 7BN United Kingdom
Nuffield Department of Clinical Medicine, Oxford University, John Radcliffe Hospital, Oxford, United Kingdom
Naval Medical Research Center, 503 Robert Grant Avenue, Silver Spring, Maryland
GlaxoSmithKline Biologicals, Rixensart, Belgium
ABSTRACT
Two different cell populations respond to potent T-cell-inducing vaccinations. The induction and loss of effector cells can be seen using an ex vivo enzyme-linked immunospot (ELISPOT) assay, but the more durable resting memory response is demonstrable by a cultured ELISPOT assay. The relationship of the early effector response to durable resting memory is incompletely understood. Effector phenotype is usually identified by gamma interferon (IFN-) production, but interleukin-2 (IL-2) has been specifically linked to the differentiation of memory cells. Here, IFN-- and IL-2-secreting effector cells were identified by an ex vivo ELISPOT assay 1 week after vaccination and compared with the resting memory responses detected by a cultured ELISPOT assay 3 months later. The different kinetics and induction of IL-2 by different vaccines and natural exposure are described. Furthermore, both early IFN- and IL-2 production independently predicted subsequent memory responses at 3 months in malaria-nave volunteers, but only IFN- predicted memory in malaria-exposed volunteers. However, dual ELISPOT assays were also performed on malaria-exposed volunteers to identify cells producing both cytokines simultaneously. This demonstrated that double-cytokine-producing cells were highly predictive of memory. This assay may be useful in predicting vaccinations most likely to generate stable, long-term memory responses.
INTRODUCTION
Experimental studies with mice (7), field studies with humans (16), and irradiated sporozoite immunization (15) suggest that T-cell responses to pre-erythrocytic malaria antigens are protective. A heterologous prime-boost strategy that induces T cells has been developed (23), using sequential immunization with different vectors delivering a common pre-erythrocytic malaria antigen. Two antigens have been used, i.e., ME-TRAP (multiple pre-erythrocytic stage epitopes joined with the whole pre-erythrocytic stage antigen TRAP) and CS (the circumsporozoite antigen with one additional P. falciparum epitope). Each antigen can be delivered by two different viral vectors, either the attenuated fowlpox strain FP9 or modified virus Ankara (MVA). In previous studies of ME-TRAP-encoding vaccinations in nonimmune volunteers, some individuals were fully protected, and many immunized groups showed mean delays in time to parasitemia after controlled bites from Plasmodium falciparum-infected mosquitoes (31).
An alternative approach to vaccination against malaria uses a particulate protein vaccine with adjuvant (RTS,S, expressing the CS antigen). This vaccination is protective in sporozoite challenge (27) and also potently induces T-cell responses (20). In further studies, it has been given in combination with a viral vector encoding CS (9).
Vaccination appears to induce two different populations of T cells. There is a primary expansion and contraction of effector T cells over several weeks and a subsequent differentiation of memory T cells (18). The initial expansion and contraction of effector T cells following prime-boost vaccination can be followed using the ex vivo enzyme-linked immunospot (ELISPOT) assay (23). This assay detects peptide-specific gamma interferon (IFN-)-producing T cells immediately following isolation during 18 h of incubation. However, a resting memory population that acquires the effector phenotype (i.e., IFN- production) after a second exposure to antigen can be identified by a cultured ELISPOT assay (19). This assay employs a 10-day culture period, initially with antigen and then with exogenous interleukin-2 (IL-2).
The cell populations identified by these two assays are different. A study of naturally acquired immunity to the CS protein demonstrated no correlation between cultured ELISPOT and ex vivo ELISPOT assay results (11). Furthermore, the cells identified by the cultured ELISPOT assay persist for at least 6 months after vaccination of malaria-nave subjects, despite waning of cells detected by the ex vivo ELISPOT assay (19). Any reactive T cell persisting beyond the initial expansion and contraction of effector cells could be described as a memory cell, but those cells requiring 10 days of culture to acquire antigen-specific IFN- secretion are referred to here as resting memory cells. This description does not directly relate to the division between central and effector memory, which refers to cell surface markers and homing patterns (26). However, experiments that selectively depleted CCR7-positive T cells before culture suggest that the cultured ELISPOT assay does detect central memory cells (14). In previous studies, we have shown that resting memory cells are expanded following vaccination in malaria-na ve subjects (19) and malaria-exposed subjects (2) and are not simply due to prior exposure.
Mean effector responses among vaccination groups predicted protection from sporozoite challenge, determined by time to parasitemia (31), but did not predict the variability in protection at an individual level. However, cultured ELISPOT assays to detect resting memory responses were recently shown to predict the individual variability in protection (19), and following vaccination with RTS,S/AS02, the cultured ELISPOT assay correlated with protection (25, 28).
Experimental data are contradictory on the relationship of initial effector cell expansion with memory populations. It is not clear whether memory cells differentiate immediately on stimulation with antigen or whether they differentiate from a subset of effector cells (18). Experiments on adoptive transfer in mice suggest that effector differentiation is not required (21), whereas other work suggests these cell populations are not distinct subsets (32). Several lines of investigation link IL-2 to memory cell differentiation. In vitro work suggests that memory cells require IL-2 for proliferation (5), but IL-2 signaling can lead to either proliferation or apoptosis (6), and the effects of IL-2 may vary according to the presence or absence of antigen (12). Exogenous IL-2 promoted resting memory in mice but accelerated cell death when given during expansion (3). In human immunodeficiency virus (HIV)-infected humans, exogenous IL-2 produces a sustained rise in CD4 count (1). Furthermore, although memory is critical to the long-term efficacy of vaccinations in the field, little is known about IL-2 production following immunization.
This study aimed to characterize the early effector response to vaccination in terms of both IL-2 and IFN- production (measured at 1 week) and to compare this with durable resting memory responses (measured at 3 months) in order to determine whether either IL-2 or IFN- production (or both) would predict memory. Phase 1 studies of candidate malaria vaccines were conducted in an area where malaria is not endemic and then in an area where it is endemic. Immunogenicity is described in detail elsewhere (2, 9). Resting memory cell responses were significantly greater after vaccination in malaria-nave (19) and malaria-exposed (2) subjects. The current manuscript describes the additional studies of IL-2 production and memory responses.
MATERIALS AND METHODS
Vaccination was not performed in a blind manner. Subjects were recruited in two locations, from the Kenya Medical Research Institute (KEMRI), Centre for Geographical Medicine Research (Coast), located in Kilifi, Kenya, an area where malaria is endemic, and from Oxford, United Kingdom, an area where malaria is not endemic. Data from 43 subjects are presented here (details of vaccination regimens given in Table 1). Ethical approval was obtained from the Kenyan Medical Research Council (KEMRI) National Ethics Committee, and OXTREC, the Oxford University Tropical Ethics Committee for studies in Kenya. The studies in Oxford were approved by both the Oxfordshire Research Ethics Committee and the Naval Medical Research Center Institutional Review Board and the U.S. Navy Surgeon General (in accordance with U.S. Navy regulations, SECNAVINST 3900.39B). Exclusion criteria included significant medical illness and HIV seropositivity.
Vaccines. Two antigen inserts were used. These were ME-TRAP, encoding the pre-erythrocytic antigen TRAP (thrombospondin-related adhesion protein) and CS (the circumsporozoite protein). TRAP is coupled to a multiple-epitope (ME) string (13). The ME string contains 14 major histocompatibility complex class I preerythrocytic epitopes, three class II epitopes (from Mycobacterium bovis BCG, tetanus toxin, and the circumsporozoite antigen), two preerythrocytic B-cell epitopes, and pb9 (a Plasmodium berghei T-cell epitope that allows preclinical potency and stability testing). CS is coupled to the P. falciparum class I epitope ls6 (from liver-stage antigen 1) and pb9.
DNA CS was manufactured and supplied by Vical Inc., San Diego, CA, and has been called the P. falciparum circumsporozoite protein (PfCSP) DNA vaccine and VCL-2510 elsewhere (10). It is full length and not coupled to ls6 or pb9. DNA ME-TRAP was manufactured by the contract manufacturer QIAGEN, Germany.
Each insert was also used in one of two viral vectors, either in the attenuated fowlpox virus (FP9) or in modified vaccinia Ankara (30, 31). Recombinant vaccine stock was supplied to the contract manufacturer IDT (Rosslau, Germany), who produced clinical lots under good manufacturing practice conditions.
The adjuvant particulate protein vaccine RTS,S/AS02 (GlaxoSmithKline Biologicals, Rixensart, Belgium) comprises the carboxyl-terminal (amino acids 207 to 395) of the 3D7 circumsporozoite protein fused to the hepatitis B surface antigen, coexpressed in yeast with the nonfused hepatitis B surface antigen. The proprietary adjuvant AS02A is composed of an oil-in-water emulsion and the two immunostimulants QS21 and monophosphoryl lipid A (27).
The different vaccine regimens used are shown in Table 1. The medical supervision of vaccination, follow-up, and adverse events is described elsewhere (2). DNA was given as a 4-mg dose (intramuscular). FP9 ME-TRAP, FP9 CS, and MVA ME-TRAP were given as 1 x 108 PFU. MVA CS was given as 1.5 x 108 PFU, and RTS,S/AS02 was given as 0.5 ml (intramuscular).
Malaria-nave volunteers. One group of malaria-nave volunteers received a priming vaccination with DNA, followed by boosting with modified vaccinia Ankara, encoding either one of two pre-erythrocytic malaria antigens (CS or ME-TRAP). Vaccinations were given 4 weeks apart.
The second group received regimens using RTS,S/AS02. This vaccine was used in two similar regimens for malaria-nave volunteers. RTS,S/AS02 was given twice, with MVA encoding CS antigen either before or after the two RTS,S/AS02 immunizations (9). Vaccinations were given 4 weeks apart.
Malaria-exposed volunteers. The malaria-exposed volunteers studied here had received two regimens using viral vectors. These were a prime-boost CS-delivering regimen, using the FP9 virus to prime and MVA to boost, or alternating-vector immunizations (i.e., sequential vaccinations with FP9, MVA, and then FP9 again) delivering ME-TRAP. Vaccinations were given 3 weeks apart for ME-TRAP and 4 weeks apart for CS.
ELISPOT assays. Peripheral blood mononuclear cells (PBMCs) were isolated from blood taken 7 days after each immunization to study the acute response (by IL-2 and ex vivo IFN-), with the exception of volunteers immunized with RTS,S/AS02-containing regimens for which PBMCs were available only 21 days after immunization. Cultured ELISPOT assays were conducted on PBMCs taken 3 months and 9 months after the final immunization. Previous studies indicate that CD8 responses are infrequent and that most responses are CD4 T cells (20, 24), and cell separations were not performed for the current study.
IFN- ELISPOT assays. IFN- ELISPOT assays used Millipore MAIP S45 plates and MabTech antibodies according to the manufacturer's instructions. A total of 4 x 105 of freshly isolated PBMCs per well were incubated in 100 μl RPMI (Sigma-Aldrich) with 10 μg ml–1 peptides for 18 to 20 h before developing. 20-mers overlapping by 10 were used for TRAP, and 15-mers overlapping by 10 were used for CS. Both TRAP and CS peptides were divided into three pools. For TRAP peptides, this was a pool of peptides from the N-terminal region, a pool from the C-terminal region, and a central pool. CS peptides were grouped in N-terminal, C-terminal, and TH2R/TH3R regions. Phytohemagglutinin (20 μg/ml) (Sigma-Aldrich) was used as a positive control, and PBMCs cultured in medium alone was used as negative control. Spot-forming cell numbers were counted with an ELISPOT plate reader (Autoimmun Diagnostika; version 3.0).
Cultured ELISPOT assays. For cultured ELISPOT assays, 1 x 106 PBMCs in 0.5 ml were incubated with 10 μg/ml/peptide of all T9/96 strain TRAP peptides or all 3D7 strain CS peptides in a 24-well plate. On day 3 and day 7, 250 μl of culture supernatant was replaced with 250 μl of 10 IU/ml recombinant IL-2. On day 9, the cells were washed three times and left overnight before the standard ELISPOT assay. Responses were presented by giving the starting cell numbers, rather than the postculture cell numbers, as for previous studies (19, 25).
PBMCs were incubated in RPMI (Sigma-Aldrich) with 10% human AB serum. All ex vivo assays were performed on fresh cells. For malaria-nave volunteers, cultured assays were performed on cells cryopreserved in 10% dimethyl sulfoxide-fetal calf serum and thawed with 25 U/ml benzonase nuclease (Novagen, Darmstadt, Germany) and RPMI with 10% fetal calf serum (Sigma-Aldrich).
The cells identified by cultured ELISPOT assay are referred to as "resting memory" cells throughout the Results and Discussion sections. These cells were identified by IFN- production after stimulation. (They did not produce detectable IL-2, even in response to phytohemagglutinin.)
IL-2 ELISPOT assays were conducted as for IFN-, but using a biotinylated polyclonal and monoclonal mouse anti-human IL-2 antibody pair (BD Pharmingen).
In dual-color ELISPOT assays, each well was coated with IL-2 and IFN- capture antibodies, incubated, and developed as for IL-2 ELISPOT assay using alkaline phosphatase-conjugated streptavidin. The wells were then incubated with RPMI to block biotin binding sites, and developed for IFN- using horseradish peroxidase-conjugated streptavidin.
Positive controls were developed from wells coated with IL-2 capture antibody only, IFN- capture antibody only, or both. This allowed comparison of spots in sample wells with spots of known color. Cells producing IL-2 only were light blue. Cells producing IFN- only were light red, and dual-cytokine-producing cells were dark purple. Developing plates briefly with the alkaline phosphatase substrate (4 min) but longer with the horseradish peroxidase substrate (10 to 15 min) produced the greatest discrimination between the dark purple and light blue spots. Wells were read in a blind manner (the reader did not know the results of cultured ELISPOT assays) and confirmed by two separate observers. Results varying by more than 10% were reread a third time, and the two closer readings were accepted.
Analysis. ELISPOT assay wells were assayed in duplicate. The mean was taken, and the negative-control well result was subtracted from each peptide well. Assays were considered failures if the positive-control well gave less than 150 spots per well or if the negative-control well gave more than 20 spots per well for ex vivo ELISPOT assays or more than 40 spots per well for cultured ELISPOT assays. Pools were summed to calculate total responses. Spot numbers given are per million PBMCs throughout.
Data were log transformed to normalize distributions. In examining for either cross-reactivity or relating resting memory and effector responses, results were paired according to the region of the CS/TRAP tested and volunteer. Each individual may contribute two or three points. This can lead to an overestimate of significance due to the linked nature of observations. To account for this, the cluster subcommand of Stata 8 (Stata Corp.) was used to adjust P values upwards.
Grouping of vaccination regimens for analysis. When the immunological responses and vaccination regimens were very similar, samples were pooled for analysis. DDM regimens had used two different antigen inserts (ME-TRAP and CS). However, immunogenicities were similar (S. J. Dunachie et al., unpublished data), and both regimens were given to malaria-nave volunteers. Samples from patients given these two regimens were therefore grouped for analysis. The regimens that combined RTS,S/AS02 and MVA CS also gave very similar immunogenicities (9), and both were given to malaria-nave volunteers, so samples were pooled for analysis.
In malaria-exposed volunteers, alternating-vector regimens (i.e., FMF and MFM) and prime-boost regimens (FM and FFM) were used. These two groups used different inserts and had previously been shown to induce quite different memory and immediate effector responses (2) so they were analyzed separately.
RESULTS
Ex vivo IL-2 production was analyzed by malaria exposure, and the magnitude and kinetics of vaccine-induced responses were compared (Fig. 1). IFN- responses are shown at key time points for comparison with IL-2 responses, but the full kinetics of IFN- production are described elsewhere (2, 9, 31). PBMCs were not available for IL-2 studies 7 days after RRM/MRR vaccinations. The samples from four subjects at baseline, two subjects at 14 days after DDM, and four subjects at 90 days after RRM/MRR had assay results with unsatisfactory positive or negative-control results.
IL-2 production at baseline differs with malaria exposure. PBMCs from malaria-nave, unimmunized subjects showed very little ex vivo IFN- or IL-2 production in response to malaria antigen (CS/TRAP). Malaria-exposed subjects had low ex vivo IFN- production and despite a few individuals with strong responses, the overall geometric mean was similar to that from malaria-nave subjects (P = 0.24). However, ex vivo IL-2 responses were significantly higher among the malaria-exposed population (P = 0.007).
IL-2 responses after viral vector immunizations are short-lived. In malaria-nave volunteers, prime-boost vaccination regimes induced both ex vivo IFN- (as previously described) and ex vivo IL-2 production (P = 0.0047). IL-2 production at 3 months was similar to baseline levels (P = 0.1).
RTS,S/AS02A induces strong, long-lived IL-2 responses. Viral vectors induced equal numbers of ex vivo IL-2- and IFN--producing cells, but the regimens using RTS,S/AS02A induced more ex vivo IL-2 than IFN- (P = 0.02). For RTS,S/AS02A regimens, PBMCs were not available at 7 days postimmunization. The earliest IL-2 response measured after RTS,S/AS02A is 21 days postvaccination. At 14 days, IL-2 production induced by DNA (prime)-MVA (boost) regimens (DDM) in malaria-nave volunteers had fallen to 30 spots per million PBMCs, whereas IL-2 production 21 days postvaccination with RTS,S/AS02A-containing regimens was 147 spots per million (P = 0.03). Furthermore, the IL-2 responses induced by RTS,S/AS02A were sustained 3 months later, when DDM-induced responses had fallen to baseline levels.
IL-2 responses were not significantly induced after a single RTS,S/AS02A immunization, or after MR (i.e., MVA CS followed by RTS,S/AS02A). PBMCs following two RTS,S/AS02A immunizations were not available, but IL-2 production after MRR (i.e., MVA CS followed by two sequential RTS,S/AS02A immunizations) was significantly higher than after MR (P = 0.02).
Responses induced among malaria-exposed volunteers. In malaria-exposed volunteers, the alternating-vector vaccinations with the ME-TRAP insert provoked a rise in ex vivo IFN- production and resting memory cells (2). The induction of IL-2 following alternating-vector vaccinations (FMF/MFM) was marginal (from 50 spots per million at baseline [95% confidence interval {95% CI }, 30 to 90 spots per million] to 102 spots per million after vaccination [95% CI, 40 to 320 spots per million]; P = 0.07). However, although FP9/MVA regimens encoding CS generated a rise in IFN- (2), there was no evidence of a rise in IL-2 production (P = 0.6).
After IL-2 responses were examined by vaccination group, the correlation of IL-2 production with IFN- and resting memory was then examined by individual responses. This aimed to determine whether the acute response, in terms of IL-2 production or IFN- production, predicted the durable resting memory response seen at 3 months.
Early IFN- and IL-2 responses are both independent predictors of durable resting memory after vaccination of malaria-nave subjects (Fig. 2). The IL-2 and IFN- effector responses at 1 week postvaccination were then compared with durable resting memory detected by cultured ELISPOT assays 3 months after the final vaccination. In nave volunteers, the effector responses at 1 week, measured either by IFN- or IL-2 production, predicted the resting memory population seen 3 months later by cultured ELISPOT assays. Since IFN- and IL-2 correlate, a multiple-regression model was examined, including data from all nave volunteers. This was to determine whether IL-2 or IFN- were independent predictors of resting memory, since the correlation between IL-2 and resting memory might have been spurious, seen only as a result of IFN- as a confounding factor. In fact, in a multiple-regression analysis, resting memory at 3 months was predicted by both IFN- production (coefficient of 0.24, P = 0.001) and by IL-2 production (coefficient of 0.21, P = 0.028), demonstrating that IL-2 production is associated with memory at 3 months, even after adjusting for the effect of IFN- production.
Naturally acquired immunity (Fig. 2). The impact of natural exposure could be assessed only at a single time point (i.e., prevaccination), and so effector IFN- and IL-2 production was compared to concurrent resting memory (rather than resting memory at 3 months as above). There was no correlation between resting memory and ex vivo IFN- production or IL-2 production prior to vaccination, but ex vivo IFN- and IL-2 production correlated highly.
Early IFN-, but not IL-2, predicted resting memory after some vaccinations in malaria-exposed subjects (Fig. 2). For both alternating-vector and prime-boost vaccinations in malaria-exposed subjects, the effector IFN- response measured at 1 week correlated with subsequent resting memory populations at 3 months, but IL-2 production 1 week after vaccination did not predict resting memory at 3 months. However, whereas the IFN- effector correlated with IL-2 after alternating-vector immunizations, no such correlation was seen after prime-boost vaccination using FP9/MVA encoding CS.
IL-2 inversely correlated with resting memory at 9 months in malaria-exposed subjects (Fig. 2). Data on long-lived resting memory (from samples taken 9 months after the final vaccination) were available for the volunteers given alternating-vector vaccinations. The correlation between IFN- production 7 days after vaccination and resting memory at 3 months was no longer present for resting memory at 9 months. Rather, there was an inverse correlation of marginal significance (P = 0.027) between IL-2 effector response at 7 days and resting memory at 9 months.
In dual-cytokine ELISPOT assays, IFN- and IL-2 coproduction by individual cells predicts memory (Fig. 3). The dual-cytokine ELISPOT assay distinguished single-cytokine-producing cells from cells producing both cytokines simultaneously by a double-color ELISPOT assay. Responses were studied from 11 volunteers, all from the area where malaria is endemic. Six of these volunteers had received alternating-vector immunizations, and five had received prime-boost CS encoding immunizations. Volunteers were chosen according to the availability of PBMCs for this additional assay. In samples from the 11 volunteers, 33% of effector cells produced IFN- only, 35% produced IL-2 only, and 32% produced both cytokines.
These early effector cells were then compared with durable resting memory, assayed on samples taken 3 months after vaccination. The double-cytokine-producing cells yielded the strongest correlations with subsequent resting memory responses (top three panels of Fig. 3). When all three populations were examined in a multiple-regression model, it was the double-cytokine-producing cells that produced the higher coefficient. The coefficients were 0.94 (95% CI, 0.2 to 2) for double-cytokine-producing cells, 0.21 (95% CI, 0.9 to 1.4) for IFN--producing cells, and 0.01 (–1.2 to 1.2) for IL-2 alone. A simple regression model for the effect of double-cytokine-producing cells alone gave a coefficient of 1.1 (95% CI, 0.4 to 1.9). Double-positive IL-2 and IFN--secreting cells at 1 week also significantly predicted the magnitude of the effector response at 3 months (bottom panels of Fig. 3), although this analysis was dominated by two outlying points.
DISCUSSION
Effector T-cell responses to immunization have previously been characterized by IFN- production. This correlates with effector function in general and is probably the key mediator of T-cell inhibition of liver-stage malaria parasites (8). However, effector T cells have been shown to produce a variety of cytokines, and this has been incompletely characterized after immunization. Furthermore, several lines of evidence implicate a role for IL-2 in the differentiation of memory cells, a key component of vaccine-induced immunity. We have previously demonstrated that resting memory cells are significantly expanded by vaccination of malaria-nave (19) and malaria-exposed (2) volunteers.
Endogenous IL-2 production was induced by vaccination in this study. Production was short-lived after DNA-primed, MVA-boosted regimens in malaria-nave volunteers but sustained over 3 months after regimens using the particulate protein vaccine RTS,S/AS02A in malaria-nave volunteers. Limited IL-2 production was induced either by a single RTS,S/AS02A immunization or by MVA followed by RTS,S/AS02A. A significant rise in IL-2 production was seen after the second RTS,S/AS02A vaccination. This temporal association and the less sustained IL-2 production seen after other viral vector regimens using MVA CS indicate that either RTS,S/AS02A itself or the combination of RTS,S/AS02A with the viral vector caused the more persistent IL-2 production.
Malaria exposure alone induced proportionally more IL-2 production than IFN-. This was probably a memory response, since the entomological inoculation rate in the area is one infective bite per year (22), and it would be unlikely that a significant proportion of the volunteers had recently been exposed to sporozoites. IL-2 and IFN- effector responses closely correlated, but resting memory responses did not correlate with either of these. This is in agreement with a previous study of IFN--producing effector and resting memory cells in an area where malaria is endemic (11).
IL-2-producing CD4-positive cells have also been identified by intracellular cytokine staining 7 days after vaccination with RTS,S/MVA CS regimens (9) and 7 days after vaccination with viral vector regimens (29). However, intracellular cytokine staining was less sensitive and less reproducible due to the low numbers of cytokine-producing cells. Although bead depletions were not repeated in this study, in previous studies CD4 cells have been shown to be the main producers of IL-2 secretion (9, 29), IFN- secretion (23, 31), and resting memory responses (19).
In malaria-nave volunteers, IFN- and IL-2 production at 1 week after vaccination predicted the resting memory population present 3 months after vaccination, analyzed alone or simultaneously in multiple regression. This relates the size of the early effector response to resting memory per se and suggests either that cells producing IL-2 are more likely to differentiate into memory cells or that the secreted IL-2 promotes memory differentiation or that common factors influence the magnitude of both the effector and memory responses that may be triggered in parallel.
Ex vivo IFN-, but not IL-2, production predicted resting memory responses 3 months after vaccination in the malaria-exposed vaccinees. However, the FP9-primed/MVA-boosted CS-encoding regimen studied in the malaria-exposed vaccinees gave lower overall immunogenicity and only weakly induced resting memory cells. Resting memory cell populations were seen before vaccination in volunteers in areas where malaria is endemic, and in other studies using a single viral vector predicted postvaccination effector responses (Bejon et al., unpublished). The limited expansion of resting memory induced by vaccination would have greatly reduced the statistical power to detect any correlation with vaccine-induced effector responses.
Results from dual-cytokine ELISPOT assays suggest that cells producing IFN- and IL-2 together are more closely related to the differentiation of memory cells than cells producing IFN- only. This assay was performed on cells from malaria-exposed donors, and the results differs from the results described using the IL-2 ELISPOT assay alone. This may be because the cells producing IL-2 only correlate weakly with memory. However, the IL-2 ELISPOT assay is unable to distinguish between these and coproducing cells and so fails to detect the apparent synergy of IFN- and IL-2 coproduction in determining durable resting memory. There was a tendency for cells producing IL-2 only to correlate with resting memory in the dual-cytokine ELISPOT assays (Fig. 3) that was not apparent in the single-cytokine ELISPOT assays (Fig. 2). However, the cell populations are slightly different (since IL-2-producing cells identified in the single-color ELISPOT assay include double-cytokine and single-cytokine-producing cells), and the correlation seen in the dual-cytokine ELISPOT assay with IL-2 production was marginally significant (P = 0.04). The correlation with dual-cytokine-producing cells was much more robust (P = 0.005) than for IL-2 production alone. Eleven of 19 volunteers for whom the dual-cytokine ELISPOT assay was conducted were selected when sufficient PBMCs had been separated, and the assays were conducted concurrently with the IL-2 ELISPOT assay. Sample bias seems an unlikely explanation for this result.
In the small number of volunteers assessed at 9 months, resting memory at 9 months inversely correlated with IL-2 production, in contrast to the positive correlation seen earlier. However, the correlation was marginal (P = 0.027), and the opposite result had been seen at 3 months.
The double-color ELISPOT assay allowed differentiation between cells producing single cytokines and those producing both IL-2 and IFN-. All three populations significantly predicted subsequent resting memory, but IFN- IL-2 double-positive cells most strongly predicted both resting and effector memory. The correlation with memory and double-positive cells had the highest independent r value, and when all three populations were included in multiple regression, the coefficient relating IFN- IL-2 double positives at 1 week to resting memory at 3 months was 5 times that for cells producing IFN- or IL-2 only. This suggests that the association between IL-2 and memory at 3 months reflects IL-2 coproducing cells themselves persisting, rather than local IL-2 secretion promoting the persistence of other T cells. IL-2 and IFN- coproduction also supports antigen-specific proliferation in chronic HIV infection (33) and is associated with nonprogression to AIDS (4).
It has previously been demonstrated in murine studies that cells expressing the IL-7 receptor are more likely to develop memory phenotype (17). However, this cannot be identified readily by the ELISPOT assay, and in the authors' hands, the lower frequencies of T cells induced by vaccination of humans have been more difficult to identify by intracellular cytokine staining using four-color FACS (S. Keating, unpublished data). Here, coproduction of IFN- and IL-2 appeared to mark potential memory cells.
The factors influencing the differentiation of memory cells are incompletely understood, and the role of IL-2 in generated memory cells is unclear. The data on vaccinated human volunteers here strongly support a role for IL-2 in memory cell differentiation and should inform the design of further vaccines capable of raising longer-lasting memory.
ACKNOWLEDGMENTS
The volunteers are thanked for their participation. Marcelle Van Mechelen (GlaxoSmithKline Biologicals, Rixensart, Belgium) is thanked for her presubmission review of the manuscript.
P. Bejon holds a Wellcome Trust training fellowship (073597), K. Marsh holds a Wellcome Trust senior fellowship (061702), and A. V. S. Hill is a Wellcome Trust Principal Research Fellow.
The study was performed with the permission of KEMRI National Ethics Committees, and OXTREC, the Oxford University Tropical Research Ethics Committee, and published with the permission of the Director of KEMRI.
A.V.S.H. is cofounder of and an equity holder in Oxxon Therapeutics, a company developing prime-boost therapeutic vaccines. J.C. and P.M. are employees of GlaxoSmithKline Biologicals. No other authors have a conflict of interest.
FOOTNOTES
Corresponding author. Mailing address: Kenya Medical Research Institute, Centre for Geographical Medical Research (Coast), P.O. Box 230, Kilifi, Kenya. Phone: 254 415 22063. Fax: 254 415 22396. E-mail: pbejon@kilifi.mimcom.net.
Published ahead of print on 11 September 2006.
REFERENCES
1. Arduino, R. C., E. C. Nannini, M. Rodriguez-Barradas, S. Schrader, M. Losso, K. Ruxrungtham, M. C. Allende, S. Emery, L. Fosdick, J. Neaton, J. A. Tavel, R. T. Davey, and H. C. Lane. 2004. CD4 cell response to 3 doses of subcutaneous interleukin 2: meta-analysis of 3 vanguard studies. Clin. Infect. Dis. 39:115-122.
2. Bejon, P., O. K. Kai, J. Mwacharo, S. Keating, T. Lang, S. C. Gilbert, N. Peshu, K. Marsh, and A. V. Hill. 2006. Alternating-vector immunizations encoding pre-erythrocytic malaria antigens enhance memory responses in a malaria endemic area. Eur. J. Immunol. 36:2264-2272.
3. Blattman, J. N., J. M. Grayson, E. J. Wherry, S. M. Kaech, K. A. Smith, and R. Ahmed. 2003. Therapeutic use of IL-2 to enhance antiviral T-cell responses in vivo. Nat. Med. 9:540-547.
4. Boaz, M. J., A. Waters, S. Murad, P. J. Easterbrook, and A. Vyakarnam. 2002. Presence of HIV-1 Gag-specific IFN-gamma+IL-2+ and CD28+IL-2+ CD4 T-cell responses is associated with nonprogression in HIV-1 infection. J. Immunol. 169:6376-6385.
5. Brenchley, J. M., D. C. Douek, D. R. Ambrozak, M. Chatterji, M. R. Betts, L. S. Davis, and R. A. Koup. 2002. Expansion of activated human naive T-cells precedes effector function. Clin. Exp. Immunol. 130:432-440.
6. Cheng, L. E., C. Ohlen, B. H. Nelson, and P. D. Greenberg. 2002. Enhanced signaling through the IL-2 receptor in CD8+ T cells regulated by antigen recognition results in preferential proliferation and expansion of responding CD8+ T cells rather than promotion of cell death. Proc. Natl. Acad. Sci. USA 99:3001-3006.
7. Doolan, D. L., and S. L. Hoffman. 2000. The complexity of protective immunity against liver-stage malaria. J. Immunol. 165:1453-1462.
8. Doolan, D. L., and S. L. Hoffman. 1997. Pre-erythrocytic-stage immune effector mechanisms in Plasmodium spp. infections. Philos. Trans. R. Soc. Lond. B 352:1361-1367.
9. Dunachie, S. J., M. Walther, J. M. Vuola, D. P. Webster, S. M. Keating, T. Berthoud, L. Andrews, P. Bejon, I. Poulton, G. Butcher, K. Watkins, R. E. Sinden, A. Leach, P. Moris, N. Tornieporth, J. Schneider, F. Dubovsky, E. Tierney, J. Williams, D. Gray Heppner, Jr., S. C. Gilbert, J. Cohen, and A. V. Hill. 2006. A clinical trial of prime-boost immunisation with the candidate malaria vaccines RTS,S/AS02A and MVA-CS. Vaccine 24:2850-2859.
10. Epstein, J. E., Y. Charoenvit, K. E. Kester, R. Wang, R. Newcomer, S. Fitzpatrick, T. L. Richie, N. Tornieporth, D. G. Heppner, C. Ockenhouse, V. Majam, C. Holland, E. Abot, H. Ganeshan, M. Berzins, T. Jones, C. N. Freydberg, J. Ng, J. Norman, D. J. Carucci, J. Cohen, and S. L. Hoffman. 2004. Safety, tolerability, and antibody responses in humans after sequential immunization with a PfCSP DNA vaccine followed by the recombinant protein vaccine RTS,S/AS02A. Vaccine 22:1592-1603.
11. Flanagan, K. L., E. A. Lee, M. B. Gravenor, W. H. Reece, B. C. Urban, T. Doherty, K. A. Bojang, M. Pinder, A. V. Hill, and M. Plebanski. 2001. Unique T cell effector functions elicited by Plasmodium falciparum epitopes in malaria-exposed Africans tested by three T cell assays. J. Immunol. 167:4729-4737.
12. Geginat, J., A. Lanzavecchia, and F. Sallusto. 2003. Proliferation and differentiation potential of human CD8+ memory T-cell subsets in response to antigen or homeostatic cytokines. Blood 101:4260-4266.
13. Gilbert, S. C., M. Plebanski, S. J. Harris, C. E. Allsopp, R. Thomas, G. T. Layton, and A. V. Hill. 1997. A protein particle vaccine containing multiple malaria epitopes. Nat. Biotechnol. 15:1280-1284.
14. Godkin, A. J., H. C. Thomas, and P. J. Openshaw. 2002. Evolution of epitope-specific memory CD4+ T cells after clearance of hepatitis C virus. J. Immunol. 169:2210-2214.
15. Herrington, D., J. Davis, E. Nardin, M. Beier, J. Cortese, H. Eddy, G. Losonsky, M. Hollingdale, M. Sztein, M. Levine, et al. 1991. Successful immunization of humans with irradiated malaria sporozoites: humoral and cellular responses of the protected individuals. Am. J. Trop. Med. Hyg. 45:539-547.
16. Hill, A. V., C. E. Allsopp, D. Kwiatkowski, N. M. Anstey, P. Twumasi, P. A. Rowe, S. Bennett, D. Brewster, A. J. McMichael, and B. M. Greenwood. 1991. Common west African HLA antigens are associated with protection from severe malaria. Nature 352:595-600.
17. Kaech, S. M., J. T. Tan, E. J. Wherry, B. T. Konieczny, C. D. Surh, and R. Ahmed. 2003. Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Nat. Immunol. 4:1191-1198.
18. Kaech, S. M., E. J. Wherry, and R. Ahmed. 2002. Effector and memory T-cell differentiation: implications for vaccine development. Nat. Rev. Immunol. 2:251-262.
19. Keating, S. M., P. Bejon, T. Berthoud, J. M. Vuola, S. Todryk, D. P. Webster, S. J. Dunachie, V. S. Moorthy, S. J. McConkey, S. C. Gilbert, and A. V. Hill. 2005. Durable human memory T cells quantifiable by cultured enzyme-linked immunospot assays are induced by heterologous prime boost immunization and correlate with protection against malaria. J. Immunol. 175:5675-5680.
20. Lalvani, A., P. Moris, G. Voss, A. A. Pathan, K. E. Kester, R. Brookes, E. Lee, M. Koutsoukos, M. Plebanski, M. Delchambre, K. L. Flanagan, C. Carton, M. Slaoui, C. Van Hoecke, W. R. Ballou, A. V. S. Hill, and J. Cohen. 1999. Potent induction of focused Th1-type cellular and humoral immune responses by RTS,S/SBAS2, a recombinant Plasmodium falciparum malaria vaccine. J. Infect. Dis. 180:1656-1664.
21. Manjunath, N., P. Shankar, J. Wan, W. Weninger, M. A. Crowley, K. Hieshima, T. A. Springer, X. Fan, H. Shen, J. Lieberman, and U. H. von Andrian. 2001. Effector differentiation is not prerequisite for generation of memory cytotoxic T lymphocytes. J. Clin. Investig. 108:871-878.
22. Mbogo, C. M., J. M. Mwangangi, J. Nzovu, W. Gu, G. Yan, J. T. Gunter, C. Swalm, J. Keating, J. L. Regens, J. I. Shililu, J. I. Githure, and J. C. Beier. 2003. Spatial and temporal heterogeneity of Anopheles mosquitoes and Plasmodium falciparum transmission along the Kenyan coast. Am. J. Trop. Med. Hyg. 68:734-742.
23. McConkey, S. J., W. H. Reece, V. S. Moorthy, D. Webster, S. Dunachie, G. Butcher, J. M. Vuola, T. J. Blanchard, P. Gothard, K. Watkins, C. M. Hannan, S. Everaere, K. Brown, K. E. Kester, J. Cummings, J. Williams, D. G. Heppner, A. Pathan, K. Flanagan, N. Arulanantham, M. T. Roberts, M. Roy, G. L. Smith, J. Schneider, T. Peto, R. E. Sinden, S. C. Gilbert, and A. V. Hill. 2003. Enhanced T-cell immunogenicity of plasmid DNA vaccines boosted by recombinant modified vaccinia virus Ankara in humans. Nat. Med. 9:729-735.
24. Moorthy, V. S., M. Pinder, W. H. Reece, K. Watkins, S. Atabani, C. Hannan, K. Bojang, K. P. McAdam, J. Schneider, S. Gilbert, and A. V. Hill. 2003. Safety and immunogenicity of DNA/modified vaccinia virus ankara malaria vaccination in African adults. J. Infect. Dis. 188:1239-1244.
25. Reece, W. H., M. Pinder, P. K. Gothard, P. Milligan, K. Bojang, T. Doherty, M. Plebanski, P. Akinwunmi, S. Everaere, K. R. Watkins, G. Voss, N. Tornieporth, A. Alloueche, B. M. Greenwood, K. E. Kester, K. P. McAdam, J. Cohen, and A. V. Hill. 2004. A CD4+ T-cell immune response to a conserved epitope in the circumsporozoite protein correlates with protection from natural Plasmodium falciparum infection and disease. Nat. Med. 10:406-410.
26. Sallusto, F., D. Lenig, R. Forster, M. Lipp, and A. Lanzavecchia. 1999. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401:708-712.
27. Stoute, J. A., M. Slaoui, D. G. Heppner, P. Momin, K. E. Kester, P. Desmons, B. T. Wellde, N. Garcon, U. Krzych, M. Marchand, W. R. Ballou, and J. D. Cohen for The RTS,S Malaria Vaccine Evaluation Group. 1997. A preliminary evaluation of a recombinant circumsporozoite protein vaccine against Plasmodium falciparum malaria. N. Engl. J. Med. 336:86-91.
28. Sun, P., R. Schwenk, K. White, J. A. Stoute, J. Cohen, W. R. Ballou, G. Voss, K. E. Kester, D. G. Heppner, and U. Krzych. 2003. Protective immunity induced with malaria vaccine, RTS,S, is linked to Plasmodium falciparum circumsporozoite protein-specific CD4+ and CD8+ T cells producing IFN-gamma. J. Immunol. 171:6961-6967.
29. Vuola, J. M., S. Keating, D. P. Webster, T. Berthoud, S. Dunachie, S. C. Gilbert, and A. V. Hill. 2005. Differential immunogenicity of various heterologous prime-boost vaccine regimens using DNA and viral vectors in healthy volunteers. J. Immunol. 174:449-455.
30. Walther, M., F. M. Thompson, S. Dunachie, S. Keating, S. Todryk, T. Berthoud, L. Andrews, R. F. Andersen, A. Moore, S. C. Gilbert, I. Poulton, F. Dubovsky, E. Tierney, S. Correa, A. Huntcooke, G. Butcher, J. Williams, R. E. Sinden, and A. V. Hill. 2006. Safety, immunogenicity, and efficacy of prime-boost immunization with recombinant poxvirus FP9 and modified vaccinia virus Ankara encoding the full-length Plasmodium falciparum circumsporozoite protein. Infect. Immun. 74:2706-2716.
31. Webster, D. P., S. Dunachie, J. M. Vuola, T. Berthoud, S. Keating, S. M. Laidlaw, S. J. McConkey, I. Poulton, L. Andrews, R. F. Andersen, P. Bejon, G. Butcher, R. Sinden, M. A. Skinner, S. C. Gilbert, and A. V. Hill. 2005. Enhanced T cell-mediated protection against malaria in human challenges by using the recombinant poxviruses FP9 and modified vaccinia virus Ankara. Proc. Natl. Acad. Sci. USA 102:4836-4841.
32. Wherry, E. J., V. Teichgraber, T. C. Becker, D. Masopust, S. M. Kaech, R. Antia, U. H. von Andrian, and R. Ahmed. 2003. Lineage relationship and protective immunity of memory CD8 T cell subsets. Nat. Immunol. 4:225-234.
33. Zimmerli, S. C., A. Harari, C. Cellerai, F. Vallelian, P. A. Bart, and G. Pantaleo. 2005. HIV-1-specific IFN-gamma/IL-2-secreting CD8 T cells support CD4-independent proliferation of HIV-1-specific CD8 T cells. Proc. Natl. Acad. Sci. USA 102:7239-7244.(Philip Bejon, Sheila Keating, Jedidah Mw)