Enhancement of dendritic cell activation via CD40 ligand-expressing γδ T cells is responsible for protective immunity to Plasmodium parasites

Abstract

Previous reports have shown that γδ T cells are important for the elimination of malaria parasites in humans and mice. However, how γδ T cells are involved in protective immunity against blood-stage malaria remains unknown. We infected γδ T-cell–deficient (TCRδ-KO) mice and control wild-type mice with Plasmodium berghei XAT, which is a nonlethal strain. Although infected red blood cells were eliminated within 30 d after infection, TCRδ-KO mice could not clear the infected red blood cells, showed high parasitemia, and eventually died. Therefore, γδ T cells are essential for clearance of the parasites. Here, we found that γδ T cells play a key role in dendritic cell activation after Plasmodium infection. On day 5 postinfection, γδ T cells produced IFN-γ and expressed CD40 ligand during dendritic cell activation. These results suggest that γδ T cells enhance dendritic cell activation via IFN-γ and CD40 ligand–CD40 signaling. This hypothesis is supported strongly by the fact that in vivo induction of CD40 signaling prevented the death of TCRδ-KO mice after infection with P. berghei XAT. This study improves our understanding of protective immunity against malaria and provides insights into γδ T-cell–mediated protective immunity against various infectious diseases.

There are two types of T-cell receptor (TCR)–expressing T cells, namely, αβ and γδ T cells. γδ T cells are thought to function as innate immune cells, which are the first line of defense against infectious pathogens (1). γδ T cells play important roles in immune responses against various infectious diseases, including those associated with protozoan parasites, bacteria, and viruses (2–5). In humans and mice, the number of γδ T cells in peripheral blood and spleen increases after infection with malaria parasites (6–8). Furthermore, in vitro experiments have demonstrated that γδ T cells recognize malaria antigens (7). Therefore, γδ T cells may have protective or pathological effects on blood-stage malaria. We previously reported that mice depleted of γδ T cells using mAb could not eliminate the nonlethal malaria parasite strain Plasmodium berghei XAT (9), which is an attenuated variant of the lethal strain P. berghei NK65 (10). Therefore, γδ T cells are important for protective immunity against blood-stage malaria (9). However, the precise cellular and molecular mechanisms of the γδ T-cell–involved protective immunity against malaria parasites remain unknown.

Previous reports have revealed requirements of other immune cells and cytokines for the elimination of P. berghei XAT. For example, depletion of CD4⁺ αβ T cells by mAbs and depletion of macrophages by a specific chemical failed to eliminate malaria parasites (11, 12). IFN-γ, but not IL-4, is essential for the elimination of malaria parasites (11, 13). Thus, Th1 cells that produce IFN-γ are critical for protective immunity. IFN-γ activates phagocytes and IgG2a production from B cells, thereby clearing P. berghei XAT parasites. Another key cytokine for protective immunity against P. berghei XAT is IL-12, which is produced by dendritic cells (DCs) after stimulation with components of the infectious organism such as LPS or CpG oligonucleotides (14). IL-12 induces naive CD4⁺ T cells to differentiate into IFN-γ–expressing Th1 cells (15). Furthermore, it stimulates natural killer (NK) cells and CD8⁺ T cells (14). However, NK cells and CD8⁺ T cells are not required for the elimination of P. berghei XAT (11, 12). Furthermore, the depletion of only NK cells would not lead to a change in the IFN-γ levels in the case of P. berghei XAT infection (11). These reports suggest that IFN-γ production from NK cells and CD8⁺ T cells is compensated for by the other cell types, especially by CD4⁺ T cells. The CD40 ligand (CD40L) is an important molecule that is ligated to CD40 on DCs and is involved in the production of IL-12 in DCs (16). Antigen stimulation and CD40/CD40L signaling synergistically induce IL-12 production in DCs (17). CD40L likely is expressed on active CD4⁺ T cells that are not induced upon initial infection. Therefore, the presence of a third type of cell (besides DCs and naive T cells) is important for activating DCs and producing IL-12 for the Th1 immune response to eliminate pathogens in vivo.

Previous in vitro studies of malaria have shown that expansion of human γδ T cells in peripheral blood depends on the presence of CD4⁺ T cells (18). In addition, in vivo studies using a rodent malaria model have shown that increasing the number of splenic γδ T cells requires CD4⁺ αβ T cells to produce IL-2 ∼2–3 wk after infection (19, 20). Thus, these reports suggest that CD4⁺ αβ T cells regulate expansion of γδ T cells. On the other hand, it has been demonstrated that malarial antigens from Plasmodium falciparum stimulate and proliferate γδ T cells in human peripheral blood, although the receptor that binds the malaria antigens on γδ T cells remains unknown. These reports imply that γδ T cells respond immediately to Plasmodium infection, produce proinflammatory cytokines, and facilitate the activation of other immune cells by responding directly to malaria parasites. However, the correlation between γδ T cells and CD4⁺ αβ T cells and the precise timing of their activation require further investigation.

The aim of this study was to elucidate the mechanism of γδ T-cell–associated protective immunity against blood-stage malaria. Using γδ T-cell depletion by anti-TCRγδ mAb, we found that γδ T cells are necessary for the elimination of P. berghei XAT parasites during the early stages of infection. Next, using γδ T-cell–deficient (TCRδ-KO) mice, we showed that γδ T cells enhance DC activation through CD40/CD40L signaling during the early stages of infection. The function of γδ T cells is to induce Th1 cell differentiation and increase the number of phagocytes, resulting in clearance of P. berghei XAT.

Results

γδT Cells Play a Critical Role in Host Survival During P. berghei XAT Infection.

To confirm that γδ T cells are required for clearance of blood-stage P. berghei XAT, we inoculated P. berghei XAT-infected red blood cells (iRBCs) into TCRδ-KO mice and control WT mice. WT mice cleared the iRBCs by day 30 postinfection (p.i.) after showing fluctuating parasitemia (Fig. 1A). On the other hand, high parasitemia was observed in TCRδ-KO mice during the observation period. WT mice did not die from P. berghei XAT infection, but TCRδ-KO mice began to die from day 18 p.i., and all TCRδ-KO mice died by day 40 p.i. (Fig. 1B). These results indicate that γδ T cells play a key role in clearing P. berghei XAT parasites.

Fig. 1.

γδ T cells are required for elimination of P. berghei XAT. (A and B) TCRδ-KO and C57BL/6 WT mice were infected with P. berghei XAT through i.v. inoculation of blood-stage parasites (10⁴ infected RBCs). (A) Time-course analysis of parasitemia (n = 5). (B) Kaplan–Meier survival curves (n = 15). (C and D) C57BL/6 WT mice were depleted of γδ T cells or CD4⁺ T cells by administration of anti-TCRγδ mAb or anti-CD4 mAb from day 0 (d-0) or day 9 (d-9) p.i. with P. berghei XAT (10⁴ infected RBCs). (C) Time-course analysis of parasitemia (n = 5). (D) Kaplan–Meier survival curves (WT anti-CD4 day 9 p.i., n = 13; WT anti-TCRγδ day 9 p.i. and day 0, n = 5). Data are representative of two or three independent experiments.

To determine when during infection γδ T cells function as protective immune cells, γδ T cells were depleted from mice using anti-TCR γδ mAb after infection with P. berghei XAT. When γδ T cells were depleted from mice at day 0 p.i., the mice could not clear the parasites and eventually died. In contrast, when γδ T cells were depleted at day 9 p.i., all mice cleared the parasites and survived (Fig. 1 C and D). These results suggest that, at least before day 9 p.i., γδ T cells act as critical protective immune cells for the clearance of P. berghei XAT parasites.

Next, to determine whether CD4⁺ T cells function as the effector cells of protective immunity against P. berghei XAT even after day 9 p.i., we depleted CD4⁺ T cells by sequential administration of anti-CD4 mAb after day 9 p.i. Because of the immunodepletion of CD4⁺ T cells after day 9 p.i., all mice showed high parasitemia and died by day 30 p.i. (Fig. 1 C and D). These results suggest that CD4⁺ T cells were indeed the effector cells of protective immunity against P. berghei XAT.

IFN-γ Production by Th1 Cells Is Influenced by γδ T Cells.

After Plasmodium infection, naive CD4⁺ T cells (Th0 cells) differentiate into a large number of Th1 cells that produce IFN-γ. Thus, to investigate whether TCRδ-KO mice had impaired Th1 cell differentiation after P. berghei XAT infection, we performed an intracellular cytokine assay to estimate the proportion of splenic CD4⁺ T cells producing IFN-γ. In WT mice, the proportion of splenic T cells expressing IFN-γ increased significantly on days 7 and 9 p.i. This increasing proportion also occurred in TCRδ-KO mice, but to significantly lower levels (Fig. 2 A and B). The number of splenic CD4⁺ T cells increased significantly from day 9 p.i. in both groups, but on day 9 p.i. there were significantly fewer CD4⁺ T cells in TCRδ-KO mice than in WT mice (Fig. 2C). Based on these data, we estimated the number of splenic CD4⁺ T cells producing IFN-γ in both groups. WT mice had more cells of this type than did TCRδ-KO mice on days 7 and 9 p.i. (Fig. 2D). The levels of IFN-γ were higher in the supernatants of cultivated spleen cells from WT mice than in those from TCRδ-KO mice on days 7 and 9 p.i. (Fig. 2E). Consistent with these data, IFN-γ levels in the plasma of WT mice were increased on days 7 and 9 p.i., whereas those of TCRδ-KO mice were not (Fig. 2F). These data suggest that γδ T cells regulate the differentiation of naive CD4⁺ T cells into Th1 cells.

Fig. 2.

γδ T cells are required for Th1 cell differentiation after infection with P. berghei XAT. TCRδ-KO mice and C57BL/6 WT mice were infected with P. berghei XAT through i.v. inoculation of blood-stage parasites (10⁴ infected RBCs). (A) Zebra plots represent CD3⁺- and CD4⁺-gated spleen cells from WT mice and TCRδ-KO mice on day 0 (naive) and day 7 p.i. Numbers indicate the proportion of CD3⁺ CD4⁺ T cells (CD4⁺ T) expressing IFN-γ. (B) Proportions of IFN-γ–expressing CD4⁺ T cells in spleens from WT mice and TCRδ-KO mice on days 0, 5, 7, 9, and 14 p.i.(n = 3). **P < 0.01. (C) Absolute number of IFN-γ–expressing CD4⁺ T cells in spleens from WT mice and TCRδ-KO mice on days 0, 3, 5, 7, 9, and 14 p.i.(n = 3–4). *P < 0.05. (D) Estimation of the number of CD4⁺ T-cells expressing IFN-γ in spleens from each mouse group after infection (n = 3). **P < 0.01. (E) IFN-γ concentration in the supernatant of cultivated splenocytes from each mouse group on days 0, 3, 5, 7, 9, and 11 p.i. (n = 3). *P < 0.05, **P < 0.01. (F) IFN-γ concentration in plasma from each mouse group on days 0, 7, and 9 p.i. (n = 3). **P < 0.01. Data are representative of three or four independent experiments.

Activation and IL-12 Production of DCs Are Regulated by γδ T Cells After Infection with P. berghei XAT.

DCs play a key role in the development of protective immunity against malaria (21). Activation of DCs is essential for the differentiation of naive T cells into effector T cells that facilitate clearing of malaria parasites. Because DCs generally proliferate after pathogen stimulation, we estimated the number of DCs at various times p.i. to investigate whether activation of DCs is induced by P. berghei XAT infection. The number of conventional DCs (cDC; CD11c⁺ B220⁻ CD19⁻ CD3⁻) (Fig. 3A) and plasmacytoid DCs (pDC; CD11c^int B220⁺ CD19⁻ CD3⁻) (Fig. 3B) in spleens of WT mice increased transiently on day 5 p.i. The expression of MHC Class II and the costimulatory molecules CD40, CD80, and CD86 (22) also are indicators of DC activation. Thus, we measured the expression of these indicators in cDCs and pDCs in spleens from WT mice after infection using flow cytometry. Flow cytometric analysis revealed that the expression levels were significantly higher in cDCs than in pDCs in both naive and infected mice (Fig. 3C). Expression of CD40, CD80, and CD86 in both DC types increased transiently on day 5 p.i., suggesting that cDCs and pDCs are activated on day 5 p.i. MHC-II expression also increased transiently in cDCs on day 5 p.i. but was reduced in pDCs on days 5 and 7 p.i. (Fig. 3C).

Fig. 3.

γδ T cells enhance DC activation after infection with P. berghei XAT. (A) Dot plots showing cDC and pDC populations in spleens from naive WT mice. (B–G) TCRδ-KO mice and C57BL/6 WT mice were infected with P. berghei XAT through i.v. inoculation of blood-stage parasites (10⁴ infected RBCs). (B) Absolute numbers of cDCs and pDCs in spleens from each mouse group on days 0 (naive), 3, 5, and 7 p.i. (n = 3–4). *P < 0.05, **P < 0.01. (C) Mean fluorescent intensity (MFI) of FITC-conjugated mAbs against MHC-II and costimulatory molecules (CD40, CD80, CD86) on cDCs and pDCs in spleens from each mouse group on days 0 (naive), 3, 5, and 7 p.i. The MFI levels indicate expression levels of each molecule in cDCs and pDCs (n = 3–4). *P < 0.05, **P < 0.01. (D) IL-12 concentration in the supernatant of cultivated splenocytes from each mouse group on days 0, 5, and 7 p.i. (n = 3). *P < 0.05, **P < 0.01. (E) Intracellular cytokine assay of cDCs and pDCs in spleens from each mouse group on days 0 (naive) and 5 p.i. Numbers represent the proportion of–cDCs and pDCs producing IL-12. (F) Proportion of cDCs and pDCs in spleens from each mouse group expressing IL-12 on days 0 and 5 p.i. (n = 3). *P < 0.05, **P < 0.01. (G) Estimation of the number of cDCs and pDCs expressing IL-12 in spleens from each mouse group after infection (n = 3). *P < 0.05, **P < 0.01. Data are representative of three or four independent experiments.

Next, to investigate whether γδ T cells regulate activation of DCs after infection, we compared the number of DCs and expression levels of the indicators on DCs in spleens from WT and TCRδ-KO mice after infection. Although both cDCs and pDCs were increased on day 5 p.i. in spleens from TCRδ-KO mice, the levels were significantly lower than in spleens from WT mice (Fig. 3B). On day 5 p.i. expression levels of all indicators were lower in cDCs of TCRδ-KO mice than in cDCs of WT mice (Fig. 3C). For pDCs, significant differences in expression levels of MHC–II and CD80 were not observed between WT and TCRδ-KO mice. However, the expression levels of CD40 and CD86 in TCRδ-KO mice were reduced on day 5 p.i. compared with those in WT mice (Fig. 3C). These results suggest that γδT cells regulate activation of DCs after P. berghei XAT infection and that the effect is stronger in cDCs than in pDCs.

After activation, DCs produce IL-12, a key cytokine that induces Th1 immune responses (15). Therefore, we compared IL-12 production in DCs from WT and TCRδ-KO mice. First, we measured IL-12 levels in the supernatant of spleen cells cultivated with P. berghei XAT antigens (Fig. 3D). Spleen cells of naive WT and TCRδ-KO mice produced similar levels of IL-12. On days 5 and 7 p.i. spleen cells from infected WT mice had significantly higher levels of IL-12 than spleen cells from naive WT mice. Although IL-12 levels were similar in spleen cells from infected and naive TCRδ-KO mice on day 5 p.i., the levels on day 7 p.i. were slightly higher in cells from infected TCRδ-KO mice than in cells from naive TCRδ-KO mice. (Fig. 3D). On days 5 and 7 p.i. IL-12 levels were significantly lower in the supernatant of spleen cells cultivated from TCRδ-KO mice than in cells cultivated from WT mice (Fig. 3D). To confirm that DCs produce IL-12 after infection and to determine if the ability of splenic DCs to produce IL-12 is influenced by γδ T cells, we performed an intracellular cytokine assay (Fig. 3 E and F). The proportion of splenic DCs producing IL-12 was similar in naive WT and TCRδ-KO mice (Fig. 3F). Consistent with the data shown in Fig. 3D, the proportion of cDCs producing IL-12 in WT mice on day 5 p.i. was increased significantly compared with naive WT mice. On the contrary, the proportion in TCRδ-KO mice on day 5 p.i. was similar to that in naive TCRδ-KO mice. The proportion of pDCs producing IL-12 in TCRδ-KO mice was similar to that in WT mice on day 0 and decreased on day 5 p.i. (Fig. 3F). Based on these data, we estimated the number of splenic DCs producing IL-12 in WT and TCRδ-KO mice. Although the absolute numbers of cDCs producing IL-12 increased in both WT and TCRδ-KO mice on day 5 p.i., the numbers were significantly higher in WT mice than in TCRδ-KO mice (Fig. 3G). For pDCs, the number of cells producing IL-12 was slightly increased in WT mice on day 5 p.i. (Fig. 3G). These results suggest that γδ T cells regulate the ability of DCs to produce IL-12 after infection. Because the majority of IL-12 was produced by cDCs on day 5 p.i., the reduced IL-12 levels in TCRδ-KO mice after infection likely is caused by impaired cDC ability.

γδ T Cells Increase Production of IFN-γ and Express CD40L During the Early Stages of P. berghei XAT Infection.

We investigated how γδ T cells enhance DC activation. Consistent with previous reports using various Plasmodium parasites (6–9), there was an increase in γδ T cells in the spleen after infection with P. berghei XAT (Fig. 4A). There was a prominent expansion of γδ T cells during the late stages of infection (e.g., on days 14 and 24 p.i.). Because γδ T cells act as critical immune cells to clear P. berghei XAT before day 9 p.i. (Fig. 1C), the prominent expansion of γδ T cells during the late stages of infection probably is not important for their function. IFN-γ is a cytokine that induces DC activation. Therefore, we performed an intracellular cytokine assay to evaluate IFN-γ–producing γδ T cells from the spleen. The ability of γδ T cells to produce IFN-γ increased from day 5 p.i. (Fig. 4 B and C), coinciding with DC activation (Fig. 3). The number of γδ T cells was increased slightly on day 5 p.i. (Fig. 4A). Although a significantly increased number of γδ T cells was observed in the spleen on day 14 p.i., the ability of γδ T cells to produce IFN-γ was reduced on day 14 p.i. compared with days 5, 7, and 9 p.i. (Fig. 4C). Thus, the number of γδ T cells may not be the only factor involved in the activation state of γδ T cells. Next, we immunostained spleen sections to observe the localization of γδ T cells and DCs. γδ T cells and DCs were detected by anti-TCRγδ and anti-CD11c mAbs, respectively. In the spleens of naive WT mice, most γδ T cells were interspersed in the red pulp region. Most CD11c⁺ DCs were localized in the red pulp region near the marginal zone at the interface between red and white pulp. Approximately 30% of γδ T cells adhered to DCs in splenic sections of naive WT mice (Fig. 4 D and F). On the other hand, on day 5 p.i. DCs in splenic sections of WT mice spread into the red pulp region (Fig. 4E). An accumulation of γδ T cells was observed around the marginal zone and red pulp region. The proportion of γδ T cells that adhered to DCs was about twofold higher (about 70%) in splenic sections from WT mice on day 5 p.i. than in sections from naive WT mice (Fig. 4F). These data suggest that many γδ T cells adhere to DCs in the spleens of both uninfected and infected mice and that the proportion of adhering cells is increased by the infection.

Fig. 4.

γδ T cells are activated during the early stages of P. berghei XAT infection. (A) Absolute numbers of γδ T cells in spleens from WT mice on days 0, 3, 5, 7, 9, 11, 14, 17, and 24 p.i. (n = 3–5). *P < 0.05, **P < 0.01. (B) (Left) CD3⁺-gated splenocytes in WT mice on days 0, 5, and 14 p.i. Numbers in each panel show the proportion of γδ T cells in CD3⁺ lymphocytes. (Right) CD3⁺- and TCRγδ⁺-gated spleen cells in WT mice on days 0, 5, and 14 p.i. Numbers in each panel show the proportion of γδ T cells expressing IFN-γ. (C) Proportions of γδ T cells expressing IFN-γ in spleens from WT mice on days 0, 3, 5, 7, 9, 11, and 14 p.i. (n = 3). **P < 0.01. (D and E) Localization of splenic γδ T cells and DCs detected by histological analyses with immunofluorescence. Confocal microscopic images show splenic sections in naive WT mice (D) and in WT mice on day 5 p.i. (E). Splenic sections were stained for CD11c (green), TCRγδ (red), and nuclei (blue). (Left) Low-power fields of splenic sections. (Right) High-power fields of splenic sections shown in boxes on left. Dotted lines indicate the marginal zone [the borderline between white pulp (WP) and red pulp (RP)]. Arrowheads indicate γδ T cells adhering to DCs. (Scale bars, 50 μm.) (F) Proportions of γδ T cells adhering to DCs in splenic sections of naive WT mice and WT mice on day 5 p.i. (n = 3). **P < 0.01. (G) Expression profiles of the CD40L molecule in CD4⁺ T cells and γδ T cells in spleens (Sp) and peripheral blood (PB) from WT mice. The blue line shows data for naive mice. The red line shows data for infected mice on day 5 p.i. The gray area represents data for isotype control. Data are representative of three to five independent experiments.

During our search for other molecules that induce γδ T cells and DC activation, we found that splenic γδ T cells (but not CD4⁺ T cells) in WT mice highly expressed CD40L on day 5 p.i. (Fig. 4G). On the contrary, CD40L-expressing γδ T cells were not observed in peripheral blood from WT mice on day 5 p.i. (Fig. 4G). This data is consistent with evidence that the spleen plays a central role in immunity against Plasmodium parasites (23). These results suggest that γδ T cells enhance DC activation by producing IFN-γ and increasing the expression of CD40L.

We hypothesized that stimulation of CD40 signaling in DCs by γδ T cells is critical for clearance of P. berghei XAT parasites. To test this hypothesis, we examined whether in vivo stimulation of CD40 by agonistic antibodies prevented TCRδ-KO mice from failing to clear the parasites. A previous report suggested that activation of the CD40 signal and stimulation by antigens synergistically induce IL-12 production by DCs (17). Furthermore, phagocytic activity of DCs for antigen presentation is reduced by DC activation signals (24). Because parasitemia increased exponentially until day 5 p.i., we supposed that injection of anti-CD40 mAb on days 1–3 p.i. is too early to stimulate DCs. Therefore, TCRδ-KO mice were infected with P. berghei XAT and then were injected i.v. with anti-CD40 agonistic mAb for stimulation of CD40 on day 4 p.i., 1 d before DC activation. As a control group, TCRδ-KO mice were injected with nonspecific control rat IgG on day 4 p.i. Approximately 70% of TCRδ-KO mice treated with anti-CD40 agonistic mAb controlled parasitemia levels and survived (Fig. 5 A and B). In contrast, all control TCRδ-KO mice treated with rat IgG had high parasitemia and eventually died (Fig. 5 A and B). Only a single injection of anti-CD40 agonistic mAb induced protection against P. berghei XAT parasites, even without γδ T cells. These results support our hypothesis that γδ T cells enhance DC activation by expressing CD40L to increase protective immunity against P. berghei XAT.

Fig. 5.

Anti-CD40 agonistic mAb can control the lethality of TCRδ-KO mice after infection with P. berghei XAT. TCRδ-KO mice were infected with P. berghei XAT through i.v. inoculation of blood-stage parasites (10⁴ infected RBCs). Anti-CD40 agonistic mAb (100 μg per mouse) was injected i.v. once on day 4 p.i. (A) Time-course analysis of parasitemia (n = 5). (B) Kaplan–Meier survival curves (n = 10). Data are representative of two or three independent experiments.

Discussion

γδ T cells react directly with pathogens as innate lymphocytes and also affect acquired immunity. Thus, γδ T cells are thought to play a central role as a bridge between innate and acquired immunity (1, 25, 26). Although the results of previous in vitro (27) and in vivo (9, 19) experiments suggest that γδ T cells play a role in protective immunity during malaria infection, the actual functions of γδ T cells remain largely unknown. Some in vitro experiments indicated that γδ T cells secrete granulysin and inhibit erythrocytic growth of P. falciparum (28). Therefore, granulysin may be an important effector of γδ T-cell–mediated protection against Plasmodium parasites. However, γδ T cells are a minor population in lymphoid organs. Therefore, it seems unlikely that γδ T cells can clear parasites using only the effect of granulysin. Here we provide evidence that γδ T cells boost DC activation for protective immunity against Plasmodium infection through CD40L of γδ T cells. γδ T cells are thought to play a key role in protective immunity in many other infectious diseases, and it is possible that enhancement of DC activation through CD40L of γδ T cells may occur in other infectious diseases also (29, 30).

Marked expansion of γδ T cells was observed on days 14, 17, and 24 p.i., during the late stages of P. berghei XAT infection, as is consistent with previous reports (9). Furthermore, previous studies showed that γδ T cells failed to proliferate in the absence of CD4⁺ T cells (20, 30). These reports suggested that CD4⁺ T cells are necessary for proliferation of γδ T cells in response to stimulation with malarial antigens and that marked expansion of γδ T cells might be required for protective immunity against Plasmodium parasites. In the present study, because a difference in parasitemia between WT and TCRδ-KO mice was observed from day 10 p.i. (Fig. 1A), we hypothesized that γδ T cells would begin to play their roles in protective immunity from day 10 p.i. We supposed that our hypothesis would be consistent with the suggestions of previous results (20, 31). However, an unexpected finding was that depletion of γδ T cells from day 9 p.i. had no effect on the clearance of P. berghei XAT. In contrast, even from day 9 p.i., CD4⁺ T cells were needed to clear the parasites. These results suggest that CD4⁺ T cells indeed were the effector cells of protective immunity against P. berghei XAT and that marked expansion of γδ T cells is not required for protective immunity against these parasites.

Artificial stimulation of CD40 signaling by administration of agonistic anti-CD40 mAb to TCRδ-KO mice enhanced protective immunity against P. berghei XAT parasites. These results strongly suggest that γδ T cells play a crucial role in the stimulation of CD40 signaling in DC through CD40L expression. Although administration of an agonistic anti-CD40 mAb led to complete clearance of parasites, the clearance took significantly longer in control WT mice (Fig. 5A versus Fig. 1A). We speculate that the artificial stimulation of CD40 signaling might not be adequate to substitute for γδ T cells in TCRδ-KO mice or to clear the Plasmodium parasites effectively. The result that about 30% of TCRδ-KO mice given anti-CD40 mAb succumbed to the infection may support our speculation. There are some possible explanations for the delay in clearance of the Plasmodium parasites. First, the single injection of anti-CD40 mAb on day 4 p.i. may have been too weak and/or too early to stimulate DCs. Second, signals by the other costimulation factors on γδ T cells might be needed to induce enough activation of DCs. Third, some other γδ T-cell–secreted factors may be involved in γδ T-cell–mediated immunity. IL-17 is one of these factors. γδ T cells are well known to have the potential to produce IL-17, and some previous reports showed that IL-17–producing γδ T cells play crucial roles in innate immunity against infections, such as those caused by Listeria monocytogenes and Mycobacterium tuberculosis (32, 33). However, whether IL-17 is produced by γδ T cells after infection with Plasmodium parasites and its significance in protective immunity against infection with these pathogens remain to be elucidated. Because mice with γδ T-cell depletion caused by anti-TCRγδ mAb administration from day 9 p.i. eliminated the parasites, the enhanced expansion of γδ T cells is not involved directly in the elimination of Plasmodium parasites. The expansion of γδ T cells during the late stages of malaria infection has some physiological implications. One possibility is that γδ T cells repair injured tissues, such as liver and spleen, during infection. A previous report showed that skin γδ T cells play an important role in wound repair by producing keratinocyte growth factors and chemokines (34).

Infection of TCRδ-KO mice with Plasmodium chabaudi leads to exacerbation of parasitemia, although infected TCRδ-KO mice can clear the parasites eventually (35). Thus, dependence on γδ T-cell–mediated protective immunity differs among Plasmodium species. However, in human cases, the number of γδ T cells is increased in the blood and spleen of patients with P. falciparum (7, 8). Therefore, γδ T cells play a critical role in protective immunity in humans. Future studies should examine whether human γδ T cells express CD40 after stimulation with P. falciparum-infected RBCs. Our results provide evidence for γδ T-cell– and DC-mediated regulation of malaria via CD40 signaling and offer insights into immunotherapeutic approaches for malaria, such as the use of specific chemicals for activating DCs via CD40 signaling.

Materials and Methods

Mice.

C57BL/6J mice (CLEA Japan) and TCRδ-KO mice (Jackson Laboratories) were bred in the specific pathogen-free unit in the animal facility of Kyorin University. All animals were female and were 8–12 wk old at the time of infection. Animal protocols were approved by the animal care committee of Kyorin University School of Medicine.

Parasites and Infection.

An attenuated derivative of P. berghei (10), P. berghei XAT, was used throughout the study. The parasites were stored as frozen stocks in liquid nitrogen. Freshly thawed parasites were passaged once through naive C57BL/6J mice, and 10⁴ iRBCs from passaged mice were injected i.v. into experimental mice. The resulting parasitemia was assessed by counting 250–10,000 RBCs in a Giemsa-stained thin blood film. The percentage of parasitemia was calculated as follows: [(number of infected RBCs)/(total number of RBCs)] × 100.

In Vivo Cell Depletion.

Hybridoma cells producing anti-TCRγδ (GL3) mAb were kindly provided by Drew Pardoll (The Johns Hopkins University, Baltimore). Cells producing anti-CD4 (GK1.5) mAb were obtained from American Type Culture Collection. The mAbs were partially purified from ascites by two precipitations in 50% (wt/vol) saturated ammonium sulfate solution at pH 7.4. For immunodepletion of γδ T cells, one group of mice was depleted of γδ T cells by i.p. injection of 0.5 mg of anti-TCRγδ mAb on days 0, 1, and 2 p.i. and twice a week thereafter. Another group of mice was administered anti-TCRγδ mAb twice a week beginning at day 9 p.i. For immunodepletion of CD4⁺ T cells, a group of mice was administered anti-TCRγδ mAb twice a week beginning at day 9 p.i. Normal hamster and rat IgG (Cappel Research Products) were used as control antibodies. Depletion of γδ T cells and CD4⁺ T cells was confirmed by flow cytometry.

Administration of Anti-CD40 Agonistic mAb to Mice Infected with Plasmodium Parasites.

Anti-CD40 mAb (100 μg per mouse) (BioLegend) was injected i.v. into P. berghei XAT-infected mice once on day 5 p.i.

Statistical Methods.

Statistical analyses were performed using the Student’s t test with Statcel (OMS Ltd.).

Materials and methods used for flow cytometry, intracellular cytokine assay, ELISA, and immunohistochemical analysis are included in SI Materials and Methods.