γδ T Cells but Not NK Cells Are Essential for Cell-Mediated Immunity against Plasmodium chabaudi Malaria

Abstract

Blood-stage Plasmodium chabaudi infections are suppressed by antibody-mediated immunity and/or cell-mediated immunity (CMI). To determine the contributions of NK cells and γδ T cells to protective immunity, C57BL/6 (wild-type [WT]) mice and B-cell-deficient (J_H^−/−) mice were infected with P. chabaudi and depleted of NK cells or γδ T cells with monoclonal antibody. The time courses of parasitemia in NK-cell-depleted WT mice and J_H^−/− mice were similar to those of control mice, indicating that deficiencies in NK cells, NKT cells, or CD8⁺ T cells had little effect on parasitemia. In contrast, high levels of noncuring parasitemia occurred in J_H^−/− mice depleted of γδ T cells. Depletion of γδ T cells during chronic parasitemia in B-cell-deficient J_H^−/− mice resulted in an immediate and marked exacerbation of parasitemia, suggesting that γδ T cells have a direct killing effect in vivo on blood-stage parasites. Cytokine analyses revealed that levels of interleukin-10, gamma interferon (IFN-γ), and macrophage chemoattractant protein 1 (MCP-1) in the sera of γδ T-cell-depleted mice were significantly (P < 0.05) decreased compared to hamster immunoglobulin-injected controls, but these cytokine levels were similar in NK-cell-depleted mice and their controls. The time courses of parasitemia in CCR2^−/− and J_H^−/− × CCR2^−/− mice and in their controls were nearly identical, indicating that MCP-1 is not required for the control of parasitemia. Collectively, these data indicate that the suppression of acute P. chabaudi infection by CMI is γδ T cell dependent, is independent of NK cells, and may be attributed to the deficient IFN-γ response seen early in γδ T-cell-depleted mice.

Malaria remains a leading cause of morbidity and mortality, annually killing about 2 million people worldwide (32, 33). Despite decades of research, malaria is a reemerging disease because of increasing drug resistance by malarial parasites and insecticide resistance by the mosquito vector. Most infected individuals do not succumb to malaria but develop clinical immunity where parasite replication is controlled to some degree by the immune system without eliciting clinical disease or sterile immunity (14, 38).

Understanding the immunologic pathways leading to the control of blood-stage parasite replication is important for defining the mechanisms of disease pathogenesis and improving vaccines currently in development. The early events of the immune response depend upon activation of the innate immune system, which regulates the downstream adaptive immune response needed to control or cure (44). Natural killer (NK) and γδ T cells function early in the immune response to pathogens as components of the innate immune system. Both cell types have been proposed to play significant roles in the subsequent clearance of blood-stage malarial parasites by activating the adaptive immune system (35, 43, 44). The mechanism by which they accomplish this appears to be mediated via their secretion of gamma interferon (IFN-γ) induced by cytokines such as interleukin-12 (IL-12), tumor necrosis factor alpha (TNF-α), and IL-6 produced by other components of the innate immune system, including macrophages and dendritic cells (17, 25, 26, 37, 49).

Blood-stage malaria parasites are cleared by mature isotypes of antibodies and/or by antibody-independent but T-cell-dependent mechanisms of immunity (2, 15, 22). Both responses require CD4⁺ αβ T cells; in addition, the expression of cell-mediated immunity (CMI) during both acute and chronic malaria is dependent on γδ T cells activated by CD4⁺ αβ T cells (29, 47, 49, 50). Wild-type (WT) mice depleted of γδ T cells by antibody treatment or gene knockout suppress P. chabaudi parasitemia by antibody-mediated immunity (AMI) (21, 52). Mice depleted of B cells by the same procedures also cure their acute infections in the same timeframe as intact control mice but then develop chronic low-grade parasitemia of long-lasting duration, indicating that B cells and their antibodies are needed to sterilize the infection as we originally reported (15, 48) and has since been confirmed by others (51). B-cell-deficient mice depleted of γδ T cells cannot suppress P. chabaudi parasitemia (49, 50, 52).

The prominent role played by IFN-γ in immunity to malaria is generally accepted by most researchers. P. chabaudi malaria is more severe in WT mice treated with neutralizing antibody and in IFN-γ^−/− mice, as indicated by the increased magnitude and duration of parasitemia and mortality in mice deficient in IFN-γ versus intact controls (24, 39, 46). In B-cell-deficient animals, the similar neutralization of IFN-γ by treatment with anti-IFN-γ monoclonal antibody (MAb) or gene knockout of IFN-γ has an even greater effect on the time course of parasitemia, which remains at high levels and fails to cure (1, 46), indicating that IFN-γ is essential for the expression of anti-parasite CMI and contributes to AMI in this model system.

The early source of IFN-γ remains controversial, with both NK cells and γδ T cells being proposed to produce this critical cytokine necessary for the activation of the adaptive immune response and the development of protective immunity (9). The results of earlier genetic studies failed to correlate susceptibility to P. chabaudi infection with NK activity (31, 44). Subsequently, Mohan et al. (25) reported that NK cell activity against tumor cell targets correlates with protection against P. chabaudi; anti-asialo GM1 polyclonal antibody depletion of NK cells results in significantly increased levels of peak parasitemia and a prolonged duration of infection compared to controls. The mode of action by which NK cells function appears to be via the secretion of cytokines (25) rather than direct cytotoxicity against the blood-stage parasites. The surface expression of lysosome-associated membrane protein 1 (LAMP-1) by subsets of human NK cells exposed to Plasmodium falciparum-infected erythrocytes may suggest otherwise (20). NK cells in collaboration with dendritic cells are responsible for optimal IFN-γ production dependent upon IL-12 (17, 36, 39, 40). In contrast to the findings of Mohan et al., other studies indicate similar P. chabaudi parasitemia in depleted mice and intact controls after NK1.1 MAb depletion of NK cells (19, 41, 53). Using microarray analysis of blood cells from P. chabaudi-infected mice, Kim et al. (18) reported a rapid production of IFN-γ and activation of IFN-γ-mediated signaling pathways as early as 8 h after infection; however, NK cells did not express IFN-γ or exhibit IFN-γ-mediated pathways in their analysis. At this time, NK cells are replicating and migrating from the spleen to the blood. In humans with P. falciparum malaria, increased production of IFN-γ by PBMC in response to parasitized RBCs correlates with protection from high-density parasitemia and clinical malaria (10, 11); early IFN-γ production by PBMC obtained from malaria naive donors is primarily by γδ T cells and not by NK cells (26). Animal models by definition do not exactly mimic the human condition, and the experimental malaria in mice uses distinct species from those that infect humans. Nevertheless, analysis of protective immunity provides important information on how a protective immune response to Plasmodium may be elicited.

Whether both NK cells and γδ T cells have essential roles during the early stages of the immune response to blood-stage malaria remains to be determined. Likewise, whether these cells function early in CMI to malaria parasites is unknown. To address these issues, we infected NK-cell- or γδ-T-cell-depleted J_H^−/− mice with blood-stage P. chabaudi. The resulting time course of parasitemia was monitored and compared to control mice. In addition, spleen cells from depleted and control mice were profiled by cytofluorimetry, and the serum levels of inflammatory cytokines were measured.

MATERIALS AND METHODS

Parasites and infections of mice.

CCR2^−/− mice, β₂m^−/− mice, and CD1d^−/− mice of both sexes on a C57BL/6 (WT) background were purchased from Jackson Laboratories (Bar Harbor, ME). B-cell-deficient J_H^−/− mice, which were generated by deletion of the J_H region in embryonic stem cells, fail to produce immunoglobulin (Ig) and are devoid of B cells because B-cell differentiation is blocked at the large CD43⁺ precursor stage (8). J_H^−/− mice originally obtained from Dennis Huszar were bred and maintained in our laboratory. J_H^−/− mice, tenth generation backcrossed to C57BL/6 (WT) mice, were crossed with CCR2^−/− mice, which lack the receptor for the chemokine, macrophage chemoattractant protein 1 (MCP-1). β₂m^−/− mice and CD1d^−/− mice were also bred with J_H^−/− mice to produce double-knockout (KO) mice, which were phenotyped by standard PCR techniques, using gene-specific primers. β₂m^−/− mice defective in major histocompatibility complex (MHC) class I expression are deficient in NK cells, NKT cells, and CD8⁺ T cells. CD1d^−/− mice are deficient in NKT cells. All mice were bred and maintained at the AAALAC-accredited University of Wisconsin-Madison vivarium and infected between 8 and 12 weeks of age. Animal studies have been reviewed and approved by the University of Wisconsin Animal Care Committee.

P. chabaudi adami 556KA, which produces nonlethal infections and sterilizing immunity in immunologically intact mice, was maintained and used as described previously (6). Unless stated otherwise, age- and sex-matched experimental and control mice were injected intravenously (i.v.) with 10⁵ P. chabaudi-parasitized erythrocytes obtained from a source mouse. The injections of both antibodies and parasites into the peritoneal cavity causes the activation of resident macrophages and results in a reduced inoculum size. This is prevented by injecting antibodies and parasites by different routes. Previous experience indicated that the kinetics of parasitemia following the injection of 10⁵ parasitized erythrocytes i.v. were equivalent to those following an injection with 10⁶ parasitized erythrocytes given intraperitoneally (i.p.).

The parasitized erythrocytes were washed prior to inoculation to prevent the carryover of cytokines and chemokines in donor blood. The resulting parasitemia was assessed by counting the number of parasitized erythrocytes in 500 to 1,000 erythrocytes in Giemsa-stained thin blood films.

Antibody depletion.

Purified anti-NK1.1 MAb (hybridoma PKI36) was a gift from Charles Czyprinski (University of Wisconsin, Madison) and anti-TCRδ (GL3) was kindly provided by Francesca Neethling and Jon Weidanz (Receptor Logic, Abilene, TX). Purified mouse Ig and hamster Ig were purchased from Thermo Scientific (Rockford, IL) and used as a control for the depleting MAbs. Depletion of NK cells was accomplished by using the following treatment regimen: 250 μg of anti-NK1.1 in phosphate-buffered saline was injected i.p. into each mouse on days −1, 0, 1, 3, 5, 10, and 15 postinoculation (p.i.) with parasitized erythrocytes. Depletion of γδ T cells was achieved by the i.p. injection of 500 μg of anti-GL3 on days −1, 0, 1, and 15 p.i. Normal mouse Ig and hamster Ig were injected i.p. into control mice according to the same regimen used for the depleting MAbs.

Flow cytometry.

Four-color flow cytometry was performed as described previously (16). Briefly, the spleen was dissected from a euthanized animal, disaggregated by passing the spleen through nylon mesh, and the erythrocytes were lysed by hypotonic shock. Washed cells were incubated with Fc block (Becton Dickinson, San Diego, CA) and then stained with the following MAbs labeled with fluorescein isothiocyanate or phycoerythrin purchased from Pharmingen: anti-CD3e (145-2C11), anti-CD4 (GK1.5), anti-CD8 (Ly-2), anti-TCRαβ (H57), anti-TCRγ (GL3), anti-NK1.1 (PK136), anti-F4/80 (M5-2C11), and anti-CD11b (M1/70). Propidium iodide was added prior to flow cytometric analysis to exclude dead cells. An equal number of cells (10, 000) from MAb-treated test mice or Ig-treated control mice was analyzed. The number of cells examined was sufficient to accurately assess the percentages of different subsets of spleen cells. This percentage was multiplied by the total number of live cells in the spleen (measured as trypan blue-negative cells by using a hemocytometer) to obtain the number of cells within selected subsets of splenocytes. More than 10,000 events calculates the percentage to 0.01% accuracy, which is beyond the machine error and interanimal variation.

Cytokine measurement by bead array.

The levels of selected cytokines (IL-6, IL-10, IL-12p70, IFN-γ, TNF-α, and MCP-1) in the sera of experimental animals and their controls obtained days 4 and 8 p.i. were assayed by using a mouse inflammation kit from BD Biosciences (La Jolla, CA) according to the manufacturer's instructions. The limit of detection of this assay is 20 pg of cytokine/ml and is indicated by a dashed line in Fig. 1C, 2C, 4C, and 6C.

FIG. 1.

P. chabaudi infection in C57BL/6 mice depleted of NK cells and mouse Ig-treated controls (n = 4/group). (A) Time course of parasitemia after the inoculation of 10⁵ parasitized erythrocytes i.v. in C57BL/6 mice treated with anti-NK 1.1 or mouse Ig (250 μg) on days −1, 0, 1, 3, 5, 10, and 15 p.i. injected i.p. (B) Cytofluorimetric analysis of spleen cells from NK cell-depleted mice and controls 19 days p.i. (C) Serum cytokine levels in NK cell-depleted mice and controls on days 4 and 8 p.i. *, P < 0.05.

FIG. 2.

P. chabaudi infection in C57BL/6 mice depleted of γδ T cells and hamster-Ig treated controls (n = 4 animals/group). (A) Time course of parasitemia after the inoculation of 10⁵ parasitized erythrocytes i.v. in C57BL/6 mice treated with anti-GL3 or hamster Ig (500 μg) on days −1, 0, 1, and 15 p.i. injected i.p. (B) Cytofluorimetric analysis of spleen cells from γδ T-cell-depleted mice and controls 21 days p.i.. (C) Serum cytokine levels in γδ T-cell-depleted mice and controls on days 4 and 8 p.i. *, P < 0.05.

IgG isotype ELISA.

The isotypic profile of the antigen-specific antibodies in the sera of infected mice depleted of NK or γδ T cells and their controls was determined by enzyme-linked immunosorbent assay (ELISA) using P. chabaudi apical membrane antigen-1 (rPCAMA-1)- or P. chabaudi merozoite surface protein-1₄₂ (rPCMSP-1)-coated wells as described previously (5). Briefly, serum from each animal harvested day 19 p.i. (NK cell-depleted and mouse Ig-treated controls) and day 21 p.i. (γδ T-cell-depleted and hamster Ig-treated controls) was assayed in duplicate on antigen-coated wells at dilutions that ranged from 1:100 to 1:1,000. Antigen-specific antibodies were detected with horseradish peroxidase-conjugated rabbit antibody specific for mouse IgM, IgG1, IgG2b, or IgG3 (Zymed Laboratories) or with horseradish peroxidase-conjugated goat anti-mouse IgG2c (IgG2a b allotype; Southern Biotechnology Associates, Inc., Birmingham, AL) (23) and ABTS [2,2′azinobis(3-ethylbenzthiazolinesulfonic acid)] as the substrate. In each assay, wells were coated with purified IgM (eBioscience, San Diego, CA) or with IgG1, IgG2b, or IgG3 (Zymed Laboratories). Isotype control antibodies were used to generate standard curves (16 ng/ml to 2 μg/ml). The IgG2c standard curve was generated by using a purified monoclonal IgG2c (IgG2a b allotype) antibody (BD Biosciences Pharmingen, San Jose, CA). The concentration of each Ig isotype was expressed in units per milliliter, where 1 U/ml was equivalent to 1 μg of myeloma standard/ml.

Statistical analysis.

Analysis of variance with the StatView program (SAS Institute, Cary, NC) was performed to statistically compare parasitemia, cell numbers, and percentages in different groups of mice obtained from depletion studies. The statistical significance of all other parasitemia data was determined by using a Student t test. A P value of <0.05 was considered significant.

RESULTS

Effect of NK cell versus γδ T-cell depletion on the time course of P. chabaudi parasitemia in WT mice.

To determine the effects of treatment with anti-NK1.1 versus treatment with anti-GL3, four parameters of P. chabaudi malaria in WT mice were compared; these parameters included the time course of parasitemia, the profile of splenic mononuclear cells, the serum levels of selected cytokines, and the antibody isotype responses to recombinant P. chabaudi antigenic constructs.

Two groups of WT mice (n = 4/group) were individually inoculated i.v. with. 10⁵ P. chabaudi-parasitized erythrocytes, followed by 250 μg i.p. of either anti-NK1.1 MAb or mouse Ig on days −1, 0, 1, 3, 5, 10, and 15 p.i. The time course of acute parasitemia in both test and control groups of mice was similar (Fig. 1A), with peak parasitemia occurring on day 9 and clearance of parasites from the blood to < 0.01% occurring on days 15 and 19, respectively. This regimen of anti-NK1.1 treatment depleted NK cells and NKT cells from the spleen; indeed, few NK1.1 cells and NKT cells were detected by flow cytometric analysis of splenocytes obtained from antibody-treated mice on day 19 p.i. (Fig. 1B). The number of cells in each splenic cell population other than NK cells and NKT cells did not differ significantly (P > 0.05) between the anti-NK1.1 MAb-treated mice and the mouse Ig control groups. Likewise, the numbers of γδ T cells and activated (B220⁺) γδ T cells, as well as the number of splenic macrophages (F4/80⁺, CD11d⁺), were not diminished by the antibody depletion of NK cells. The difference in numbers observed between test and control groups was not significant. To determine whether NK cell depletion affected the levels of selected cytokines (IL-6, IL-10, IL-12, TNF-α, and IFN-γ) and the chemokine, MCP-1, we quantified these proteins in the sera of test and control mice bled on days 4 and 8 p.i. Day 4 corresponds to the period of ascending parasitemia when the protective immune response is being activated by the innate immune system to control parasite replication. By day 8 p.i., the activated immune system appears to be functioning as indicated by the decrease in the slope of parasitemia (Fig. 1A). The levels of all cytokines and chemokines were below the limits of resolution (<20 pg/ml) in uninfected animals. The concentration of IFN-γ increased significantly (P < 0.05) in both groups of mice on days 4 and 8 p.i., with the greatest increase seen on day 4 p.i. (Fig. 1C). The levels of serum MCP-1 also increased markedly on day 4 p.i. and then declined by day 8 p.i. IL-10 was elevated on day 8 p.i. but not day 4. The remaining cytokines (IL-12, TNF-α, and IL-6) were not detected above baseline. There were no significant differences (P > 0.05) in IL-10, IFN-γ, or MCP-1 concentrations between NK-depleted and control groups. Similar results were obtained upon repeating the experiment.

Depletion of γδ T cells from WT mice decreases the levels IFN-γ and MCP-1 early during the course of P. chabaudi malaria.

Previously, we and others have reported that the time course of P. chabaudi parasitemia was prolonged for several days in WT mice depleted of γδ T cells by MAb treatment or gene knockout compared to infected controls (21, 52). To determine the role of γδ T cells in the control of acute P. chabaudi parasitemia, we inoculated two groups of WT mice (n = 4 animals/group) i.v., with each mouse receiving 10⁵ P. chabaudi parasitized erythrocytes, followed by the i.p. injection of 500 μg of either anti-TCR γδ MAb or hamster Ig on days −1, 0, 1, and 15 p.i. The time course of parasitemia was similar in both groups of mice, with peak parasitemia occurring on day 9 p.i. (Fig. 2A). However, the mice injected with anti-GL3 showed significantly (P < 0.05) increased parasitemia on days 13 and 15 p.i. compared to control mice but exhibited modest differences in suppressing the duration of patent parasitemia (day 15 p.i. for hamster Ig-treated controls and day 19 p.i. for γδ T-cell-depleted mice), a result reminiscent of the published findings from previous studies.

Flow cytometric analysis of spleen cell populations from anti-GL3-treated mice and controls revealed a statistically significant (P < 0.05) depletion of γδ T cells in the antibody-treated mice compared to hamster Ig-treated controls. The number of cells in each αβ T-cell subset (CD4⁺ and CD8⁺) was greater in the γδ T-cell-depleted animals, but the differences did not reach significance (Fig. 2B). Likewise, the number of splenic macrophages increased in γδ T-cell-depleted animals compared to controls (Fig. 2B), whereas they were approximately the same in NK-depleted animals compared to their corresponding controls (Fig. 1B).

Depletion of γδ T cells affected cytokine production. The concentrations of serum IFN-γ and MCP-1 were significantly (P < 0.05) lower on day 4 p.i. in γδ T-cell-depleted animals compared to controls; on day 8 p.i., the level of IL-10 was lower in antibody-treated animals than in controls (Fig. 2C). Although IL-12 was decreased in the sera of the γδ T-cell-depleted animals compared to controls, the levels were too low to make any conclusions with certainty. The other cytokines (TNF-α and IL-6) were not detected above baseline. We repeated the experiment with nearly identical results.

Depletion of NK cells or γδ T cells from WT mice increases selected antibody isotype responses to malaria antigens.

Antibody isotypes specific for rPCAMA-1 and rPCMSP-1 were measured in serum samples from the same mice harvested on day 19 or day 22 p.i. The results indicate that both NK cell depletion and γδ T-cell depletion caused marked increases (P < 0.05) in the levels of IgG2c and IgG3 antibodies to both antigens compared to infected Ig-treated controls (Fig. 3).

FIG. 3.

Antibody isotype responses to rPCAMA-1 (top panel) and rPCMSP-1 (bottom panel) determined by ELISA in P. chabaudi-infected C57BL/6 mice depleted of NK cells or γδ T cells and their controls using serum harvested at days 19 and 21 p.i., respectively. *, P < 0.05.

NK cells are not essential for expression of CMI against P. chabaudi malaria.

To determine whether NK cells play an essential role in the activation of CMI that suppresses acute parasitemia in B-cell-deficient mice, we inoculated two groups of J_H^−/− mice (n = 4 animals/group), with each mouse receiving 10⁵ P. chabaudi-parasitized erythrocytes i.v., followed by either anti-NK1.1 MAb or mouse Ig i.p., using the regimen described above. The time course of acute parasitemia was similar in both groups of mice (Fig. 4 A), with peak parasitemia occurring on day 9 p.i. and the suppression of parasitemia to levels of <0.01% occurring in NK-cell-depleted mice by day 15 p.i. and by day 17 p.i. in control mice. There were few splenic NK1.1 cells in anti-NK1.1-treated mice compared to control mice, as detected by flow cytometry on day 21 p.i. (Fig. 4B).

FIG. 4.

P. chabaudi infection in B-cell-deficient mice depleted of NK cells and mouse Ig-treated controls (n = 4 animals/group). (A) Time course of parasitemia after the inoculation of 10⁵ parasitized erythrocytes i.v. in B-cell-deficient mice treated with anti-NK 1.1 or mouse Ig (250 μg) on days −1, 0, 1, 3, 5, 10, and 15 p.i. injected i.p. (B) Cytofluorimetric analysis of spleen cells from NK cell-depleted mice and controls at 19 day p.i. (C) Serum cytokine levels in NK cell-depleted mice and controls on days 4 and 8 p.i. *, P < 0.05.

To determine the effects of NK cell depletion on other splenic cell populations, we assessed the number of CD4⁺, CD8⁺, αβ, and γδ T cells on day 21 p.i. by flow cytometry. The number of cells in each population was similar in the anti-NK1.1-treated and mouse Ig control groups (Fig. 4B). Although NK cell depletion did not alter γδ T-cell activation, a decrease in the number of splenic macrophages (F4/80⁺, CD11d⁺ cells) as a result of NK cell depletion was observed (Fig. 4B). The differences in macrophage numbers in test and control mice did not reach significance.

We also determined whether NK cell depletion in J_H^−/− mice affected the levels of selected cytokines (IL-6, IL-10, IL-12, TNF-α, and IFN-γ) and the chemokine MCP-1 on days 4 and 8 p.i. Similar to the results observed in WT mice, the concentrations of IFN-γ increased significantly on both day 4 p.i. and day 8 p.i.; the greatest concentration was on day 4 p.i. (Fig. 4C). The levels of serum MCP-1 also increased markedly on day 4 p.i. and then declined by day 8 p.i. IL-10 was elevated only on day 8 p.i.. The other cytokines (IL-12, TNF-α, and IL-6) were not detected above baseline levels. There were no significant differences in IL-10, IFN-γ, or MCP-1 concentration between NK-depleted and control groups.

CMI against P. chabaudi is fully functional in knockout mice deficient in NK cells or NKT cells.

To determine the contribution of NK, NKT, and CD8⁺ T cells to control of P. chabaudi parasitemia, we injected 10⁶ parasitized erythrocytes into groups of B-cell-deficient J_H^−/− × β₂m^−/−, J_H^−/− × CD1d^−/− double KO mice and their B-cell-deficient, heterozygous controls (J_H^−/− × β₂m^+/−, J_H^−/− × CD1d^+/−). β₂m^−/− lack MHC class I expression and exhibit severely diminished numbers of NK cells, NKT cells, and CD8⁺ T cells (30, 45). CD1d^−/− mice lack NKT cells (3). The parasitemia was similar in the B-cell-deficient mice lacking β₂m or CD1d and their controls; both groups suppressed their acute parasitemia with a similar time course (Fig. 5).

FIG. 5.

Time course of P. chabaudi parasitemia in double KO mice lacking NK cells, NKT cells, or CD8⁺ T cells. (A) Time course of parasitemia in J_H^−/− × β₂m^−/− double KO mice (n = 7) and their J_H^−/− × β₂m^+/− controls (n = 4) inoculated i.p. with 10⁶ P. chabaudi-parasitized erythrocytes. (B) Time course of parasitemia in J_H^−/− × CD1^−/− double KO mice (n = 5) and their J_H^−/− × CD1d^+/− controls (n = 5) inoculated i.p. with 10⁶ P. chabaudi-parasitized erythrocytes.

Suppression of CMI against P. chabaudi is mediated by γδ T cells.

To confirm that γδ T-cell function in the activation of CMI that suppresses acute parasitemia as we reported previously (49, 50, 52), we inoculated B-cell-deficient mice i.v. with 10⁵ P. chabaudi-parasitized erythrocytes and then injected the mice i.p. with either anti-GL3 or hamster Ig as described above. The parasitemia was similar throughout the period of ascending parasitemia (Fig. 6 A), with both test and control groups exhibiting similar magnitudes of peak parasitemia, Subsequently, the time course of parasitemia diverged significantly with γδ T-cell-depleted animals exhibiting significantly greater parasitemia (P < 0.05) that did not resolve. There were significantly fewer γδ T cells detected by flow cytometry in the spleens of anti-GL3-treated mice compared to controls (Fig. 6B).

FIG. 6.

P. chabaudi infection in B-cell deficient mice depleted of γδ T cells and hamster Ig-treated controls (n = 4 animals/group). (A) Time course of parasitemia after the inoculation of 10⁵ parasitized erythrocytes i.v. in B-cell-deficient mice treated with anti-GL3 or hamster Ig (500 μg) on days −1, 0, 1, and 15 p.i. injected i.p. (B) Cytofluorimetric analysis of spleen cells from γδ T-cell-depleted mice and controls 21 days p.i. (C) Serum cytokine levels in γδ T-cell-depleted mice and controls on days 4 and 8 p.i. *, P < 0.05.

To determine whether the depletion of γδ T cells affected the profile of other cell populations that might contribute to the suppression of CMI to P. chabaudi infection, we assessed the numbers of splenic CD4⁺ T cells, CD8⁺ T cells, NK cells, NKT cells, and dendritic cells on day 21 p.i. by flow cytometry. The number of cells in each αβ T-cell subset (CD4⁺ and CD8⁺) was greater in the γδ T-cell-depleted compared to controls (Fig. 6B). We also observed an increase in the numbers of splenic macrophages in γδ T-cell-depleted animals. In contrast, the numbers of NK cells and NKT cells were approximately the same in the absence of γδ T cells compared to controls. Except for the depletion of γδ T cells in test versus control mice, these changes in spleen cell numbers of the different cell populations assayed in test and control mice were modest and did not reach statistical significance.

To determine whether γδ T-cell depletion affected the levels of selected cytokines (IL-6, IL-10, IL-12, TNF-α, and IFN-γ) and the chemokine MCP-1, we measured the levels of these proteins in serum on days 4 and 8 p.i. in groups of mice injected i.p. with either anti-Τ-cell γδ MAb or hamster Ig. The concentration of IFN-γ was significantly (P < 0.05) lower on day 4 p.i. in γδ T-cell-depleted animals than in controls (Fig. 6C); on days 4 and 8 p.i., the levels of MCP-1 and IL-10 were also depressed in γδ T-cell-depleted animals compared to controls. The concentration of IL-10 was too low on day 4 p.i. to draw a firm conclusion. However, the other cytokines assayed (IL-12, TNF-α, and IL-6) were not detected above baseline.

The MCP-1 receptor, CCR2, does not play an essential protective role in immunity to P. chabaudi malaria.

Although the role of IFN-γ in suppression of P. chabaudi parasitemia is established, the observation that the levels of MCP-1 in the sera of γδ T-cell-depleted mice were also depressed suggests that this chemokine, which has been reported to play an important role in monocyte migration and polarization of the type 1 cytokine response (7), may have an important role in curing acute parasitemia. To test this supposition, we injected both CCR2^−/− mice and heterozygous controls i.p. with 10⁶ P. chabaudi-parasitized erythrocytes and monitored parasitemia over time; the time course of parasitemia was similar in both CCR2^−/− mice and controls (Fig. 7 A).

FIG. 7.

Comparison of time courses of P. chabaudi parasitemia in CCR2 KO mice (n = 6) and C57BL/6 controls (n = 6) (A) and in J_H^−/− × CCR2^−/− double KO mice (n = 4) and their J_H^−/− × CCR2^+/− controls (n = 4) (B) after the inoculation of 10⁶ P. chabaudi-parasitized erythrocytes i.p.

We also bred J_H^−/− × CCR2^−/− double KO mice to determine whether MCP-1 was essential for the expression of CMI against the parasite. Double KO mice and J_H^−/− × CCR2^+/− controls inoculated i.p. with 10⁶ parasitized erythrocytes had similar time courses of parasitemia (Fig. 7B).

Antibody depletion of γδ T cells in mice chronically infected with P. chabaudi causes an immediate and pronounced exacerbation of parasitemia.

γδ Τ cells are required to suppress parasitemia during acute and chronic P. chabaudi malaria in B-cell-deficient C57BL/6 mice (49, 50, 52). As indicated in Fig. 4A, depletion of γδ T cells prevented the suppression of acute parasitemia, as seen in control mice. In addition, the depletion of γδ T cells during chronic malaria causes the exacerbation of parasitemia. To gain insight on how γδ cells function, we injected chronically infected mice i.p. on day 60 p.i. with either anti-TCRγδ MAb or hamster Ig and monitored their parasitemia daily during the first week after injection (Fig. 8). The time course of acute parasitemia was similar in both groups of mice (data not shown), and both experimental and control mice exhibited similar parasitemia prior to the injection of the depleting antibody (Fig. 8). However, there was an immediate and marked increase in parasitemia which rose dramatically to high levels (i.e., ∼45%) in antibody-treated J_H^−/− mice by day 5 after treatment with anti-GL3, whereas parasitemia remained at preinjection levels in the control mice. A repeat of this experiment yielded nearly identical results.

FIG. 8.

Immediate exacerbation of P. chabaudi parasitemia in chronically infected B-cell-deficient mice, following the injection i.p. of 1 mg of anti-GL3 versus 1 mg of hamster Ig. The mice had been inoculated 60 days previously with 10⁶ P. chabaudi-parasitized erythrocytes i.p. *, P <0.05.

DISCUSSION

Both NK cells and γδ T cells have been purported to function early in the innate immune response to blood-stage malaria parasites (35, 43, 44). These assumptions in part are based on the outcomes of cell depletion studies utilizing antibodies and/or gene KO in mouse models of blood-stage malaria. Whether either cell type dominates in protection is unknown. Likewise, the function of these cells in AMI and CMI to malaria remains to be determined.

Previously, it was reported that the depletion of NK cells either enhanced the severity of the AS strain of P. chabaudi chabaudi infections in terms of mortality and/or parasitemia or had little, if any, effect on these parameters of infection compared to intact controls (27, 44). A number of factors may have contributed to the divergent results reported in the literature, including differences in the strains and species of Plasmodium used, different mouse strains utilized, and differences in the target(s) of NK cell-depleting antibodies used. Parasitemia is of greater magnitude and longer in duration when NK cells are depleted with anti-asialo GM1 polyclonal rabbit antibody but nearly identical to controls when NK cells are depleted with anti-NK1.1 MAb. Anti-NK1.1 MAb is specific for NK cells and NK T cells expressing the NK1.1 antigen. It has been suggested that anti-asialo GM1 polyclonal antibody may recognize and/or cross-react with other carbohydrate moieties (27). This could affect the function of other cell types such as macrophages, dendritic cells, neutrophils, and T cells, in addition to NK cells, resulting in the enhancement of parasitemia and the severity of infection seen in depleted mice.

We utilized the adami subspecies of P. chabaudi in studies reported here and observed that the depletion of NK cells from either WT mice or J_H^−/− mice failed to alter the time course of parasitemia compared to controls given mouse Ig. Treatment with anti-NK1.1 MAb is reported to also deplete NKT cells from liver and spleen (30). Our data indicate that both of these cell types were depleted in the spleens of WT mice and B-cell-deficient mice treated with anti-NK1.1 MAb, suggesting a minimal role, if any, for NK cells and NKT cells in the activation of either AMI or CMI against P. chabaudi adami. Otherwise, the profiles of splenic cell populations in depleted mice were similar to mouse Ig-treated controls. There were no significant differences in serum cytokine levels, including IFN-γ, between MAb-treated and control mice. Our results agree with those of others who observed that the depletion of splenic cells with anti-NK1.1 MAb does not alter the time course of P. chabaudi parasitemia (19, 41, 53).

The conclusion that NK cell or NKT cells play little if any role in protective immunity against malaria was further supported by the observations that the time course of P. chabaudi adami parasitemia was similar in J_H^−/− × β₂m^−/− double KO mice and J_H^−/− × β₂m^+/− controls, as well as in J_H^−/− × CD1d^−/− mice and their J_H^−/− × CD1d ^+/− controls. NK cells, NKT cells, and CD8⁺ T cells are lacking in J_H^−/− × β₂m^−/− mice, and NKT cells are lacking in J_H^−/− × CD1^−/− mice. The depletion of CD8⁺ T cells from WT mice by antibody treatment or gene KO of β₂m has been reported to have a minimal effect on the time course of parasitemia (45). The role of CD8⁺ T cells in CMI to P. chabaudi in B-cell-deficient mice, lacking antibody, had not been determined previously. The data obtained in the present study indicate that CD8⁺ T cells do not play an essential role in suppressing parasitemia by cytotoxicity or regulating the function of other cell types implicated in protection.

Our results confirm and extend earlier observations that γδ T cells may play an important role in malaria. They are found in elevated numbers in the blood of humans acutely infected with P. falciparum and remain at high levels for prolonged periods of time (16, 28). During experimental P. chabaudi malaria, the depletion of γδ T cells from WT mice results in significantly elevated parasitemia at several time points prior to curing (21, 52). Moreover, γδ T cells are required for the CMI-mediated suppression of P. chabaudi parasitemia in B-cell-deficient mice (49, 50, 52). The profiles of other splenic mononuclear cells in γδ T-cell-depleted mice (WT and J_H^−/− mice) infected with P. chabaudi did not differ significantly from that of control mice. In contrast to mice depleted of NK cells, the levels of serum IFN-γ and MCP-1 were significantly depressed in γδ T-cell-depleted WT and J_H^−/− mice compared to control mice.

Diverse antibody isotypes specific for AMA-1 and MSP-1 malaria antigenic constructs were present in the sera of NK cell- and γδ T-cell-depleted mice. Both depleted models displayed multiple antibody isotypes to rPCAMA-1 and rPCMSP-1, sharing similar changes to those measured in control sera. These data indicate that mice depleted of either cell type are capable of mounting a robust antibody response that is purported to suppress parasitemia by AMI (2, 22).

The critical role of IFN-γ in immunity to malaria is uncontested. The neutralization of IFN-γ by MAb in WT mice or depression of IFN-γ production in KO mice prolongs the duration of P. chabaudi parasitemia, which eventually cures (39, 46), but in J_H^−/− mice leads to high levels of noncuring parasitemia (1, 46). These data indicate that IFN-γ plays an important role in AMI immunity to malaria and an essential role in CMI to the parasite. In contrast, the function of the chemokine MCP-1 in immunity to malaria has only recently been described (34). Chemokines and their receptors, such as MCP-1 and its receptor, CCR2, play an important role in the mobilization of bone marrow cells, monocyte migration, and the polarization of the type 1 cytokine response (4). Sponaas et al. (34) observed that peak parasitemia was similar in CCR2 mice infected with the more virulent AS strain of P. chabaudi chabaudi and their controls. After reaching peak parasitemia, the clearance of parasitemia in CCR2^−/− mice is delayed, suggesting that “CD11b^high cells are part of a second wave of mechanisms controlling post-peak parasitemia, rather than part of the first line of defense” (34). As indicated above, the levels of MCP-1 were severely suppressed in both WT and J_H^−/− mice infected with P. chabaudi adami and depleted of γδ T cells by treatment with anti-GL3 (Fig. 2C). However, the time course of parasitemia in CCR2^−/− mice and heterozygous controls was nearly identical, and parasitemia again was similar in J_H^−/− mice, lacking the CCR2 receptor for MCP-1, and their controls. Taken together, these findings suggest little, if any, role for MCP-1, in immunity to P. chabaudi malaria.

Robinson et al. (26) suggest that γδ T cells are an important cell type for producing IFN-γ. γδ T cells contribute to the resolution of P. chabaudi parasitemia in immunologically intact animals and are essential for protective CMI. When γδ T cells were depleted from WT mice or J_H^−/− mice in the present study, the serum levels of IFN-γ were significantly depressed. We can speculate that early IFN-γ produced by γδ T cells is necessary for the recruitment and activation of as-yet-unidentified effector cell populations that are responsible for killing blood-stage parasites or inhibiting their growth. On the other hand, the immediate, marked and significant (P < 0.05) increase in parasitemia seen after depleting γδ T cells from J_H^−/− mice chronically infected with P. chabaudi suggests a different possibility. Our observation of exacerbating parasitemia immediately after γδ T-cell depletion in chronically infected mice suggests that at this stage of infection γδ T cells are functioning directly as killer cells. Human γδ T cells have been reported to kill blood-stage P. falciparum parasites in vitro through a granulysin-dependent mechanism (12, 13, 42). Previous studies have shown that B220⁺ γδ T cells activated by CD4⁺ T cells allow chronically infected J_H^−/− mice to control their parasitemia at low levels for extended periods, indicating that γδ T cells are continually required for control of parasitemia. Should future studies reveal the mechanisms of how these cells control and/or suppress parasitemia, it may be possible to utilize these cells or their receptors for the immunotherapy of malaria. In addition, vaccines and/or immunomodulators aimed at activating γδ T cells to proliferate and become effector cells capable of killing blood-stage malaria parasites could provide new approaches for the suppression of parasitemia during malaria.