Detection of EXP1-Specific CD4+ T Cell Responses Directed Against a Broad Range of Epitopes Including Two Promiscuous MHC Class II Binders During Acute Plasmodium falciparum Malaria

Background: T cells are thought to play a major role in conferring immunity against malaria. This study aimed to comprehensively define the breadth and specificity of the Plasmodium falciparum (P. falciparum)-specific CD4+ T cell response directed against the exported protein 1 (EXP1) in a cohort of patients diagnosed with acute malaria. Methods: Peripheral blood mononuclear cells of 44 patients acutely infected with P. falciparum, and of one patient infected with P. vivax, were stimulated and cultured in vitro with an overlapping set of 31 P. falciparum-specific 13-17-mer peptides covering the entire EXP1 sequence. EXP1-specific T cell responses were analyzed by ELISPOT and intracellular cytokine staining for interferon-γ production after re-stimulation with individual peptides. For further characterization of the epitopes, in silico and in vitro human leukocyte antigen (HLA) binding studies and fine mapping assays were performed. Results: We detected one or more EXP1-specific CD4+ T cell responses (mean: 1.09, range 0–5) in 47% (21/45) of our patients. Responses were directed against 15 of the 31 EXP1 peptides. Peptides EXP1-P13 (aa60-74) and P15 (aa70-85) were detected by 18% (n = 8) and 27% (n = 12) of the 45 patients screened. The optimal length, as well as the corresponding most likely HLA-restriction, of each of these two peptides was assessed. Interestingly, we also identified one CD4+ T cell response against peptide EXP1-P15 in a patient who was infected with P. vivax but not falciparum. Conclusions: This first detailed characterization of novel EXP1-specific T cell epitopes provides important information for future analysis with major histocompatibility complex-multimer technology as well as for immunomonitoring and vaccine design.

INTRODUCTION 219 million cases of malaria and 435,000 deaths were recorded in 2017, mostly in Africa, and half of the world's population is at risk of the disease (1). Plasmodium falciparum (P. falciparum) is the species that causes the majority of deaths, and drug and insecticide resistance is increasing. A consistent solution would be the broad employment of a -thus far elusive-effective and durable protective malaria vaccine (2).
The exact mechanisms that lead to protection against malaria have not been fully defined, but both specific CD4+ and CD8+ T cells are thought to play a major role in conferring immunity against malaria (3)(4)(5)(6). Obstacles in vaccine development are the large genome of P. falciparum, that encodes for more than 5,000 proteins (7). These malaria proteins exhibit great diversity and are partially only expressed at certain stages of the plasmodial life cycle, making them difficult targets for the immune system and vaccine design (8).
As a consequence, only relatively few P. falciparum-specific T cell epitopes located on a small number of malaria antigens have been identified and characterized in detail (9). Additionally, studies on the phenotype and function of P. falciparum-specific T cells are largely missing. Furthermore, detailed immunological studies that assess the T cell repertoire in correlation with patient characteristics [such as the origin and the human leukocyte antigen (HLA) alleles] and the clinical course of the disease [complicated vs. uncomplicated course, anemia, parasitaemia, or the C-reactive protein (CRP)] are lacking. Thus, fine mapping of the optimal length of T cell epitopes and investigation of their HLA-restriction will allow the synthesis of suitable major histocompatibility complex (MHC)-multimers for phenotypical analysis that are necessary to understand the function of T cells in malaria (10).
The 162 amino acid long exported protein 1 (EXP1) plays a pivotal role for the parasite at different stages of the life cycle. It is, for example, a component within the parasitophorous vacuole membrane (PVM) that separates and protects the parasite from the cytosol of the host cell (11). The sequence of EXP1 is well conserved across different strains of Plasmodium species (Supplementary Table 1). The expression of EXP1 during the liver and the blood stage of the plasmodial life cycle inside the human host potentially makes it an ideal target for P. falciparumspecific CD4+ and CD8+ T effector cells. CD8+ T cells are thought to play an important role during the liver stage because they are able to attack infected hepatocytes (that present peptides on MHC class I molecules) (3). CD4+ T cells are considered to mainly confer immunity during the blood stage in which erythrocytes (that do not carry MHC class I molecules) are being invaded by the parasite (12).
Former studies have shown that antibodies against EXP1 were able to inhibit parasite growth in vitro as well as in vivo (13,14). Also, DNA vaccines containing the P. falciparum EXP1 molecule and synthetic peptides from the C-terminal region of EXP1 conferred protection in mice (15,16). EXP1 is expressed at two different life cycle stages where it is assumed to have important physiological functions (17), and it has shown encouraging results in previous vaccine studies. This makes EXP1 a promising target antigen for vaccine development (18)(19)(20). So far, only four CD4+ and six CD8+ EXP1-specific T cell epitopes have been described (Supplementary Figure 1) (21)(22)(23)(24)(25).
In this study we comprehensively defined 15 different P. falciparum-specific EXP1 CD4+ T cell epitopes using a previously established in vitro culture protocol for pathogenspecific T cells (26)(27)(28)(29), and evaluated clinical parameters as well as patients' characteristics. We also performed fine mapping and HLA binding experiments that will allow the design of suitable MHC-multimers to characterize the phenotype of the T cell response in future studies (29). '

Patient Cohort
PBMCs (peripheral blood mononuclear cells) of P. falciparum infected patients (n = 44) and one patient infected with Plasmodium vivax, as well as uninfected healthy controls (n = 10) collected at University Medical Center Hamburg-Eppendorf, were either used freshly or stored in liquid nitrogen (−196 • C). P. falciparum infection was verified microscopically by experienced lab technicians in the diagnostic department of the Bernhard-Nocht-Institute for Tropical Medicine, Hamburg, Germany. Thick and thin blood smears were stained with 4% Giemsa and examined under oil immersion (original magnification ×100).
Patients were further sub-stratified into two groups for some of the analysis. Group 1 (n = 11) included patients originally from Germany and group 2 (n = 31) included patients who were born in Africa. Two patients came from the Philippines and Jamaica. We also compared patients with a complicated (30) (n = 11) and uncomplicated (n = 33) course of the disease and we distinguished between patients that had the first malaria infection (n = 17) and patients that have had prior malaria infections (n = 27). Other clinical parameters including parasitaemia, CRP, hemoglobin and thrombocytes were also assessed and are shown in Table 1 (clinical details for patient HH-43 are unknown).

HLA Typing
High definition molecular HLA class I and II typing was performed at the Institute of Transfusion Medicine at University Medical Center Hamburg-Eppendorf, by polymerase chain reaction-sequence specific oligonucleotide (PCR-SSO) using the commercial kit SSO LabType as previously described (One Lambda, Canoga Park, CA, USA) (33).

ELISPOT Assay
ELISPOT assays were performed as previously described (26,28,34). We used 20,000 cells per well and responses were considered positive if the number of spots was at least three times the number of spots in the negative control. Single peptides were used at a concentration of 10 µg/ml. Anti-CD3-antibodies served as a positive control, DMSO and R10 as a negative control (35). All positive responses were confirmed by ICS assays following stimulation with the respective peptide.

Intracellular IFNγ Staining and Flow Cytometry
ICS was performed as previously described (26,28). 5 × 10 5 PBMCs were stimulated with the corresponding EXP1 peptide at a final concentration of 10 µg/ml before blocking the secretion with 10 µg/ml Brefeldin A (Sigma Aldrich) 1 h We defined a T cell response as positive when the percentage of CD4+ T cells within the gate for IFNγ was three times higher than the negative control and the population could be clearly separated from the negative control (26,28,36). DMSO and R10 were added to the negative control. Figure 1A shows an exemplary ICS result of an EXP1-specific CD4+ T cell response.

In vitro HLA Binding Assays and in silico Prediction
In vitro binding assays with 14 of the peptides that elicited a response were performed using purified HLA-DR molecules, as previously described (

Statistical Analysis
All flow cytometric data were analyzed using FlowJo 10.5.0 software (Treestar, Ashland, OR, USA). Statistical analyses were carried out using the Prism 7.0 software (GraphPad software, San Diego, CA). Mann Whitney test or the Kruskal-Wallis test with Dunn's post-test was performed throughout all samples for inter group comparisons. Spearman's correlation was performed for bivariate correlation analyses. Data are expressed as means with standard deviations (SD) or with standard error of mean (SEM). P-values less than or equal to 0.05 were considered significant.

Clinical Features of the Study Cohort
The clinical data of all patients as well as other characteristics are summarized in   The binding capacity is displayed as IC50 value deduced from in vitro experiments with selected purified HLA-DR molecules and overlapping EXP1 peptides. High affinity binding is defined as IC50 Initial parasitaemia was on average 2.37% (range: <1-11%), CRP 89 mg/dl (range: 5-239 mg/dl), hemoglobin 12.1 g/dl (range: 7.1-16.2 g/dl), and thrombocytes 111.300 /µl (range: 7,400-404,000 /µl). The average number of days between the start of symptoms and the blood sampling for our study was ∼8.6 (range: 2-28 days). The number of days between the start of therapy and the blood sampling for our study was about 3.1 (range: 0-28 days). Patient HH-17 had a blood draw before therapy for malaria was initiated.
Detection of EXP1-Specific T Cell Responses After in vitro Cultivation With IL-2 The P. falciparum-specific T cell response has so far not been comprehensively assessed for neither the majority of P. falciparum-specific antigens in general, nor for EXP1 in particular (9). In order to evaluate the EXP1-specific T cell response on an epitope level, we first looked at the ex vivo IFNγ production of PBMCs of acutely infected malaria patients after stimulation with single peptides of an overlapping 13-17-mer peptide set covering the entire P. falciparum antigen EXP1. Most samples were drawn 1-4 days after admission to the hospital (average 3.07 days) shortly before, during or after initiation of anti-malaria therapy, only 3 samples were taken at later timepoints.
However, we detected only one T cell response with the ex vivo ELISPOT in a total of ten patients (data not shown). The fresh blood sample of patient HH-38 showed an ex vivo response against EXP1-P15. After the in vitro culture, the same patient responded again to EXP1-P15 and additionally to EXP1-P10. These results are in line with the results of previous studies that showed that the magnitude of the ex vivo P. falciparum-specific CD4+ and CD8+ T cell response is low (23,(45)(46)(47). No EXP1specific T cell responses were detectable after 14 days of in vitro culture of 10 healthy controls that had never traveled to a malaria endemic area (data not shown).
Due to the presumably low ex vivo precursor frequencies of P. falciparum-specific T cells, we employed a previously used, ultra-sensitive, in vitro approach (26)(27)(28)48). In short: our peptide set was divided into three different peptide pools ( Table 2) and we started three parallel cell cultures per patient, stimulating them with each pool for 14 days (Pool 1: EXP1-P1-10, Pool 2: EXP1-P11-20, Pool 3: EXP1-P21-31). This was followed by a single peptide ELISPOT for IFNγ-production against re-stimulation with the individual peptides. The 31 overlapping single peptides covered the entire EXP1 protein, and PBMCs from 45 patients were tested. Every positive response in the ELISPOT assay was confirmed by re-stimulation with the according single peptide and subsequent intracellular cytokine staining (ICS) assay for IFNγ production. Figure 1A shows an exemplary ICS result of an EXP1-specific CD4+ T cell response.
We also longitudinally analyzed, in selected patients, the EXP1-specific T cell responses in the long-term follow-up. The patient HH-06 showed a large response against EXP1-P15 (frequency: 1.86% IFNγ+CD4+) and EXP1-P25 (frequency: 1.88% IFNγ+CD4+) during acute infection. However, 12 months after the first sampling a specific T cell response could not be detected from a cell culture started directly with fresh blood (data not shown). The patient had stayed in Germany and no further malaria infections had occurred in the intervening 12 months. By contrast, in patient HH-27, who responded to EXP1-P16 (frequency: 3.91% IFNγ+CD4+), we could detect responses against EXP1-P16 (frequency: 9.73% IFNγ+CD4+) and EXP1-P10 (frequency: 0.25% IFNγ+CD4+) in a sample collected 7 years prior to the present assays. During these 7 years, several subsequent malaria infections had occurred as reported by the patient and patient's health provider.
Of the 45 patient samples utilized, seven samples were cultured from fresh samples, of which three patients showed a CD4+ T cell response (42.9%, 3/7). The other 38 samples were stored in liquid nitrogen before analysis, and of these 18 patients showed a T cell response (47.4%, 18/38). Thus, there were no significant differences in the breadth of the detected responses as a function of sample storage conditions (data not shown).

The Breadth of the EXP1-Specific CD4+ T Cell Response Does Not Correlate With the Clinical Course of Acute Malaria
Several clinical parameters, including parasitaemia, CRP, hemoglobin, and thrombocyte count, were examined, but no significant correlation between the number of T cell responses and any of the parameters could be identified (Figure 2). Similarly, no correlation was found between age or gender and the CD4+ T cell response (Supplementary Figure 5). The patients' origin, the duration of the malaria therapy and the clinical course of the disease (uncomplicated and complicated) did also not correlate with the breadth of the CD4+ T cell response (Supplementary Figures 6A-C). When comparing the number of in vitro responses between different patient groups, we could not see significant differences between patients with their first episode of diagnosed malaria infection (52.9%, 9/17) vs. patients who had a documented history of a past malaria infection (44.4%, 12/27) (Supplementary Figure 6D). Table 4 lists the 45 donors examined in the present study along with their HLA-DRB1 * typing data and summarizes the individual patient-specific EXP1-peptide response patterns. The finding that certain peptides were recognized by a large proportion of subjects with acute P. falciparum infection, despite the fact that they expressed different HLA class II molecules, let us hypothesize that these were comparatively promiscuous MHC class II binding EXP1-specific CD4+ T cell epitopes (i.e., peptides recognized in the context of multiple HLA class II molecules).

HLA Binding and Restriction Experiments
To test this hypothesis, we measured the in vitro capacity of the most frequently recognized EXP1 peptides to bind to a panel of 17 of the most prevalent human HLA-DR molecules, since the majority of HLA class II restricted CD4+ T cell responses are presented by these molecules (49,50). Peptides with an IC50 of 1,000 nM or lower, a threshold previously found associated with the vast majority of HLA class II-restricted T cell epitopes, were considered binders to the respective DR molecule (51-53). As shown in Table 3, in general the peptides located in the middle of the protein (i.e., EXP1-P14-18; aa65-99) were associated with the most promiscuous binding (range: 5-13 of the 17 DR tested). These include EXP1-P14 which bound to five, EXP1-P15 to 13, EXP1-P16 to 10 and EXP1-P17 to nine different HLA molecules ( Table 3).
The most frequently recognized peptides were promiscuous in binding to multiple HLA-DR molecules, with each of the peptides binding between zero and 13 different DR molecules. Furthermore, some of these peptides were able to bind to several HLA-DR molecules with extremely high affinities (IC50 < 10 nM) ( Table 3) 0 nM). This binding pattern is in line with the HLA molecules of the patients that responded to EXP1-P15. Peptides that bound to zero alleles in our DRB in vitro assay may be restricted by DQ or DP molecules.
Supplementary Table 2B shows the most likely HLArestriction considering the in vitro binding data as well as the responders' HLA molecules. Nine out of 12 patients who recognized EXP1-P15 expressed at least one of the HLA molecules mentioned above. The most common HLA molecule expressed in patients of our cohort who recognized EXP1-P15 was DRB1 * 07:01 (six patients). Interestingly, six patients from our cohort also expressed DRB1 * 07:01 but did not recognize EXP1-P15. EXP1-P13 bound to DRB1 * 11:01 in the in vitro assay (142 nM). This is consistent with the fact that out of eight patients who recognized EXP1-P13, five were carrying the HLA-DRB1 * 11 molecule. These results indicate that certain EXP1 peptides, especially those that are located in the center of the protein like EXP1-P15 are potentially promiscuous T cell epitopes.

Fine Mapping of EXP1-P13 and P15
To narrow down the minimal length of the epitopes within EXP1-P13 and P15, we tested the magnitude of the frequency of responding CD4+ T cells by the use of truncations (Figure 3). We first cultivated PBMCs of patients who had shown a strong response against the corresponding peptide with the original peptide (concentration 10 µg/ml) and then tested the magnitude of the IFNγ responses by ICS. The highest CD4+ T cell frequency for EXP1-P13 was reached by patient HH-09 by EXP1-P13 truncation 1 (13-mer truncation: EELVEVNKRKSKY). For EXP1-P15 the 14-mer EXP1-P15 truncation 1 (SKYKLATSVLAGLL) triggered the biggest response by patient HH-03 and in vitro binding data as well as preliminary restriction experiments suggest that this epitope was likely restricted by DRB1 * 04 (Supplementary Figure 7) in this patient. However, restriction by other molecules like DP and DQ molecules were not tested and cannot be excluded as alternative but less likely restricting molecules.
For both EXP1-P15 and EXP1-P20 we also tested variants found on UniProt (32). EXP1-P15 variant 2 contained serine (S) instead of alanine (A) at position 13 and showed the biggest response. EXP1-P20 variant 1 contained arginine (R) instead of histidine (H) at position 14 and EXP1-P20 variant 2 contained arginine (R) instead of lysine (K) at position 11. EXP1-P20 variant 2 triggered a much bigger response than variant 1. These results show that the change of a single amino acid can lead to an improvement of binding and therefore recognition and are in line with prior publications (54). No variants for EXP1-P13 could be found in online databases (Supplementary Table 1).

DISCUSSION
Here, we present an immunological study of the P. falciparum EXP1-specific T cell response on a single epitope level using PBMCs of a large and well-characterized patient cohort. In concordance with the results of previous studies, frequencies of ex vivo EXP1-specific T cell responses were low. However, with our sensitive in vitro testing approach we detected T cell responses in 21 of the 45 patients, and discovered responses directed against 15 of 31 peptides (48.4%) spanning the entire EXP1 protein, making this the first study to comprehensively describe responses against a broad range of EXP1-specific epitopes. For most P. falciparum antigens previous studies only studied vaccine-primed T cell responses in malaria-naïve volunteers within human vaccine trials (25,55) which cannot be directly compared with P. falciparum T cell responses generated during a natural malaria infection. Other studies used small and partly HLA-A2 pre-selected study populations (54). HLA-A2 is the most common expressed class I HLA in Caucasians, but not as commonly expressed in African individuals. As a result, these previous studies have limited applicability for vaccine design and immunomonitoring in malaria endemic countries. Other studies used in silico predicted epitopes or peptide pools instead of comprehensive, overlapping peptide sets (24,56,57). However, this approach does not allow the identification of single epitopes  and immunogenic parts of antigens. Published P. falciparum specific CD4+ and CD8+ T cell epitopes as well as HLArestriction (if known) and newly detected CD4+ T cell epitopes from this study are shown in Supplementary Figure 1. We also assessed important clinical parameters such as the HLA type, origin and parasitaemia in our study cohort, but no significant correlation with the CD4+ T cell response could be shown. Of note, EXP1-P15 (aa70-85) and EXP1-P13 (aa60-74) elicited a CD4+ T cell response in a quarter of our study group that had a diverse HLA-background. The minimal length was experimentally narrowed down in two patients and the most likely restricting HLA-molecule of each epitope was evaluated using our in vitro HLA class II binding data ( Table 3 Table 2A) and comparing it with the HLA molecules of responding patients (Supplementary Table 2B). EXP1-P15 is likely to be restricted by HLA-DRB1 * 01:01, DRB1 * 04:01, DRB1 * 04:02, DRB1 * 04:05, DRB1 * 07:01, DRB1 * 10:01, DRB1 * 13:03, DRB1 * 15:02, and DRB1 * 15:03. Interestingly, six patients in our cohort also expressed DRB1 * 07:01 but did not respond to EXP1-P15. Heterology of the protein sequence of the autologous infecting strain, or slightly differing immunogenic epitope cores between different patients with the same HLA molecules (29) could explain, why some patients were able to recognize a certain epitope unlike others even though they express the same HLA.

), in silico predictions (Supplementary
In this study we have only tested the binding to DRB-molecules, since they restrict the majority of class II responses (∼50%), but acknowledge that restrictions by HLA-DQ and DP molecules are also possible (58). One study examining the HLA-DP, DQ, and DRB3/4/5 restricted responses showed that Dengue virus-specific CD4+ T cell responses were of lower magnitude compared to HLA-DRB1 (58,59). This is consistent with their lower levels of expression (58). Future studies should experimentally confirm the HLA-restriction of these EXP1 T cell epitopes and possibly evaluate restriction by HLA-DP or DQ molecules.
Further detailed in vitro HLA-restriction and peptide truncation experiments to establish optimal peptide length and HLA-restriction (i.e., with single HLA-molecule transfected cell lines or by blockade with anti-DR, and-DQ or anti-DP antibodies) of peptide specificities defined in this study need to be conducted in future follow-up studies.
We still might underestimate the breadth of the P. falciparum EXP1-specific T cell response. In this study design we aimed to expand and define dominant peptide-specific T cell responses producing IFNγ to define the breadth of the CD4+ T cell repertoire after short term cell cultures and using exogenous IL-2 which is known to create a type 1 T helper (Th1) cell cytokine shift (60) and no statement can be made about the original ex vivo phenotype or functionality of these cells. Future studies should therefore further characterize T cell responses by directly assessing ex vivo secretion (e.g., with ELISPOT) of different, additional cytokines (e.g., IL-4, IL-10) that potentially could have been missed using the current approach.
Further subdominant responses might be detectable using peptide sets of peptide variants matching autologous sequences or by applying more sensitive, but far more labor-intensive techniques, like single cell dilution cloning. Patient HH-21 was infected with P. vivax and not with P. falciparum but the EXP1specific cultures nevertheless elicited a CD4+ T cell response against EXP1-P15 indicating that immunity against malaria can indeed be cross-species specific. Of note, the sequence of EXP1-P15 (aa70-85) is well-conserved (Supplementary Table 1), potentially making this epitope an interesting candidate for immunomonitoring by MHC-multimer technology or an epitope to be used for a suitable subunit vaccine.
Also, our analysis is limited to samples drawn directly after hospital admission during acute malaria and it is not known how long each patient had been infected. Of note, if we stratify our data according to patients with and without a response, surprisingly the time after malaria treatment did not significantly differ between either group at this early phase of infection (Supplementary Figure 6B). The single cases for which we were able to analyze the T cell response at later time points would let us hypothesize that the naturally primed EXP1 specific T cell response rapidly wanes over the weeks after successful treatment.
Although EXP1 is likely to be essential for the development and the survival of the parasite, its exact function in the P. falciparum malaria life cycle is far from clear and more investigation as well as linkage between important and conserved regions and immunopotent epitopes are necessary (61). Since EXP1 is expressed at the liver stage during a malaria infection we also expected to detect CD8+ T cell responses. However, all CD8+ T cell responses detected in this study were weak and coincided with a parallel CD4+ T cell response directed against the same peptide in the same patient and we did not follow-up on fine-mapping these subdominant T cell responses (Supplementary Figure 4). It has to be considered that the design of our overlapping peptide set, with an average length of 15 amino acids, is biased toward detection of CD4+ T cell responses and could therefore explain low CD8+ T cell responses. Furthermore, prior publications showed that the blood phase of a malaria infection suppresses immune cells of the liver stage (62,63). This could also explain the lack of CD8+ T cell responses in our cohort since all patients had an extensive blood phase with strong symptoms and the need to be hospitalized. It has been discussed that CD4+ T cells might play an important role in priming other immune cells (i.e., NK cells) and in evoking a strong antibody response which is necessary for immune control (64,65). If an efficient induction of immunity against plasmodial infection by CD8+ T cells depends on CD4+ T cells, is therefore an important question to address in the development of an effective vaccine.
EXP1 is also called circumsporozoite-related antigen due to a similarity between the amino acid sequence of EXP1 and the circumsporozoite protein (CSP). The CSP is by far the most investigated malaria antigen and it is also part of the RTS,S vaccine trial (66). The amino acid sequence NANPDADSESNGEPN of EXP1 is similar to the NANP-repeat region of the CSP which has so far only shown few T cell responses (67,68). Interestingly, we detected T cell responses against EXP1-P25 which is located in this NANP-related region of EXP1 and future studies are necessary to evaluate if not only CD4+ T cell and antibody responses against the NANPrepeat region but also CD8+ T cell responses can be found. The cross reactivity of vaccine-induced CSP and EXP1-primed T cell responses should be tested in future studies.
The RTS,S vaccine only showed short lived efficacy and also naturally acquired immunity only lasts a few months (69). Epidemiological evidence described that antibodies to Plasmodium antigens are inefficiently generated and rapidly lost without continued parasite exposure (70). We also investigated longitudinal follow-up samples of a subset of patients and could show that the EXP1-specific T cell response primed by natural infection waned over time and was not detectable 12 months after the diagnosis and therapy of acute malaria (data not shown). Unfortunately, it was difficult in the present study to collect samples at later timepoints after the patients were discharged. And future, well-structured and prospective longitudinal studies that describe the dynamics of the priming and the breadth, magnitude and quality of the T cell response directed against different malaria antigens are necessary.
This current study aimed to primarily discover novel P. falciparum-specific epitopes with a frequency high enough to be detectable e.g., after enrichment by MHC class II multimer technology. This technology will allow the characterization of the phenotype of T cells which could help to understand the role but also the complications of T cells during malaria infections. In previous studies, EXP1 has shown to generate a strong antibody response in naturally exposed individuals and high antibody titers specific to EXP1 aa73-162 correlated with a high level of IL-6 production to the same peptide (20). This strong recognition by B and T cells suggest that the whole sequence of EXP1 may represent a suitable malaria antigen for a subunit vaccine construct (20).
In summary, a broad range of P. falciparum EXP1-specific CD4+ T cell responses can be detected after in vitro expansion in nearly half of our patients of different origin and within a diverse HLA molecule background. We did not find any correlation of the number of T cell responses and relevant clinical parameters in this cohort. These detailed data on P. falciparum EXP1-specific T cell epitopes will be helpful for the development of tools like MHC class II multimers or to monitor the immune response on an epitope level during future malaria vaccine trials.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
All subjects gave written informed consent in accordance with the Declaration of Helsinki and the study was approved by the local ethics board of the Ärztekammer Hamburg (PV 4238).

ACKNOWLEDGMENTS
We thank all patients who participated in this study. We also thank Jonas Arand, Tom Diedrich, and Maria Sophia Mackroth for helping with the recruitment of the patients, Silke Kummer and Robin Woost for technical assistance, Sophia Schulte for performing further multi-cytokine ICS assays and Georg Lauer for his critical advice and discussion during the revision process.  (31). The sequence of P. vivax EXP1 that resembles P. falciparum EXP1-P15 is framed. Some sequences are only fragments and therefore shorter than 162 amino acids.

SUPPLEMENTARY MATERIAL
Supplementary Table 2 | In silico (A) and most likely (B) binding predictions for EXP1 peptides that elicited a CD4+ T cell response. (A) In silico binding predictions for HLA-DRB1 molecules. In silico binding predictions were performed as described for 15 EXP1 peptides that at least one patient's CD4+ T cells responded to. The threshold for the binding affinity was defined as 5,0; the most probable binders are displayed in bold. The MHC class II in silico binding predictions were made using the IEDB analysis resource consensus tool (38)(39)(40) combining predictions from ANN (41,42) and SMM (43). (B) Most likely HLA-restrictions of the EXP1 peptides that elicited a response as determined by in silico and in vitro predictions matching the patients' HLA molecules.