Our understanding of the human immune response to malaria remains incomplete. Clinical trials using whole-sporozoite-based vaccination approaches such as the Sanaria PfSPZ Vaccine, followed by controlled human malaria infection (CHMI) to assess vaccine efficacy offer a unique opportunity to study the immune response during Plasmodium falciparum infection. Diverse populations of T cells that are not restricted to classical HLA (unconventional T cells) participate in the host response during Plasmodium infection. Although several populations of unconventional T cells exist, the majority of studies focused on TCR Vγ9Vδ2 cells, the most abundant TCR γδ cell population in peripheral blood. In this study, we dissected the response of three TCR γδ cell subsets and mucosal-associated invariant T cells in healthy volunteers immunized with PfSPZ Vaccine and challenged by CHMI using Sanaria PfSPZ Challenge. Using a flow cytometry-based unbiased analysis followed by T cell cloning, several findings were made. Whereas major ex vivo alterations were not detectable after immunization with PfSPZ Vaccine, TCR Vδ2, and mucosal-associated invariant T cells expanded after asexual blood-stage parasitemia induced by CHMI. CHMI, but not vaccination, also induced the activation of TCR Vδ1 and Vδ1Vδ2 γδ T cells. The activated TCR Vδ1 cells were oligoclonal, suggesting clonal expansion, and upon repeated CHMI, showed diminished response, indicating long-term alterations induced by blood-stage parasitemia. Some TCR Vδ1 clones recognized target cells in the absence of parasite-derived Ags, thus suggesting recognition of self-molecules. These findings reveal the articulate participation of different populations of unconventional T cells to P. falciparum infection.

Compared with the year 2010, global malaria disease burden has decreased significantly. In recent years, however, this reduction has stalled and, in some regions, even reversed. In 2017, 219 million cases of malaria were reported, leading to 435,000 deaths worldwide (1).

A valuable tool to fight malaria would be an effective vaccine that leads to long-lasting protection. RTS,S, the most advanced malaria vaccine, received a positive scientific opinion under Article 58 from the European Medicines Agency and is being further assessed in pilot implementation programs in Ghana, Kenya, and Malawi, starting in 2019 (2). However, RTS,S/AS01 provides only partial protection against clinical malaria episodes in African children and infants (3). Recently, Sanaria PfSPZ Vaccine, which is composed of aseptic, purified, cryopreserved, metabolically active, whole, live, irradiation-attenuated Plasmodium falciparum sporozoites (PfSPZ), has led to more promising results in European and United States volunteers (4, 5). Results from clinical trials in malaria pre-exposed volunteers from sub-Saharan Africa indicate significantly lower immunogenicity than achieved in European or United States volunteers using similar vaccination regimen (69). The reasons for the differences in immunogenicity and protective efficacy remain obscure, demonstrating our incomplete understanding of the interactions between malaria pre-exposure, malaria vaccination, and vaccination outcome.

The human immune response to Plasmodium spp. is parasite-stage dependent and involves myeloid cells, NK cells, B cells, and T cells (10). Many studies of T cell biology in human malaria have focused on conventional CD4+ and CD8+ T cells bearing a TCR αβ (10). However, apart from Ab responses (11), no clear correlate of protection either under natural conditions or upon vaccination has been identified (4).

Expansion of TCR γδ T cells upon asexual erythrocytic stage P. falciparum infection has been observed in several settings (1217). In a population of adult malaria patients in Thailand, both Vδ1+ and Vδ1 γδ T cells were found to be expanded after treatment, compared with the day of admission (15). In contrast, an increase in TCR γδ cells observed in Ethiopian malaria patients was mainly attributed to an increase in Vδ1 T cells, but not Vγ9Vδ2 T cells (16). Similar findings were made in a study in Ghanaian children, where an increase of γδ T cells after treatment for malaria was found, an effect that was mainly attributed to expansion of TCR Vδ1 cells. However, length spectratyping of the CDR3δ of these TCR Vδ1 cells did not reveal a dominant public clonotype to be expanded across all patients in a cross-sectional analysis (17).

TCR Vγ9Vδ2 cells are the most abundant TCR γδ cell population in human peripheral blood and are activated in the presence of small, phosphorylated compounds, including P. falciparum–derived Ags, leading to expansion of TCR Vγ9Vδ2 cells in response to P. falciparum both in vitro and in vivo (1821). In addition, T cells expressing the TCR Vγ9Vδ2 present at the time of immunization and after immunization have been suggested to be a possible correlate of protection after PfSPZ Vaccine administration (4, 22).

Mucosal-associated invariant T (MAIT) cells, a second population of unconventional T cells, expand after treatment for asexual erythrocytic stage infection induced by controlled human malaria infection (CHMI) (23). MAIT cells recognize microbial-derived riboflavin metabolites presented by the MHC class I–like molecule MR1 (24) and express semi-invariant TCRs αβ and high levels of CD161.

Taken together, the involvement of unconventional T cell subsets in human malaria infection is evident. However, the diversity of study designs and experimental approaches taken as well as the differences in the examined study populations make it difficult to reach a clear understanding of their functions and contributions to malaria immunity.

To overcome these gaps, we made use of the highly controlled setting of the PfSPZ Vaccine studies BSPZV1 (8) (ClinicalTrials.gov identifier: NCT02132299) and BSPZV2 (25) (ClinicalTrials.gov identifier: NCT02613520) in Tanzania, which included CHMI to assess vaccine efficacy. PBMC collected longitudinally during the study were analyzed by two multicolor flow cytometry panels in combination with an unbiased analysis pipeline to dissect the responses of three major TCR γδ cell subsets and MAIT cells to immunization with PfSPZ Vaccine and CHMI. Our results showed that these T cells are distinctly activated after erythrocytic stage parasitemia in malaria pre-exposed volunteers. We also found that TCR Vδ1 cells show hallmarks of adaptive-like T cells associated with expansion of distinct TCR clonotypes in a donor-dependent manner.

The HC-04 human hepatocyte cell line was obtained from BEI Resources. The human cell lines THP-1 (monocytic leukemia) and HeLa (cervical cancer) were obtained from American Type Culture Collection.

PBMC were isolated from participants in the BSPZV1 clinical trial (ClinicalTrials.gov identifier: NCT02132299). Details are published in Jongo et al. (8). All participants were healthy, male Tanzanians between 18 and 35 y of age. Samples analyzed in this study were derived from the high-dose group that received 2.7 × 105 PfSPZ of PfSPZ Vaccine per injection. CHMI consisted of 3200 PfSPZ of PfSPZ Challenge administered by direct venous inoculation. The second sample set that was analyzed was from the BSPZV2 clinical trial (25) (ClinicalTrials.gov identifier: NCT02613520). From this study, analyzed samples were derived from vaccinated individuals from groups 1a and 1b, who received three times 9 × 105 or three times 1.8 × 106 PfSPZ Vaccine, respectively. Details about the donors are published in Jongo et al. (25).

PBMC were isolated by density gradient centrifugation and were cryopreserved in 90% FCS with 10% DMSO. The cells were stored and transported in liquid nitrogen vapor phase until their usage in the assays.

Monocytes were isolated using a Human CD14 Positive Selection Kit (STEMCELL Technologies) according to the manufacturer’s instructions. Monocyte-derived dendritic cells (Mo-DC) were generated by culturing CD14-positively selected monocytes in the presence of human rIL-4 and GM-CSF (BioLegend) for 5 d. Differentiation was controlled by cell surface staining of CD209 (DC-SIGN).

T cell lines and clones were established as previously described (26).

Flow cytometry staining, analyses, and cell sorting were performed using standard protocols. Data were acquired using an LSR II Fortessa (BD Biosciences). The following Abs were obtained from BioLegend: CCR6 (fluorochrome BV421, clone G034E3), CCR7 (APC/Cy7, G043H7), CD3 (BV650, OKT3), CD4 (Alexa 700, OKT4), CD27 (BV785, O323), CD28 (Alexa 700, CD28.2), CD38 (APC/Cy7, HB-7), CD57 (PE/dazzle, HNK-1), CD69 (PE/Dazzle, HNK-1), CD94 (PerCP/Cy5.5, DX22), CD161 (BV605, HP-3G10), CD209 (PE, 9E9A8), CD294 (BV421, BM16), HLA-DR (Alexy700, L243), KLRG1 (APC, SA231A2), NKp80 (APC, 5D12), pan-γδ (PE, B1), PD-1 (BV785, EH12.2H7), TCR Vα7.2 (BV510, 3C10), and TCR Vδ2 (BV711, B6). Anti–LAG-3 (PE/Cy7, 3DS223H) and LILRB1 (PE/Cy7, GHI/75) were obtained from eBioscience, TCR Vδ1 (FITC, TS8.2) from GeneTex, and CD8 (BUV496, RPA-T8) from BD Biosciences. Cell proliferation was assessed in vitro using the CellTrace Violet Proliferation Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions.

Standard flow cytometry analyses were performed using FlowJo (Tree Star). Dimensionality reduction and unsupervised clustering was performed using a custom R script. Briefly, flow cytometry data were gated on live, single cells, and the subsets of interest (MAIT, TCR Vδ1, TCR Vδ2, or TCR Vδx cells) were exported separately from FlowJo and imported into R. Fluorescent parameters were transformed using an individually centered inverse hyperbolic sine function for each fluorochrome. The t–distributed stochastic neighbor embedding (t-SNE) dimensionality reduction (27) was performed using the Rtsne package (28). Clustering was performed by multiple rounds of density-based spatial clustering of applications with noise (DBSCAN) with custom parameters using the DBSCAN package (29) in R. Then, cell frequencies for each cluster were calculated for each sample and statistical analyses performed. Cluster frequencies between visits were compared using Wilcoxon signed-rank test adjusted for multiple comparisons using the Benjamini–Hochberg method. Additionally, only clusters with a fold change of the mean >2.5 were considered to be significant. To assess vaccination-induced changes, only the vaccinated volunteers were analyzed (n = 20), and to assess CHMI-induced changes, only the volunteers who developed blood-stage parasitemia upon CHMI were included (n = 20).

For the clonality analysis, the D50 value was defined as the number of distinct clonotypes necessary to account for 50% of the total cells within one T cell population. This value was then normalized by the total number of cells per population.

Tree maps were constructed using the R treemapify package (30).

In T cell activation assays with cell lines, 5 × 104 T cells were cocultured with 5 × 104 stimulatory cells in a final volume of 200 μl. Adherent cells were plated and allowed to adhere for 2 h before addition of T cells. T cells and stimulatory cells were cocultured for 16 h.

For proliferation assays, 2 × 105 Mo-DC were cocultured with 1 × 106 PBMC and 2 × 105 uninfected erythrocytes (uRBC) or P. falciparum–infected erythrocytes (PfRBC) in a final volume of 2 ml, and proliferation was assessed after 6 d. Activation of T cell clones was performed by coculture of 5 × 104 Mo-DC and 5 × 104 uRBC or PfRBC in a final volume of 200 μl. For the screening of the T cell clones, T cells were not counted.

Sorting and CDR3δ sequencing were performed as previously described (31). CD38+PD-1+ or CD38PD-1 TCR Vδ1 cells were single-cell sorted into 96-well plates containing 2 μl of SuperScript IV VILO Master Mix (Invitrogen) using a FACSAria (BD Biosciences). The CDR3δ was amplified by nested PCR with the primers as previously described (31) and sequenced using Sanger sequencing (Microsynth).

P. falciparum parasites strain NF54 from fresh cultures were enriched using MACS magnetic columns as published (32). Briefly, fresh parasite cultures were passed on top of the MACS magnetic column and eluted using RMPI 1640 after removing the column from the magnetic stand. Enrichment of infected erythrocytes was assessed by Giemsa staining. Typical parasitemia after enrichment was ∼90%.

We assessed ex vivo frequencies of four groups of unconventional T cells before and after immunization with PfSPZ Vaccine and after CHMI with PfSPZ Challenge that consists of the same strain of P. falciparum (NF54) as PfSPZ Vaccine (Fig. 1A). Cell frequencies were analyzed by multicolor flow cytometry, focusing on a group of 24 adult, male, healthy Tanzanian volunteers who received five doses of 2.7 × 105 PfSPZ Vaccine during the BSPZV1 study. We grouped the TCR γδ cells according to expression of the TCRδ chain into TCR Vδ1, TCR Vδ2 T cells, and TCR Vδ1Vδ2 (Vδx) cells. MAIT cells were identified by the expression of TCR Vα7.2 and high levels of CD161. Between the baseline visit and 2 wk after the fifth vaccination, there were no significant changes in the frequencies of unconventional T cells in peripheral blood (Supplemental Fig. 1A). However, in PBMC collected 28 d after initiation of CHMI, TCR Vδ2 cells and MAIT cells were markedly expanded in the volunteers who developed asexual blood-stage parasitemia (infected individuals), but not in the participants who were protected from blood-stage parasitemia (protected individuals). In contrast, the frequencies of TCR Vδ1 cells and TCR Vδx cells were not significantly altered after CHMI in any group of volunteers (Supplemental Fig. 1B).

FIGURE 1.

Subsets of Vδ1 T cells are activated after asexual blood-stage infection induced by PfSPZ Challenge. (A) Timeline of the clinical trial BSPZV1. (B) t-SNE plot of TCR Vδ1 cells from all volunteers color coded by visit: prevaccination (yellow), postvaccination (green), and post-CHMI (magenta). Regions with increased cell frequency after CHMI by PfSPZ Challenge appear in magenta (n = 24). (C) t-SNE plot after clustering by DBSCAN. Clusters are color coded and assigned numbers. (D) Clusters with significantly altered frequency in infected volunteers after CHMI are highlighted. Clusters are considered significant with p < 0.01 (Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method), and at least 2.5-fold change of the mean frequency across the 20 infected volunteers. (E) Frequency of the five clusters that are significantly increased in infected volunteers after CHMI. Infected volunteers (black dots) are separated from protected volunteers (white dots). Horizontal lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest point still within 1.5 interquartile ranges from the box. Infected, n = 20; protected, n = 4. ****p < 0.0001, ***p < 0.001, **p < 0.01, Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method. (F) Heat map showing the median normalized surface expression of all clusters. The five significantly increased clusters are highlighted with a red frame and labeled with their cluster identifications.

FIGURE 1.

Subsets of Vδ1 T cells are activated after asexual blood-stage infection induced by PfSPZ Challenge. (A) Timeline of the clinical trial BSPZV1. (B) t-SNE plot of TCR Vδ1 cells from all volunteers color coded by visit: prevaccination (yellow), postvaccination (green), and post-CHMI (magenta). Regions with increased cell frequency after CHMI by PfSPZ Challenge appear in magenta (n = 24). (C) t-SNE plot after clustering by DBSCAN. Clusters are color coded and assigned numbers. (D) Clusters with significantly altered frequency in infected volunteers after CHMI are highlighted. Clusters are considered significant with p < 0.01 (Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method), and at least 2.5-fold change of the mean frequency across the 20 infected volunteers. (E) Frequency of the five clusters that are significantly increased in infected volunteers after CHMI. Infected volunteers (black dots) are separated from protected volunteers (white dots). Horizontal lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest point still within 1.5 interquartile ranges from the box. Infected, n = 20; protected, n = 4. ****p < 0.0001, ***p < 0.001, **p < 0.01, Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method. (F) Heat map showing the median normalized surface expression of all clusters. The five significantly increased clusters are highlighted with a red frame and labeled with their cluster identifications.

Close modal

Next, we examined the activation state of each T cell subset after immunization with PfSPZ Vaccine and after CHMI by PfSPZ Challenge ex vivo. To this purpose, we designed a multicolor flow cytometry panel (Supplemental Table I, panel 1) that, in addition to Abs identifying TCR γδ cells and MAIT cells, included Abs targeting NK receptors as well as markers for lymphocyte activation and exhaustion (CD38, CD69, CD94, CD161, HLA-DR, LAG-3, NKp80, and PD-1). We used an unbiased approach to identify significant phenotypic changes upon vaccination and CHMI. First, to reduce the high-dimensional flow cytometry data to two dimensions, we used the t-SNE algorithm (27). Then, clusters of phenotypically similar T cells were identified using the clustering algorithm DBSCAN (33), and the cell frequency within each cluster was calculated. Finally, statistically significant alterations in cell frequency upon vaccination and CHMI were identified. Consistent with the absence of expansion upon immunization with PfSPZ Vaccine (Supplemental Fig. 1), we did not observe significant vaccination-induced phenotypic changes in any T cell subset (Fig. 1B, Supplemental Figs. 2A, 3A, 4A). However, in infected individuals post-CHMI, several clusters of T cells were significantly increased in all T cell subsets. These alterations were not observed in the protected volunteers. The most prominent common feature of all TCR γδ cells and MAIT cells was an increase of CD38 expression in infected volunteers after CHMI, an effect that was not observed in non-γδ, non-MAIT cells (Supplemental Fig. 5). A more detailed analysis revealed distinct patterns of surface marker expression in the different T cell subsets. Within TCR Vδ1 cells, we identified five clusters that were increased after CHMI (Fig. 1D). Notably, these clusters expanded across all 20 infected volunteers after CHMI, but not in the four protected volunteers (Fig. 1E). All clusters were negative for CD161, showed expression of CD38 and PD-1, and, in some cases, coexpressed LAG-3 or HLA-DR, indicative of strong in vivo activation upon erythrocytic stage parasitemia (Fig. 1F). In addition, the increase of CD38+PD-1+ expression on TCR Vδ1 cells after blood-stage parasitemia was positively correlated with the peak parasite density (Supplemental Fig. 6).

Among TCR Vδx cells, we identified two main groups of cells that were expanded in infected volunteers (Supplemental Fig. 2). One group was phenotypically similar to the expanded TCR Vδ1 cells, coexpressing CD38 and PD-1, but not CD161 (Supplemental Fig. 2E). The other group of cells expressed CD161 and CD38, but was negative for PD-1. Thus, the mutually exclusive expression of PD-1 and CD161 divided the Vδx subset into two main groups. The upregulation of CD38+PD-1+ in the Vδ1 and Vδx T cell subsets was in marked contrast with the phenotype of expanded TCR Vδ2 cells and MAIT cells, which were positive for CD38, but negative for PD-1 (Supplemental Figs. 3, 4).

We aimed to confirm the differences in frequency of CD38+PD-1+ T cells using conventional flow cytometry analysis based on the same dataset. Although there was considerable interdonor variability, TCR Vδ1 cells clearly showed the strongest expansion of CD38+PD-1+ cells upon blood-stage parasitemia (Fig. 2). Consistent with the t-SNE analysis, the Vδx T cell subset showed a lower, but still significant, increase in CD38+PD-1+ cells. In contrast, no significant increase of this phenotype was detectable in TCR Vδ2 cells and MAIT cells.

FIGURE 2.

TCR Vδ1 cells and TCR Vδx cells show an increased frequency of CD38+PD-1+ cells after blood-stage infection induced by PfSPZ Challenge. (A) Contour plot of TCR γδ cells and MAIT cells at baseline, postvaccination, and post-CHMI from an infected volunteer with a strong increase of CD38+PD-1+ TCR Vδ1 cells after CHMI. (B) Frequency of CD38+PD-1+ T cells across all infected volunteers (n = 20). Lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest point still within 1.5 interquartile ranges from the box. Data were analyzed by two-way ANOVA corrected for multiple comparisons using Dunnett test. ****p < 0.0001, *p = 0.011.

FIGURE 2.

TCR Vδ1 cells and TCR Vδx cells show an increased frequency of CD38+PD-1+ cells after blood-stage infection induced by PfSPZ Challenge. (A) Contour plot of TCR γδ cells and MAIT cells at baseline, postvaccination, and post-CHMI from an infected volunteer with a strong increase of CD38+PD-1+ TCR Vδ1 cells after CHMI. (B) Frequency of CD38+PD-1+ T cells across all infected volunteers (n = 20). Lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest point still within 1.5 interquartile ranges from the box. Data were analyzed by two-way ANOVA corrected for multiple comparisons using Dunnett test. ****p < 0.0001, *p = 0.011.

Close modal

To investigate the differentiation state of TCR γδ cells and MAIT cells during immunization with PfSPZ Vaccine and CHMI, we designed a second flow cytometry panel that included Abs specific for CD27, CD28, CD57, LILRB1, KLRG1, CCR6, and CCR7 (Supplemental Table I, panel 2). Using the same detailed analysis including dimensionality reduction and clustering, we confirmed that there are no significant ex vivo phenotypic alterations upon immunization with PfSPZ Vaccine (Fig. 3A, Supplemental Fig. 7). In contrast, in infected individuals post-CHMI, TCR Vδ1 cells showed an increased frequency of four clusters, defined by the lack of surface expression of CD27, CD28, CCR6, and CCR7, indicative of an effector phenotype (Fig. 3). Interestingly, the expanded cells were also negative for CD57, an Ag that is often present on late effector cells (34), in particular on chronically stimulated T cells that have reached replicative senescence (35, 36). The CD27CD57 TCR Vδ1 cells that expanded after blood-stage parasitemia could be further subdivided by the expression of KLRG1, a marker that is upregulated on effector cells differentiating into senescent cells (Fig. 3E). When KLRG1 and CD57 were measured on total CD27 TCR Vδ1 cells, both markers were found to be expressed on the large majority of cells (Supplemental Fig. 8A), confirming increased expression on Ag-experienced, late effector cells. Thus, the absence of CD27 and CD57 expression on the expanded T cells might identify a population of recently differentiated early effector T (TEE) cells in contrast to chronically stimulated CD27CD57+ terminal effector T cells. The expansion upon blood-stage parasitemia was specific to the CD27CD57 subset, as the frequency of CD27CD57+ TCR Vδ1 cells remained unchanged during the course of the CHMI (Supplemental Fig. 8B).

FIGURE 3.

Vδ1 TEE cells are expanded after erythrocytic stage infection induced by CHMI. (A) t-SNE plot of TCR Vδ1 cells from all volunteers color coded by visit: prevaccination (yellow), postvaccination (green), and post-CHMI (magenta). Regions with increased cell frequency after CHMI appear in magenta (n = 24). (B) t-SNE plot after clustering by DBSCAN. Clusters are color coded and assigned numbers. (C) Clusters with significantly altered frequency after CHMI are highlighted. Clusters are considered significant with p < 0.01 (Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method) and at least 2.5-fold change of the mean. (D) Frequency of the four clusters that are significantly increased in infected volunteers after CHMI. Infected volunteers (black dots) are separated from protected volunteers (white dots). Horizontal lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest point still within 1.5 interquartile ranges from the box. Infected, n = 20; protected, n = 4. ***p < 0.001, **p < 0.01, Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method. (E) Heat map showing the median normalized surface expression of all clusters. The four significantly increased clusters are highlighted with a red frame and labeled with their cluster identification.

FIGURE 3.

Vδ1 TEE cells are expanded after erythrocytic stage infection induced by CHMI. (A) t-SNE plot of TCR Vδ1 cells from all volunteers color coded by visit: prevaccination (yellow), postvaccination (green), and post-CHMI (magenta). Regions with increased cell frequency after CHMI appear in magenta (n = 24). (B) t-SNE plot after clustering by DBSCAN. Clusters are color coded and assigned numbers. (C) Clusters with significantly altered frequency after CHMI are highlighted. Clusters are considered significant with p < 0.01 (Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method) and at least 2.5-fold change of the mean. (D) Frequency of the four clusters that are significantly increased in infected volunteers after CHMI. Infected volunteers (black dots) are separated from protected volunteers (white dots). Horizontal lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest point still within 1.5 interquartile ranges from the box. Infected, n = 20; protected, n = 4. ***p < 0.001, **p < 0.01, Wilcoxon signed-rank test adjusted with the Benjamini–Hochberg method. (E) Heat map showing the median normalized surface expression of all clusters. The four significantly increased clusters are highlighted with a red frame and labeled with their cluster identification.

Close modal

An increase in the frequency of TEE cells in infected volunteers was detected in TCR Vδ1 and Vδx cells, but not in Vδ2 or MAIT cells (Fig. 4). This finding is reminiscent of the increased expression of CD38+PD-1+ that occurred only on TCR Vδ1 and Vδx cells (see Fig. 2). Indeed, the increase in frequency of these two phenotypes upon CHMI correlated significantly (Supplemental Fig. 9B, 9D), whereas no association was detectable upon immunization with PfSPZ Vaccine (Supplemental Fig. 9A, 9C).

FIGURE 4.

TCR Vδ1 cells and TCR Vδx cells show an increased frequency of early effector cells after erythrocytic stage infection induced by CHMI. (A) Contour plot of TCR γδ cells and MAIT cells at baseline, postvaccination, and post-CHMI from a representative, infected donor. (B) Frequency of CD27CD57 cells across all infected volunteers (n = 20). Horizontal lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest points still within 1.5 interquartile ranges from the box. Data were analyzed by two-way ANOVA corrected for multiple comparisons using Dunnett test. ****p < 0.0001, **p = 0.0026.

FIGURE 4.

TCR Vδ1 cells and TCR Vδx cells show an increased frequency of early effector cells after erythrocytic stage infection induced by CHMI. (A) Contour plot of TCR γδ cells and MAIT cells at baseline, postvaccination, and post-CHMI from a representative, infected donor. (B) Frequency of CD27CD57 cells across all infected volunteers (n = 20). Horizontal lines represent the median value, boxes span the interquartile range, and whiskers extend to the furthest points still within 1.5 interquartile ranges from the box. Data were analyzed by two-way ANOVA corrected for multiple comparisons using Dunnett test. ****p < 0.0001, **p = 0.0026.

Close modal

To confirm that CD38+PD-1+ and CD27CD57 T cells represented the same subset, Abs against those markers were combined in a single flow cytometry panel (Supplemental Table I, panel 3) and used to analyze PBMC from adult volunteers enrolled into the clinical trial BSPZV2. In this study, volunteers underwent three immunizations with PfSPZ Vaccine, followed by two consecutive homologous CHMIs to assess vaccine efficacy (Fig. 5A). The first CHMI (CHMI no. 1) was conducted 3–11 wk after last immunization with PfSPZ Vaccine, and the second CHMI (CHMI no. 2) was conducted 37–41 wk after last vaccination.

FIGURE 5.

TCR Vδ1 TEE cells are specifically activated after asexual erythrocytic stage infection and show long-lasting alterations. (A) Timeline of the BSPZV2 cohort. (B) Dot plot of CD38+PD-1+ TCR Vδ1 cells of a volunteer who was infected after both CHMIs. (C) Frequency of CD38+PD-1+ Vδ1 TEE cells and CD38+PD-1+ Vδ1 non-TEE cells in volunteers who were infected after both CHMIs (n = 3). Plotted is the mean ± SEM. ***p = 0.002, *p < 0.05. (D) Fold expansion of CD38+PD-1+ Vδ1 TEE cells and other CD38+PD-1+ Vδ1 T cells upon CHMI no. 1. **p = 0.0099. (E) Fold expansion of CD38+PD-1+ Vδ1 TEE cells during CHMI no. 1 and CHMI no. 2. Fold changes were compared by ratio paired t test. Shown are the three volunteers who were infected after CHMI no. 1 as determined by thick blood smear. ***p = 0.0003. Frequencies between visits were compared by one-way ANOVA, adjusted for multiple comparisons using the Holm–Sidak method.

FIGURE 5.

TCR Vδ1 TEE cells are specifically activated after asexual erythrocytic stage infection and show long-lasting alterations. (A) Timeline of the BSPZV2 cohort. (B) Dot plot of CD38+PD-1+ TCR Vδ1 cells of a volunteer who was infected after both CHMIs. (C) Frequency of CD38+PD-1+ Vδ1 TEE cells and CD38+PD-1+ Vδ1 non-TEE cells in volunteers who were infected after both CHMIs (n = 3). Plotted is the mean ± SEM. ***p = 0.002, *p < 0.05. (D) Fold expansion of CD38+PD-1+ Vδ1 TEE cells and other CD38+PD-1+ Vδ1 T cells upon CHMI no. 1. **p = 0.0099. (E) Fold expansion of CD38+PD-1+ Vδ1 TEE cells during CHMI no. 1 and CHMI no. 2. Fold changes were compared by ratio paired t test. Shown are the three volunteers who were infected after CHMI no. 1 as determined by thick blood smear. ***p = 0.0003. Frequencies between visits were compared by one-way ANOVA, adjusted for multiple comparisons using the Holm–Sidak method.

Close modal

We first assessed whether asexual blood-stage infection was associated with specific activation and expansion of Vδ1 TEE cells. Indeed, CD38+PD-1+ Vδ1 TEE cells expanded selectively in the volunteers who developed blood-stage parasitemia after CHMI no. 1, whereas the frequency of other (non-TEE) CD38+PD-1+ TCR Vδ1 cells remained unchanged (Fig. 5B–D). Consistent with these findings, in the volunteers who developed blood-stage parasitemia only after CHMI no. 2, an increase of CD38+PD-1+ Vδ1 TEE cells was detectable only after CHMI no. 2 (Supplemental Fig. 10). Again, the effect was specific for CD38+PD-1+ TEE cells.

Strikingly, these alterations were long-lasting, as the frequency of CD38+PD-1+ Vδ1 TEE cells was still elevated before initiation of CHMI no. 2, 6 mo after CHMI no. 1 (Fig. 5C, left panel). Furthermore, fold expansions of CD38+PD-1+ Vδ1 TEE cells upon CHMI no. 1 and CHMI no. 2 in the volunteers who became infected during both CHMIs were compared. Whereas a strong expansion was observed during CHMI no. 1, no significant increase occurred upon CHMI no. 2 in the same volunteers (Fig. 5E). Taken together, these findings indicate an altered immune response upon parasite rechallenge and suggested long-term functional alterations of TCR Vδ1 cells upon malaria infection. A similar analysis of TCR Vδx cells in the same cohort led to comparable results (Supplemental Fig. 11).

Interestingly, in this cohort, we detected a slight, but significant, decrease in the frequency of TCR Vδ1 and Vδx cells upon immunization with PfSPZ Vaccine. However, this observation was not associated with protection (Supplemental Fig. 12).

Loss of CD27 expression on TCR Vδ1 cells has been linked with oligoclonal expansion, accompanied by functional changes such as acquisition of cytotoxicity (31). We hypothesized that the observed expansion of CD38+PD-1+ Vδ1 TEE cells after blood-stage parasitemia was also associated with oligoclonal expansion, possibly following Ag-dependent stimulation. Therefore, we assessed whether expansion of CD38+PD-1+ TCR Vδ1 cells upon blood-stage parasitemia was accompanied by expansion of distinct Vδ1 clonotypes. We selected the three volunteers from the BSPZV1 cohort with the highest increase of CD38+PD-1+ expression in TCR Vδ1 cells upon asexual blood-stage parasitemia. From each of the three volunteers, we sorted 32 single TCR Vδ1 cells from both the CD38+PD-1+ and CD38PD-1 subsets of PBMC-collected post-CHMI. The CDR3 of the TCRδ chains was determined, and the clonotype frequencies were analyzed (Supplemental Table II). In all three examined volunteers, the CD38+PD-1+ TCR Vδ1 cells were more oligoclonal than the CD38PD-1 population, probably as a consequence of clonotype-dependent activation of TCR Vδ1 cells after asexual erythrocytic stage parasitemia (Fig. 6).

FIGURE 6.

TCR Vδ1 cells are oligoclonally expanded after asexual blood-stage infection. (A) Tree maps indicating CDR3δ clonotype usage of CD38+PD-1+ and CD38PD-1 TCR Vδ1 cells after CHMI. Three infected BSPZV1 donors were chosen because of their strong expansion of CD38+PD-1+ TCR Vδ1 cells post-CHMI. (B) Clonality is quantified by the D50 value, the number of clonotypes required to cover 50% of analyzed cells, normalized by the total number of analyzed cells per condition. Data were analyzed by paired, two-tailed Student t test. **p = 0.009.

FIGURE 6.

TCR Vδ1 cells are oligoclonally expanded after asexual blood-stage infection. (A) Tree maps indicating CDR3δ clonotype usage of CD38+PD-1+ and CD38PD-1 TCR Vδ1 cells after CHMI. Three infected BSPZV1 donors were chosen because of their strong expansion of CD38+PD-1+ TCR Vδ1 cells post-CHMI. (B) Clonality is quantified by the D50 value, the number of clonotypes required to cover 50% of analyzed cells, normalized by the total number of analyzed cells per condition. Data were analyzed by paired, two-tailed Student t test. **p = 0.009.

Close modal

Our data suggested that the expansion of CD38+PD-1+ TCR Vδ1 cells after asexual blood-stage parasitemia could be due to Ag-dependent activation of these cells. Thus, we aimed to assess T cell expansion upon parasite exposure using an in vitro model. Mo-DC from four BSPZV1 donors were used to stimulate autologous PBMC in the presence of either uRBC or PfRBC. PfRBC were infected with the P. falciparum strain NF54 that is also used in PfSPZ Vaccine and PfSPZ Challenge. Robust proliferation of CD4+ T cells and TCR Vδ2 T cells was observed after stimulation with Mo-DC and PfRBC for 6 d (Fig. 7A). MAIT cells also proliferated, albeit to a lesser extent (Fig. 7A). Mo-DC and uRBC failed to induce comparable responses. Interestingly, a marked PfRBC-dependent proliferation was observed in TCR Vδ1 and Vδx cells (Fig. 7A). Reminiscent of the ex vivo phenotype observed after CHMI, in vitro proliferating TCR Vδ1 and Vδx cells showed increased expression of CD38 and PD-1 (Fig. 7B).

FIGURE 7.

TCR Vδ1 cells proliferate and upregulate CD38 and PD-1 in response to Mo-DC and PfRBC as well as distinct human cell lines. (A) Proliferation of PBMC from BSPZV1 volunteers in response to autologous Mo-DC and PfRBC or uRBC. Proliferation was assessed after 6 d of coculture (n = 4). Histograms show CellTrace Violet staining of T cells from one representative volunteer. Filled histograms, coculture with Mo-DC and PfRBC; dashed lines, coculture with Mo-DC and uRBC. ****p < 0.0001, **p = 0.0037. (B) Frequency of CD38+PD-1+ cells in proliferating and nonproliferating cells after coculture with Mo-DC and PfRBC. ****p < 0.0001. (C) Percentage of CD137+ T cells upon coculture of Vδ1 T cell clones or TCRαβ+ CD4+ T cell clones with autologous Mo-DC and uRBC or PfRBC (upper panel). Rescreen of selected candidate clones using IFN-γ ELISA (lower panel). ****p < 0.0001. (D) Frequency of proliferating TCR Vδ1 cells and CD38+PD-1+ TCR Vδ1 cells (left panels) as well as TCR Vδ2 cells (right panels) upon coculture of ex vivo PBMC with indicated cell lines. PBMC from three BSPZV1 donors were derived from prevaccination, postvaccination, and post-CHMI visits and cocultured for 6 d with the cell lines. Shown is the mean ± SEM. Data were analyzed by two-way ANOVA, adjusted for multiple comparisons using the Holm–Sidak methods.

FIGURE 7.

TCR Vδ1 cells proliferate and upregulate CD38 and PD-1 in response to Mo-DC and PfRBC as well as distinct human cell lines. (A) Proliferation of PBMC from BSPZV1 volunteers in response to autologous Mo-DC and PfRBC or uRBC. Proliferation was assessed after 6 d of coculture (n = 4). Histograms show CellTrace Violet staining of T cells from one representative volunteer. Filled histograms, coculture with Mo-DC and PfRBC; dashed lines, coculture with Mo-DC and uRBC. ****p < 0.0001, **p = 0.0037. (B) Frequency of CD38+PD-1+ cells in proliferating and nonproliferating cells after coculture with Mo-DC and PfRBC. ****p < 0.0001. (C) Percentage of CD137+ T cells upon coculture of Vδ1 T cell clones or TCRαβ+ CD4+ T cell clones with autologous Mo-DC and uRBC or PfRBC (upper panel). Rescreen of selected candidate clones using IFN-γ ELISA (lower panel). ****p < 0.0001. (D) Frequency of proliferating TCR Vδ1 cells and CD38+PD-1+ TCR Vδ1 cells (left panels) as well as TCR Vδ2 cells (right panels) upon coculture of ex vivo PBMC with indicated cell lines. PBMC from three BSPZV1 donors were derived from prevaccination, postvaccination, and post-CHMI visits and cocultured for 6 d with the cell lines. Shown is the mean ± SEM. Data were analyzed by two-way ANOVA, adjusted for multiple comparisons using the Holm–Sidak methods.

Close modal

To further characterize the proliferating TCR Vδ1 cells and investigate their responsiveness to P. falciparum Ags, we interrogated T cell clones derived from sorted cells that proliferated after stimulation with Mo-DC and PfRBC. One group of cells was sorted as TCR γδ+ Vδ2 negative and a control group as CD4+ T cells. Both Vδ1 and CD4+ TCRαβ+ clones were established and screened for activation in response to autologous Mo-DC and PfRBC. The activation marker that was selected for this screening is CD137, whose upregulation allows for detection of Ag-specific T cells (37). The CD4+ T cell clones showed a consistent increase of CD137 expression only upon stimulation with PfRBC, indicating Ag-dependent activation and proving the validity of our approach to isolate P. falciparum–responsive T cell clones (Fig. 7C, upper panel). In contrast, the Vδ1 T cell clones showed marked expression of CD137 when cocultured with autologous Mo-DC, independently of the presence of blood-stage parasites (Fig. 7C, upper panel). Similar reactivity was observed when TCR Vδ1 clones were investigated for IFN-γ release (Fig. 7C, lower panel). These unexpected findings indicate that at least some TCR Vδ1 cells that expanded upon coculture with Mo-DC and PfRBC are activated by self-encoded ligands independently of parasite-derived Ags. The PfRBC-dependent proliferation observed with freshly isolated PBMC might have been facilitated by soluble factors or additive signals provided by CD4+ and TCR Vδ2 cells responding to P. falciparum Ags.

To further investigate the possible self-reactivity of TCR Vδ1 cells, we assessed the ex vivo response of TCR Vδ1 cells to THP-1, HC-04 and HeLa cell lines, which are derived from different human tissues. PBMC collected at baseline, postvaccination, and post-CHMI during the BSPZV1 study were cocultured for 6 d with each cell line. Analysis by flow cytometry showed that TCR Vδ1 cells proliferated and upregulated CD38 and PD-1 (Fig. 7D, left panels), an effect that was almost absent in the TCR Vδ2 cell subset (Fig. 7D, right panels). Whereas all cell lines induced some degree of proliferation and activation, the strongest response was induced by the hepatocyte-derived cell line HC-04. Furthermore, activation and proliferation of TCR Vδ1 cells was remarkably pronounced in PBMC derived from the post-CHMI visit of donor A, who displayed the highest frequency of CD38+PD-1+ TCR Vδ1 cells and profound oligoclonal expansion of TCR Vδ1 cells upon CHMI (see Fig. 6).

CHMI conducted in malaria pre-exposed populations is an excellent tool to dissect malaria parasite–host interactions under highly defined conditions (38). The role of T cells during malaria remains poorly understood. In particular, the function of unconventional T cells, comprising CD1- and MR1-restricted T cells (39), TCR Vγ9Vδ2 cells stimulated by butyrophilin 3A1 (40), and other subsets of TCR γδ cells remains unclear. We analyzed these populations in depth and found important changes for some of them. During CHMI in P. falciparum pre-exposed adult Tanzanian volunteers, we confirmed previous findings that T cells expressing the TCR Vγ9Vδ2 heterodimer as well as MAIT cells expand upon asexual blood-stage parasitemia (12, 23).

We also investigated TCR γδ cells expressing a TCR Vδ1 chain and the TCR γδ populations expressing other Vδ chains (in this study indicated as Vδx), which represent a minority of TCR γδ cells in most individuals (41). Very little is known regarding the Ag specificity and the restriction molecules of TCR Vδ1 and Vδx cells (42). In addition, their function in infectious diseases, including malaria, remains to be determined (17). TCR Vδ1 cells may recognize different types of Ags and expand clonally after Ag recognition. When activated, these T cells release a variety of cytokines, may exert helper functions, and may also kill target cells, thus resembling MHC-restricted adaptive T cells (43). Activated TCR γδ cells may also express surface markers identical to those expressed by MHC-restricted T cells. We found that only TCR Vδ1 and Vδx cells showed expansion of the CD38+PD-1+ population, a phenotype not observed on expanded TCR Vδ2 and MAIT cells. Both CD38 and PD-1 are markers of cell activation, and although they exert different functions, they both contribute to regulation of expressing T cells. The unique coexpression of CD38 and PD-1 implies a qualitatively distinct activation of TCR Vδ1 and Vδx cells upon CHMI compared with other expanded T cells such as MAIT and TCR Vγ9Vδ2 cells.

Both CD38 and PD-1 can be upregulated by soluble factors (44, 45), and their expression is more efficiently induced on Ag-stimulated T cells either within the tumor microenvironment or during chronic infectious diseases (46, 47). The presence of both CD38 and PD-1 might result from a continuous Ag stimulation in donors who developed blood-stage parasitemia, possibly following recognition of either parasite-derived Ags or self-antigens induced by P. falciparum infection. Alternatively, the expansion of unique T cells might be initially Ag driven and then facilitated by cytokines present in the microenvironment, thus representing a bystander effect. For this reason, we investigated the TCR repertoire of the expanded TCR Vδ1 CD38+PD-1+ cells. Our analysis was performed on single cells to reduce the bias introduced by the variable quantities of TCR gene mRNAs present in different cells. This approach forced us to limit the study to a relatively small number of cells. We investigated ∼30 individual T cells isolated from both TCR Vδ1 CD38+PD-1+ and TCR Vδ1 CD38PD-1 populations of three donors. The data clearly indicated a significantly reduced breadth of TCR repertoire in the CD38+PD-1+ cells in comparison with TCR Vδ1 CD38PD-1 cells in all examined donors.

The possibility that the expansion of distinct clonotypes is Ag-driven is also indicated by the noted alterations of the T cell differentiation state upon erythrocytic stage parasitemia. Naive T cells generally express CD27 and CD28 but are negative for CD57 (34, 48). Upon repeated antigenic stimulation, expression of CD28 and CD27 is lost, and concomitantly, CD57 expression is gained. Late effector or senescent T cells are mostly CD27CD28CD57+ (49). We found that expanded TCR Vδ1 and Vδx cells are CD27CD28CD57, thus representing a relatively rare phenotype of Ag-experienced cells that are in a transition stage and are not yet terminally differentiated and have not reached replicative senescence (49). CD28CD57 T cells have been described in the context of HIV infection, retain proliferative capacity, and have limited replicative history (35). This phenotype of intermediate differentiation fits well with Ag-dependent activation of TCR Vδ1 and Vδx cells in volunteers who develop blood-stage parasitemia after CHMI.

Within TCR Vδx cells, we found a second cell cluster that was significantly expanded after erythrocytic stage infection and was characterized by the absence of PD-1 and expression of CD161 (Supplemental Fig. 2). The nature of this cell population is difficult to assign. CD161 is usually expressed on both TCR αβ and γδ cells with high responsiveness to IL-12 and IL-18 stimulation (50). During asexual blood-stage malaria, both these cytokines are produced in large amounts (51), and it is tempting to speculate that this second population of TCR Vδx CD161+ cells expands during this stage. Also MAIT cells are prone to cytokine-mediated activation (52), and a similar cytokine-dependent mechanism of activation and proliferation might explain our finding of increased MAIT cells in volunteers who develop blood-stage parasitemia during CHMI. This possibility is supported by the fact that P. falciparum does not encode the enzymes of the riboflavin biosynthesis pathway (53) necessary for generation of the canonical MAIT cell Ags (54).

In the BSPZV2 cohort, we detected a slight, but significant, decrease in the frequency of Vδ1 and Vδx T cells postvaccination (Supplemental Fig. 12A, 12B). This might imply that these T cells get recruited to the liver during early liver stage, when the development of sporozoites from PfSPZ Vaccine is thought to arrest (55). Their decrease in peripheral blood was not associated with the protection status of the volunteers (Supplemental Fig. 12C, 12D). The accumulation of CD38+PD-1+ Vδ1 and Vδx T cells post-CHMI might indicate that Ag-stimulation occurs between mid–liver stage and blood stage. Analysis of samples from volunteers who received PfSPZ Challenge during antimalarial chemoprophylaxis (56), in which P. falciparum is allowed to develop through the entire liver stage, might help to narrow down the time point of Vδ1 T cell stimulation and/or their migration to peripheral blood.

Of note, the phenotypic alterations that we detected were exclusive to volunteers who experienced a recent asexual blood-stage infection, whereas volunteers who remained blood-stage negative after CHMI did not show any detectable changes. These findings indicate that the stimulatory Ags that induce the expansion of TCR Vδ1 and Vδx CD38+PD-1+ cells might be of microbial origin. Alternatively, other classes of Ags of self-origin might be expressed during the blood-stage infection and stimulate these cells. Consistent with both of these hypotheses, the volunteers with the strongest increase of CD38+PD-1+ TCR Vδ1 cells after CHMI were the ones with the highest peak parasitemia (Supplemental Fig. 6). Our in vitro experiments showed that TCR Vδ1 and Vδx cell expansion depended on the presence of P. falciparum–infected RBC. Thus, P. falciparum was important for the proliferation of these cells, possibly through an indirect effect, such as the induction of the release of cytokines by P. falciparum–specific non-Vδ1 T cells. However, TCR Vδ1 clones established from CHMI donors also reacted to Mo-DC in the absence of P. falciparum, suggesting a self-reactive capacity. In addition, fresh TCR Vδ1, but not Vδ2, cells from CHMI donors responded to the hepatocyte HC-04 cell line and, in a reduced manner, to the myelomonocytic cell line THP-1 in the absence of P. falciparum, thus confirming the ability of these cells to recognize self-molecules. These findings may indicate that, during the liver stage, some mechanisms already operate to activate the immune response. Our unpublished, preliminary results of vaccination by PfSPZ Challenge during chemoprophylaxis of malaria pre-exposed volunteers in Equatorial Guinea during the EGSPZV2 trial indicate that liver-resident, malaria-specific immunity might be induced under natural malaria exposure (Olotu et al., submitted for publication; ClinicalTrials.gov identifier: NCT02859350). Furthermore, data from mouse models suggest that the pre-erythrocytic stage of malaria is not immunologically silent but that liver cells may sense Plasmodium infection through recognition of Plasmodium RNA and respond with release of type I IFNs (57). Thus, liver-resident TCR Vδ1 cells might also be alerted by innate immune mechanisms during P. falciparum liver stage.

Our findings raised multiple questions regarding the nature of the Ags recognized by self-reactive TCR Vδ1 cells, the elements that restrict their response, and the physiological role of these T cells. All these aspects warrant further investigations.

The expansion of TCR Vδ1 cells in nonprotected patients argues in favor of an indirect role for these T cells during malaria immunity. An intriguing possibility is that they are involved in the enhancement of the immune response within the organs where they become activated. In a normal human liver, TCR γδ cells are very abundant (10–15% of total CD3+ cells), and TCR Vδ1 cells represent the largest population of TCR γδ cells (58). These cells release a variety of effector molecules, including perforin and granzyme B, and are oligoclonal. Whether they recognize Ags derived from gut microbiota or self-antigens was not investigated in these studies.

The expansion of TCR Vδ1 cells in peripheral blood of malaria patients and their release of a variety of cytokines, including IFN-γ, TNF, and IL-10, has been observed (14, 15, 17, 59). TCR Vδ1 and Vδx cells that become activated after blood-stage parasitemia might participate in recruiting other effector cells, as recently suggested (60). For example, they might release chemokines and M-CSF that attract and enhance the microbicidal activity of macrophages, as found in mice infected with P. chabaudi (61). Finally, TCR Vδ1 cells might directly inhibit the in vitro replication of P. falciparum, as previously shown (62), even not recognizing parasite-derived Ags.

In conclusion, we found that populations of TCR Vδ1 and Vδx cells become activated and show long-term phenotypic changes in volunteers who develop blood-stage parasitemia. These TCR γδ cells show features similar to classical adaptive T cells, and at least a fraction of them are activated in the absence of parasite-derived molecules. It will be important to identify the mechanisms of activation to shed light on the biological functions of these T cell populations.

We thank the members of the Flow Cytometry Core Facility at the Department of Biomedicine for their collaboration. We thank Sergio Wittlin and Christian Scheurer of Swiss Tropical and Public Health Institute for supply with P. falciparum–infected RBC cultures. We thank L. Mori for scientific discussions and P. Cullen for revision of the manuscript.

This work was supported by Grant SNF 310030-173240 and EU Horizon 2020 TBVAC2020 (Grant 643381) and DBM core funding to G.D.L. This work was supported by the Swiss Vaccine Research Institute, Switzerland, to C.D.

The online version of this article contains supplemental material.

Abbreviations used in this article:

CHMI

controlled human malaria infection

DBSCAN

density-based spatial clustering of applications with noise

MAIT

mucosal-associated invariant T

Mo-DC

monocyte-derived dendritic cell

PfRBC

P. falciparum–infected erythrocyte

PfSPZ

metabolically active, whole, live, irradiation-attenuated Plasmodium falciparum sporozoite

TEE

early effector T

t-SNE

t–distributed stochastic neighbor embedding

uRBC

uninfected erythrocyte.

1
World Health Organization
.
2018
. World malaria report 2018. Geneva, Switzerland: World Health Organization. Available at https://apps.who.int/iris/handle/10665/275867.
2
World Health Organization
.
2018
.
Malaria vaccine: WHO position paper, January 2016 - recommendations.
Vaccine
36
:
3576
3577
.
3
RTS,S Clinical Trials Partnership
.
2015
.
Efficacy and safety of RTS,S/AS01 malaria vaccine with or without a booster dose in infants and children in Africa: final results of a phase 3, individually randomised, controlled trial. [Published erratum appears in 2015 Lancet 386: 30.]
Lancet
386
:
31
45
.
4
Seder
,
R. A.
,
L.-J.
Chang
,
M. E.
Enama
,
K. L.
Zephir
,
U. N.
Sarwar
,
I. J.
Gordon
,
L. A.
Holman
,
E. R.
James
,
P. F.
Billingsley
,
A.
Gunasekera
, et al
VRC 312 Study Team
.
2013
.
Protection against malaria by intravenous immunization with a nonreplicating sporozoite vaccine.
Science
341
:
1359
1365
.
5
Lyke
,
K. E.
,
A. S.
Ishizuka
,
A. A.
Berry
,
S.
Chakravarty
,
A.
DeZure
,
M. E.
Enama
,
E. R.
James
,
P. F.
Billingsley
,
A.
Gunasekera
,
A.
Manoj
, et al
.
2017
.
Attenuated PfSPZ vaccine induces strain-transcending T cells and durable protection against heterologous controlled human malaria infection.
Proc. Natl. Acad. Sci. USA
114
:
2711
2716
.
6
Sissoko
,
M. S.
,
S. A.
Healy
,
A.
Katile
,
F.
Omaswa
,
I.
Zaidi
,
E. E.
Gabriel
,
B.
Kamate
,
Y.
Samake
,
M. A.
Guindo
,
A.
Dolo
, et al
.
2017
.
Safety and efficacy of PfSPZ vaccine against Plasmodium falciparum via direct venous inoculation in healthy malaria-exposed adults in Mali: a randomised, double-blind phase 1 trial.
Lancet Infect. Dis.
17
:
498
509
.
7
Epstein
,
J. E.
,
K. M.
Paolino
,
T. L.
Richie
,
M.
Sedegah
,
A.
Singer
,
A. J.
Ruben
,
S.
Chakravarty
,
A.
Stafford
,
R. C.
Ruck
,
A. G.
Eappen
, et al
.
2017
.
Protection against Plasmodium falciparum malaria by PfSPZ vaccine.
JCI Insight
2
: e89154.
8
Jongo
,
S. A.
,
S. A.
Shekalaghe
,
L. W. P.
Church
,
A. J.
Ruben
,
T.
Schindler
,
I.
Zenklusen
,
T.
Rutishauser
,
J.
Rothen
,
A.
Tumbo
,
C.
Mkindi
, et al
.
2018
.
Safety, immunogenicity, and protective efficacy against controlled human malaria infection of Plasmodium falciparum sporozoite vaccine in Tanzanian adults.
Am. J. Trop. Med. Hyg.
99
:
338
349
.
9
Olotu
,
A.
,
V.
Urbano
,
A.
Hamad
,
M.
Eka
,
M.
Chemba
,
E.
Nyakarungu
,
J.
Raso
,
E.
Eburi
,
D. O.
Mandumbi
,
D.
Hergott
, et al
.
2018
.
Advancing global health through development and clinical trials partnerships: a randomized, placebo-controlled, double-blind assessment of safety, tolerability, and immunogenicity of PfSPZ vaccine for malaria in healthy Equatoguinean men.
Am. J. Trop. Med. Hyg.
98
:
308
318
.
10
Crompton
,
P. D.
,
J.
Moebius
,
S.
Portugal
,
M.
Waisberg
,
G.
Hart
,
L. S.
Garver
,
L. H.
Miller
,
C.
Barillas-Mury
,
S. K.
Pierce
.
2014
.
Malaria immunity in man and mosquito: insights into unsolved mysteries of a deadly infectious disease.
Annu. Rev. Immunol.
32
:
157
187
.
11
Cohen
,
S.
,
I. A.
McGregor
,
S.
Carrington
.
1961
.
Gamma-globulin and acquired immunity to human malaria.
Nature
192
:
733
737
.
12
Ho
,
M.
,
H. K.
Webster
,
P.
Tongtawe
,
K.
Pattanapanyasat
,
W. P.
Weidanz
.
1990
.
Increased γ δ T cells in acute Plasmodium falciparum malaria.
Immunol. Lett.
25
:
139
141
.
13
Roussilhon
,
C.
,
M.
Agrapart
,
J.-J.
Ballet
,
A.
Bensussan
.
1990
.
T lymphocytes bearing the γ δ T cell receptor in patients with acute Plasmodium falciparum malaria.
J. Infect. Dis.
162
:
283
285
.
14
Roussilhon
,
C.
,
M.
Agrapart
,
P.
Guglielmi
,
A.
Bensussan
,
P.
Brasseur
,
J. J.
Ballet
.
1994
.
Human TcR γ δ+ lymphocyte response on primary exposure to Plasmodium falciparum.
Clin. Exp. Immunol.
95
:
91
97
.
15
Ho
,
M.
,
P.
Tongtawe
,
J.
Kriangkum
,
T.
Wimonwattrawatee
,
K.
Pattanapanyasat
,
L.
Bryant
,
J.
Shafiq
,
P.
Suntharsamai
,
S.
Looareesuwan
,
H. K.
Webster
,
J. F.
Elliott
.
1994
.
Polyclonal expansion of peripheral gamma delta T cells in human Plasmodium falciparum malaria.
Infect. Immun.
62
:
855
862
.
16
Worku
,
S.
,
A.
Björkman
,
M.
Troye-Blomberg
,
L.
Jemaneh
,
A.
Färnert
,
B.
Christensson
.
1997
.
Lymphocyte activation and subset redistribution in the peripheral blood in acute malaria illness: distinct gammadelta+ T cell patterns in Plasmodium falciparum and P. vivax infections.
Clin. Exp. Immunol.
108
:
34
41
.
17
Hviid
,
L.
,
J. A. L.
Kurtzhals
,
V.
Adabayeri
,
S.
Loizon
,
K.
Kemp
,
B. Q.
Goka
,
A.
Lim
,
O.
Mercereau-Puijalon
,
B. D.
Akanmori
,
C.
Behr
.
2001
.
Perturbation and proinflammatory type activation of V δ 1(+) γ δ T cells in African children with Plasmodium falciparum malaria.
Infect. Immun.
69
:
3190
3196
.
18
Goerlich
,
R.
,
G.
Häcker
,
K.
Pfeffer
,
K.
Heeg
,
H.
Wagner
.
1991
.
Plasmodium falciparum merozoites primarily stimulate the V γ 9 subset of human γ/δ T cells.
Eur. J. Immunol.
21
:
2613
2616
.
19
Behr
,
C.
,
P.
Dubois
.
1992
.
Preferential expansion of V gamma 9 V delta 2 T cells following stimulation of peripheral blood lymphocytes with extracts of Plasmodium falciparum.
Int. Immunol.
4
:
361
366
.
20
Behr
,
C.
,
R.
Poupot
,
M. A.
Peyrat
,
Y.
Poquet
,
P.
Constant
,
P.
Dubois
,
M.
Bonneville
,
J. J.
Fournie
.
1996
.
Plasmodium falciparum stimuli for human gammadelta T cells are related to phosphorylated antigens of mycobacteria.
Infect. Immun.
64
:
2892
2896
.
21
Costa
,
G.
,
S.
Loizon
,
M.
Guenot
,
I.
Mocan
,
F.
Halary
,
G.
de Saint-Basile
,
V.
Pitard
,
J.
Déchanet-Merville
,
J.-F.
Moreau
,
M.
Troye-Blomberg
, et al
.
2011
.
Control of Plasmodium falciparum erythrocytic cycle: γδ T cells target the red blood cell-invasive merozoites.
Blood
118
:
6952
6962
.
22
Zaidi
,
I.
,
H.
Diallo
,
S.
Conteh
,
Y.
Robbins
,
J.
Kolasny
,
S.
Orr-Gonzalez
,
D.
Carter
,
B.
Butler
,
L.
Lambert
,
E.
Brickley
, et al
.
2017
.
γδ T cells are required for the induction of sterile immunity during irradiated sporozoite vaccinations.
J. Immunol.
199
:
3781
3788
.
23
Mpina
,
M.
,
N. J.
Maurice
,
M.
Yajima
,
C. K.
Slichter
,
H. W.
Miller
,
M.
Dutta
,
M. J.
McElrath
,
K. D.
Stuart
,
S. C.
De Rosa
,
J. P.
McNevin
, et al
.
2017
.
Controlled human malaria infection leads to long-lasting changes in innate and innate-like lymphocyte populations.
J. Immunol.
199
:
107
118
.
24
Keller
,
A. N.
,
A. J.
Corbett
,
J. M.
Wubben
,
J.
McCluskey
,
J.
Rossjohn
.
2017
.
MAIT cells and MR1-antigen recognition.
Curr. Opin. Immunol.
46
:
66
74
.
25
Jongo
,
S. A.
,
L. W. P.
Church
,
A. T.
Mtoro
,
S.
Chakravarty
,
A. J.
Ruben
,
P. A.
Swanson
,
K. R.
Kassim
,
M.
Mpina
,
A.-M.
Tumbo
,
F. A.
Milando
, et al
.
2019
.
Safety and differential antibody and T-cell responses to the Plasmodium falciparum sporozoite malaria vaccine, PfSPZ vaccine, by age in Tanzanian adults, adolescents, children, and infants.
Am. J. Trop. Med. Hyg.
100
:
1433
1444
.
26
Lepore
,
M.
,
A.
Kalinichenko
,
S.
Calogero
,
P.
Kumar
,
B.
Paleja
,
M.
Schmaler
,
V.
Narang
,
F.
Zolezzi
,
M.
Poidinger
,
L.
Mori
,
G.
De Libero
.
2017
.
Functionally diverse human T cells recognize non-microbial antigens presented by MR1. [Published erratum appears in 2017 Elife 6.]
eLife
6
1433
1444
.
27
van der Maaten
,
L.
,
G.
Hinton
.
2008
.
Visualizing data using t-SNE.
J. Mach. Learn. Res.
9
:
2579
2605
.
28
Krijthe
,
J. H.
2015
. Rtsne: T-distributed stochastic neighbor embedding using a Barnes-Hut implementation. R package version 0.15. Available at https://github.com/jkrijthe/Rtsne. Accessed: September 30, 2019.
29
Hahsler
,
M.
,
M.
Piekenbrock
.
2018
. dbscan: Fast density-based clustering with R. J. Stat. Softw.
91
:
1
30
.
30
Wilkins
,
D.
2018
. treemapify: Draw Treemaps in “ggplot2.” R package version 2.5.2. Available at https://CRAN.R-project.org/package=treemapify. Accessed: May 31, 2019.
31
Davey
,
M. S.
,
C. R.
Willcox
,
S. P.
Joyce
,
K.
Ladell
,
S. A.
Kasatskaya
,
J. E.
McLaren
,
S.
Hunter
,
M.
Salim
,
F.
Mohammed
,
D. A.
Price
, et al
.
2017
.
Clonal selection in the human Vδ1 T cell repertoire indicates γδ TCR-dependent adaptive immune surveillance.
Nat. Commun.
8
:
14760
.
32
Coronado
,
L. M.
,
N. M.
Tayler
,
R.
Correa
,
R. M.
Giovani
,
C.
Spadafora
.
2013
.
Separation of Plasmodium falciparum late stage-infected erythrocytes by magnetic means.
J. Vis. Exp.
73
: e50342.
33
Ester
,
M.
,
H.-P.
Kriegel
,
J.
Sander
,
X.
Xu
.
1996
. A density-based algorithm for discovering clusters in large spatial databases with noise. In Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining,
AAAI Press
,
Palo Alto, CA
, p.
226
231
.
34
Mahnke
,
Y. D.
,
T. M.
Brodie
,
F.
Sallusto
,
M.
Roederer
,
E.
Lugli
.
2013
.
The who’s who of T-cell differentiation: human memory T-cell subsets.
Eur. J. Immunol.
43
:
2797
2809
.
35
Brenchley
,
J. M.
,
N. J.
Karandikar
,
M. R.
Betts
,
D. R.
Ambrozak
,
B. J.
Hill
,
L. E.
Crotty
,
J. P.
Casazza
,
J.
Kuruppu
,
S. A.
Migueles
,
M.
Connors
, et al
.
2003
.
Expression of CD57 defines replicative senescence and antigen-induced apoptotic death of CD8+ T cells.
Blood
101
:
2711
2720
.
36
Wherry
,
E. J.
,
M.
Kurachi
.
2015
.
Molecular and cellular insights into T cell exhaustion.
Nat. Rev. Immunol.
15
:
486
499
.
37
Wolfl
,
M.
,
J.
Kuball
,
W. Y.
Ho
,
H.
Nguyen
,
T. J.
Manley
,
M.
Bleakley
,
P. D.
Greenberg
.
2007
.
Activation-induced expression of CD137 permits detection, isolation, and expansion of the full repertoire of CD8+ T cells responding to antigen without requiring knowledge of epitope specificities.
Blood
110
:
201
210
.
38
Stanisic
,
D. I.
,
J. S.
McCarthy
,
M. F.
Good
.
2017
.
Controlled human malaria infection: applications, advances, and challenges.
Infect. Immun.
86
 e00479-17.
39
Mori
,
L.
,
M.
Lepore
,
G.
De Libero
.
2016
.
The immunology of CD1- and MR1-restricted T cells.
Annu. Rev. Immunol.
34
:
479
510
.
40
De Libero
,
G.
,
S.-Y.
Lau
,
L.
Mori
.
2015
.
Phosphoantigen presentation to TCR γδ cells, a conundrum getting less gray zones.
Front. Immunol.
5
:
679
.
41
Casorati
,
G.
,
G.
De Libero
,
A.
Lanzavecchia
,
N.
Migone
.
1989
.
Molecular analysis of human gamma/delta+ clones from thymus and peripheral blood.
J. Exp. Med.
170
:
1521
1535
.
42
Davey
,
M. S.
,
C. R.
Willcox
,
A. T.
Baker
,
S.
Hunter
,
B. E.
Willcox
.
2018
.
Recasting human Vδ1 lymphocytes in an adaptive role.
Trends Immunol.
39
:
446
459
.
43
Vantourout
,
P.
,
A.
Hayday
.
2013
.
Six-of-the-best: unique contributions of γδ T cells to immunology.
Nat. Rev. Immunol.
13
:
88
100
.
44
Mihara
,
K.
,
T.
Yoshida
,
S.
Ishida
,
Y.
Takei
,
A.
Kitanaka
,
K.
Shimoda
,
K.
Morishita
,
Y.
Takihara
,
T.
Ichinohe
.
2016
.
All-trans retinoic acid and interferon-α increase CD38 expression on adult T-cell leukemia cells and sensitize them to T cells bearing anti-CD38 chimeric antigen receptors.
Blood Cancer J.
6
: e421.
45
Bally
,
A. P. R.
,
J. W.
Austin
,
J. M.
Boss
.
2016
.
Genetic and epigenetic regulation of PD-1 expression.
J. Immunol.
196
:
2431
2437
.
46
Barber
,
D. L.
,
E. J.
Wherry
,
D.
Masopust
,
B.
Zhu
,
J. P.
Allison
,
A. H.
Sharpe
,
G. J.
Freeman
,
R.
Ahmed
.
2006
.
Restoring function in exhausted CD8 T cells during chronic viral infection.
Nature
439
:
682
687
.
47
Wykes
,
M. N.
,
S. R.
Lewin
.
2018
.
Immune checkpoint blockade in infectious diseases.
Nat. Rev. Immunol.
18
:
91
104
.
48
Larbi
,
A.
,
T.
Fulop
.
2014
.
From “truly naïve” to “exhausted senescent” T cells: when markers predict functionality.
Cytometry A
85
:
25
35
.
49
Strioga
,
M.
,
V.
Pasukoniene
,
D.
Characiejus
.
2011
.
CD8+ CD28- and CD8+ CD57+ T cells and their role in health and disease.
Immunology
134
:
17
32
.
50
Fergusson
,
J. R.
,
K. E.
Smith
,
V. M.
Fleming
,
N.
Rajoriya
,
E. W.
Newell
,
R.
Simmons
,
E.
Marchi
,
S.
Björkander
,
Y.-H.
Kang
,
L.
Swadling
, et al
.
2014
.
CD161 defines a transcriptional and functional phenotype across distinct human T cell lineages.
Cell Rep.
9
:
1075
1088
.
51
Lyke
,
K. E.
,
R.
Burges
,
Y.
Cissoko
,
L.
Sangare
,
M.
Dao
,
I.
Diarra
,
A.
Kone
,
R.
Harley
,
C. V.
Plowe
,
O. K.
Doumbo
,
M. B.
Sztein
.
2004
.
Serum levels of the proinflammatory cytokines interleukin-1 beta (IL-1beta), IL-6, IL-8, IL-10, tumor necrosis factor alpha, and IL-12(p70) in Malian children with severe Plasmodium falciparum malaria and matched uncomplicated malaria or healthy controls.
Infect. Immun.
72
:
5630
5637
.
52
Ussher
,
J. E.
,
M.
Bilton
,
E.
Attwod
,
J.
Shadwell
,
R.
Richardson
,
C.
de Lara
,
E.
Mettke
,
A.
Kurioka
,
T. H.
Hansen
,
P.
Klenerman
,
C. B.
Willberg
.
2014
.
CD161++ CD8+ T cells, including the MAIT cell subset, are specifically activated by IL-12+IL-18 in a TCR-independent manner.
Eur. J. Immunol.
44
:
195
203
.
53
Kanehisa
,
M.
,
S.
Goto
.
2000
.
KEGG: kyoto encyclopedia of genes and genomes.
Nucleic Acids Res.
28
:
27
30
.
54
Kjer-Nielsen
,
L.
,
O.
Patel
,
A. J.
Corbett
,
J.
Le Nours
,
B.
Meehan
,
L.
Liu
,
M.
Bhati
,
Z.
Chen
,
L.
Kostenko
,
R.
Reantragoon
, et al
.
2012
.
MR1 presents microbial vitamin B metabolites to MAIT cells.
Nature
491
:
717
723
.
55
Richie
,
T. L.
,
P. F.
Billingsley
,
B. K. L.
Sim
,
E. R.
James
,
S.
Chakravarty
,
J. E.
Epstein
,
K. E.
Lyke
,
B.
Mordmüller
,
P.
Alonso
,
P. E.
Duffy
, et al
.
2015
.
Progress with Plasmodium falciparum sporozoite (PfSPZ)-based malaria vaccines.
Vaccine
33
:
7452
7461
.
56
Mordmüller
,
B.
,
G.
Surat
,
H.
Lagler
,
S.
Chakravarty
,
A. S.
Ishizuka
,
A.
Lalremruata
,
M.
Gmeiner
,
J. J.
Campo
,
M.
Esen
,
A. J.
Ruben
, et al
.
2017
.
Sterile protection against human malaria by chemoattenuated PfSPZ vaccine.
Nature
542
:
445
449
.
57
Liehl
,
P.
,
V.
Zuzarte-Luís
,
J.
Chan
,
T.
Zillinger
,
F.
Baptista
,
D.
Carapau
,
M.
Konert
,
K. K.
Hanson
,
C.
Carret
,
C.
Lassnig
, et al
.
2014
.
Host-cell sensors for Plasmodium activate innate immunity against liver-stage infection.
Nat. Med.
20
:
47
53
.
58
Hunter
,
S.
,
C. R.
Willcox
,
M. S.
Davey
,
S. A.
Kasatskaya
,
H. C.
Jeffery
,
D. M.
Chudakov
,
Y. H.
Oo
,
B. E.
Willcox
.
2018
.
Human liver infiltrating γδ T cells are composed of clonally expanded circulating and tissue-resident populations.
J. Hepatol.
69
:
654
665
.
59
Taniguchi
,
T.
,
K.
Md Mannoor
,
D.
Nonaka
,
H.
Toma
,
C.
Li
,
M.
Narita
,
V.
Vanisaveth
,
S.
Kano
,
M.
Takahashi
,
H.
Watanabe
.
2017
.
A unique subset of γδ T cells expands and produces IL-10 in patients with naturally acquired immunity against falciparum malaria.
Front. Microbiol.
8
:
1288
.
60
Deroost
,
K.
,
J.
Langhorne
.
2018
.
Gamma/delta T cells and their role in protection against malaria.
Front. Immunol.
9
:
2973
.
61
Mamedov
,
M. R.
,
A.
Scholzen
,
R. V.
Nair
,
K.
Cumnock
,
J. A.
Kenkel
,
J. H. M.
Oliveira
,
D. L.
Trujillo
,
N.
Saligrama
,
Y.
Zhang
,
F.
Rubelt
, et al
.
2018
.
A macrophage colony-stimulating-factor-producing γδ T cell subset prevents malarial parasitemic recurrence.
Immunity
48
:
350
363.e7
.
62
Elloso
,
M. M.
,
H. C.
van der Heyde
,
J. A.
vande Waa
,
D. D.
Manning
,
W. P.
Weidanz
.
1994
.
Inhibition of Plasmodium falciparum in vitro by human gamma delta T cells.
J. Immunol.
153
:
1187
1194
.

S.L.H. and B.K.L. are salaried and full-time employees of Sanaria, Inc., the developer and sponsor of Sanaria PfSPZ Vaccine. The other authors have no financial conflicts of interest.