Plasmodium–Host Interactions Directly Influence the Threshold of Memory CD8 T Cells Required for Protective Immunity

Nathan W. Schmidt, Noah S. Butler and John T. Harty
J Immunol May 15, 2011, 186 (10) 5873-5884; DOI: https://doi.org/10.4049/jimmunol.1100194

Abstract

Plasmodium infections are responsible for millions of cases of malaria and ∼1 million deaths annually. Recently, we showed that sterile protection (95%) in BALB/c mice required Plasmodium berghei circumsporozoite protein (CS252–260)-specific memory CD8 T cells exceeding a threshold of 1% of all PBLs. Importantly, it is not known if Plasmodium species affect the threshold of CS-specific memory CD8 T cells required for protection. Furthermore, C57BL/6 mice immunized with radiation-attenuated parasites are more difficult to protect against Plasmodium sporozoite challenge than similarly immunized BALB/c mice; however, it is not known whether this is the result of different CD8 T cell specificity, functional attributes of CD8 T cells, or mouse strain-specific factors expressed in nonhematopoietic cells. In this article, we show that more CS-specific memory CD8 T cells are required for protection against P. yoelii sporozoite challenge than for protection against P. berghei sporozoite challenge. Furthermore, P. berghei CS252-specific CD8 T cells exhibit reduced protection against P. berghei sporozoite challenge in the context of C57BL/6 and C57BL/10 non-MHC-linked genes in CB6F1 and B10.D2 mice, respectively. Generation and immunization of reciprocal chimeric mice between BALB/c and B10.D2 strains revealed that B10 background factors expressed by nonhematopoietic cells increased the threshold required for protection through a CD8 T cell-extrinsic mechanism. Finally, reduced CS-specific memory CD8 T cell protection in P. yoelii-infected BALB/c or P. berghei-infected B10.D2 mice correlated with increased rates of Plasmodium amplification in the liver. Thus, both Plasmodium species and strain-specific background genes in nonhematopoietic cells determine the threshold of memory CD8 T cells required for protection.

Plasmodium infections are a global health crisis resulting in ∼250 million cases of malaria and the death of ∼1 million people annually (1). Human malaria is caused by infection with multiple species of Plasmodium parasites. Importantly, memory CD8 T cells have been shown to provide protection against Plasmodium sporozoite challenge after immunization of mice with either attenuated whole parasites (2–4) or subunit vaccines (5–8). Plasmodium berghei and P. yoelii are the two predominant rodent Plasmodium species used to study liver-stage infections. Notably, there are distinct differences between these two species that may potentially mimic the diversity of Plasmodium–human infections. For example, the ID50 of sporozoites required for subsequent development of blood-stage parasitemia in BALB/c mice is ∼100 times lower for P. yoelii (∼3 sporozoites) than that for P. berghei (∼370 sporozoites) (9) However, it is not known if this Plasmodium species-specific difference in infectivity impacts the threshold of memory CD8 T cells required for resistance against sporozoite challenge.

Radiation-attenuated sporozoite (RAS) vaccines can protect mice (2) [and humans (10)] from developing blood-stage parasitemia after viable sporozoite challenge. However, RAS immunization appears to elicit more potent immunity in inbred BALB/c compared with that in C57BL/6 mice (11, 12). Several studies suggest that T cell and/or Ab responses to the circumsporozoite protein (CS) play important albeit nonessential roles in protection of RAS-immune BALB/c mice against sporozoite infection (13, 14). BALB/c and C57BL/6 mice express different MHC alleles, and thus the specificity of the immune response could account for differences in resistance. For example, transgenic mice expressing the P. yoelii CS are unable to mount CS-specific T cell or switched Ab responses to RAS immunization (13). However, RAS immunization elicited equivalent protection (defined as control of parasite burden in the liver) in control or CS-transgenic C57BL/6 mice (13). These data argue that, in contrast to BALB/c mice, C57BL/6 mice do not mount CS-specific CD8 T cell responses. Consistent with this notion, we have been unable to identify CD8 T cell responses against an overlapping CS peptide library in P. berghei RAS-immune C57BL/6 mice (N.S. Butler and J.T. Harty, unpublished observations). Thus, RAS-immune C57BL/6 mice may be more difficult to protect against sporozoite challenge because they fail to elicit a CS-specific CD8 T cell response.

Alternatively, it is well established that mouse strain-specific background genes (i.e., host factors) shape the immune response after infection with pathogens. This has been shown after infection with Leishmania major where CD4 T cells are skewed toward a Th1 or Th2 response, which influences protective immunity, in C57BL/6 and BALB/c mice, respectively (15, 16). Another example of host factors influencing protection against a pathogen is seen in the murine CMV model. Through extensive genetic mapping, it was shown that C57BL/6 mice have NK cells that encode a protective activating receptor (Ly49H), which is lacking in susceptible BALB/c mice (17).

Consistent with these observations, there are data that suggest that strain-specific background genes also may regulate protective immunity against Plasmodium sporozoite challenge in mice. For example, P. yoelii RAS-immunized B10.D2 mice are harder to protect against sporozoite challenge than similarly immunized BALB/c mice, despite sharing the H-2d MHC haplotype (11, 18). However, interpretation of these data is complicated because the magnitude of the immune response to RAS immunization was not evaluated in the published studies. Although CD8 T cells are central to RAS-induced protection against liver-stage Plasmodium infection, CD4 T cells, Abs, and NK cells also are induced by RAS and can contribute to protection (11, 13, 19–23). Thus, an additional complication relates to the unknown relative contribution of these immune responses to resistance against Plasmodium in different mouse strains. Clearly, host factors could independently influence any of these arms of the immune system. Therefore, the RAS-immunization model is not well suited to determine whether T cell specificity or host–parasite interactions controlled by strain-specific background genes specifically impact memory CD8 T cell-dependent protection against sporozoite challenge.

Using a previously described immunization approach capable of eliciting large numbers of epitope-specific CD8 T cells (24), we recently defined the threshold of P. berghei CS252-specific (25) memory CD8 T cells in BALB/c mice required for lifelong sterilizing protection against P. berghei sporozoite challenge (26). We demonstrated that BALB/c mice with a high frequency (≥1% P. berghei CS252-specific memory CD8 T cells of all PBLs) had a 95% chance of being protected against stringent P. berghei sporozoite challenge. These data defined a critical quantitative characteristic (i.e., threshold) of protective Plasmodium-specific memory CD8 T cells. Importantly, this quantitative, CD8 T cell-dependent (27) approach is well suited to address the knowledge gaps pertaining to how Plasmodium species, memory CD8 T cell specificity, or host–parasite interactions controlled by host factors influence the threshold of memory CD8 T cells required for sterilizing immunity against sporozoite challenge.

Materials and Methods

Mice and immunizations

BALB/cJ, DBA/2, and CB6F1 mice were purchased from the National Cancer Institute (Frederick, MD). B10.D2 and C3D2F1 mice were purchased from The Jackson Laboratory (Frederick, MD). Mice were housed at the University of Iowa animal care unit under the appropriate biosafety level. Mice were primed with 1 × 106 splenic dendritic cells (DCs) coated with either P. berghei CS252–260 (DC-CS252) or P. yoelii CS280–288 (DC-CS280) and boosted 7 d later with 1 × 104 to 1 × 107 CFU recombinant Listeria monocytogenes expressing either CS252–260 (LM-CS252) or CS280–288 (LM-CS280), respectively, as described previously (26). The Institution Animal Care and Use Committee approved animal experiments.

Quantification and phenotypic analysis of Ag-specific T cells

Spleens were disrupted into single-cell suspensions. Livers were perfused with cold HBSS through the hepatic portal vein and made into single-cell suspensions. Liver mononuclear cells were collected by spinning cells in 35% Percoll/HBSS. Spleen and liver mononuclear cells were treated with Tris-ammonium chloride to lyse RBCs. PBLs were obtained by treating blood with Tris-ammonium chloride to lyse RBCs. Tissues were harvested at the indicated day after DC immunization. Total spleen CS252- or CS280-specific CD8 T cells were determined by intracellular cytokine staining (ICS) for IFN-γ after 5 h of incubation in brefeldin A (Biolegend, San Diego, CA), in the presence or absence of 200 nM CS252–260 or CS280–288, respectively. Total liver CS252- or CS280-specific CD8 T cells and the percentage of PBLs that were CS252- or CS280-specific CD8 T cells were determined by ICS for IFN-γ after 5 h of incubation in brefeldin A with P815 APCs, in the presence or absence of CS252–260 or CS280–288 , respectively. After incubation, cells were surface-stained with the indicted Abs, then fixed and permeabilized with BD Cytofix/Cytoperm (BD Biosciences, San Diego, CA). ICS for IFN-γ, TNF, and IL-2 were done in the presence of BD Perm/Wash (BD Biosciences). Cell surface expression of CD62L was detected by incubating cells with 0.1 mM TAPI-2 (Peptides International, Louisville, KY) for 30 min before and during stimulation with CS252–260 or CS280–288.

TCR Vβ analysis

Memory P. berghei CS252-specific CD8 T cells were identified by staining splenocytes with Kd-CS252–260 MHC class I tetramers followed by Abs specific for Thy1.2 and CD8. CS252-specific Thy1.2+CD8+ T cells then were analyzed for expression of Vβ2-, Vβ3-, Vβ4-, Vβ5.1/5.2-, Vβ6-, Vβ7-, Vβ8.1/8.2-, Vβ8.3-, Vβ9-, Vβ10b-, Vβ11-, Vβ12-, Vβ13-, Vβ14-, and Vβ17a-FITC (BD Pharmingen).

Abs

The following Abs were used from Biolegend: IgG2A-PE/allophycocyanin (RTK2758), CD43-PE (1B11), CD127-PE (SB/199), IL-2-PE (JES6-5H4), and TNF-PE (MP6-XT22). The following Abs were used from eBioscience (San Diego, CA): CD8-FITC/PE (53-6.7), CD27-PE (LG.7F9), KLRG1-allophycocyanin (2F1), Golden Syrian Hamster IgG-allophycocyanin, and IFN-γ-allophycocyanin (XMG1.2). The following Ab was used from BD Pharmingen: CD62L-PE (MEL-14).

Sporozoite challenge

P. berghei ANKA (clone 234) or P. yoelii 17XNL sporozoites were isolated from the salivary glands of infected Anopheles stephensi mosquitoes. Naive and immunized mice were challenged i.v. with the indicated number of P. berghei or P. yoelii sporozoites. Parasitized RBCs were identified by Giemsa stain 10 d after challenge. Protection is defined as the absence of blood-stage parasites. At least 10 fields were examined for each mouse designated as protected.

Functional avidity

Splenocytes from immunized BALB/c and B10.D2 mice were stimulated with log10 dilutions of CS252–260 peptide for 5 h. CS252-specific CD8 T cells responding to peptide stimulation were identified by ICS for either IFN-γ or TNF as described above. To calculate the percentage of maximum response, each sample was normalized to the peptide dilution that generated the maximum response. Geometric mean fluorescence intensity (MFI) was calculated by gating on CD8+IFN-γ+ or CD8+TNF+ cells using FlowJo (Tree Star).

In vivo cytolytic assay

BALB/c splenocytes (Thy1.1+) were left untreated or coated with 1 μM CS252–260 for 1 h at 37°C. Cells were washed in PBS. CS252–260-coated cells were labeled with 0.8 μM CFSE, whereas uncoated cells were labeled with 0.02 μM CFSE. Labeled splenocytes (8 × 106 total, 4 × 106 CFSElo and CFSEhi, respectively) were injected i.v. into naive or CS252–260 immune BALB/c (Thy1.2+) or B10.D2 (Thy1.2+) mice. Killing of Thy1.1+ CS252–260-coated cells was detected in the blood 2 and 4 h after injection and in the spleen at 4 h after injection by analyzing CFSE-labeled Thy1.1+ cells. The percentage killing was calculated as: 100 − (100 × [(% CFSEhi/% CFSElo)/(% CFSEhi in naive mice/% CFSElo in naive mice)]).

Bone marrow chimeric mice

Bone marrow was isolated from the femur and tibia of BALB/c and B10.D2 mice. BALB/c mice were irradiated with 8 Gy (800 rad), and B10.D2 mice were irradiated with 10 Gy (1000 rad). Irradiated mice were injected i.v. with 0.6 × 107 to 1 × 107 bone marrow cells from donor B10.D2 or BALB/c mice. Mice were rested for 8–10 wk before immunization with 5 × 105 DC-CS252. Mice were boosted 7 d later with 1 × 107 LM-CS252.

Parasite burden

Mice were injected i.v. with 1000 P. berghei or 1000 P. yoelii sporozoites. The lower left lobe of the liver was excised 20 or 40 h after infection. Liver RNA was extracted using TRIzol (Invitrogen) and purified with RNeasy Cleanup (Qiagen) for quantitative real-time PCR analysis for P. berghei and P. yoelii 18S rRNA. P. berghei 18S rRNA was detected using 5′-CGCAAGCGAGAAAGTTAAAAGAA-3′ (forward primer), 5′-GAGTCAAATTAAGCCGCAAGCT-3′ (reverse primer), and 5′-FAM-TGACGGAAGGGCACCACCAGG-TAMRA-3′ (probe), which generates a 71-bp fragment. P. yoelii 18S rRNA was detected using 5′-GGGGATTGGTTTTGACGTTTTTGCG-3′ (forward primer), 5′-AAGCATTAAATAAAGCGAATACATCCTTAT-3′ (reverse primer), and 5′-FAM-CAATTGGTTTACCTTTTGCTCTTT-TAMRA-3′ (probe), which generates a 133-bp fragment, as described previously (28).

Statistical analysis

Data were analyzed using Prism4 software.

Results

Threshold for memory CD8 T cell protection against P. yoelii sporozoite challenge is greater than that against P. berghei in BALB/c mice

Previous studies demonstrated that CS280–288-specific CD8 T cells were capable of conferring protection against P. yoelii sporozoite challenge (29). Therefore, we first determined if our short-interval DC prime–recombinant L. monocytogenes boost immunization approach could generate large populations of P. yoelii CS280-specific memory CD8 T cells. BALB/c mice were immunized with DC coated with P. yoelii CS280–288 (DC-CS280) and boosted 7 d later with LM-CS280–288 (107, a maximal dose). The DC-CS280 + LM-CS280 immunization generated robust P. yoelii CS280-specific effector and memory CD8 T cell populations in the spleen, liver, and blood (Fig. 1A–C). At a memory time point, P. yoelii CS280-specific CD8 T cells represented 5–20% of all of the CD8 T cells in the spleen, liver, and blood (Fig. 1A), resulting in ∼106 Ag-specific CD8 T cells in the spleen and ∼105 cells in the liver (Fig. 1B). Importantly, this immunization approach generated P. yoelii CS280-specific memory CD8 T cells exceeding 1% of all PBLs (Fig. 1C). P. yoelii CS280-specific memory CD8 T cells exhibited a secondary memory-like phenotype (CD27lo, CD62Llo, KLRG1hi) and function (few IL-2–producing cells) (30–32) at day 78 (71 d after booster immunization) (Fig. 1D, 1E). Thus, DC-CS280 + LM-CS280 immunization generated a P. yoelii CS280-specific memory CD8 T cell response in BALB/c mice similar in size and phenotype to that obtained for P. berghei CS252-specific memory CD8 T cells (26).

FIGURE 1.
FIGURE 1.

DC-CS280 + LM-CS280 immunization generates a large P. yoelii CS280-specific CD8 T cell population. BALB/c mice were primed with 1 × 106 DC-CS280 and boosted 7 d later with 1 × 107 LM-CS280. A, Representative contour plots of P. yoelii CS280-specific CD8 T cells in the spleen, liver, and blood as determined by ICS for IFN-γ. The number is the percentage of CD8 T cells that are P. yoelii CS280-specific. B, Total number of P. yoelii CS280-specific CD8 T cells in the spleen and liver. C, Frequency of P. yoelii CS280-specific CD8 T cells among total PBLs. B and C, Data (mean ± SD) are from three mice per day. Data are representative of two independent experiments. D, Representative histograms of phenotypic (CD27, CD43, CD62L, CD127, and KLRG1) and functional (ability of IFN-γ+ cells to coproduce IL-2 or TNF) characteristics of P. yoelii CS280-specific CD8 T cells after stimulation with CS280–288. Shaded histograms are isotype controls, and filled histograms are the indicated markers. E, Percentage marker positive of P. yoelii CS280-specific memory CD8 T cells. Data (mean ± SD) are from three mice.

Next, we determined if these large numbers of P. yoelii CS280-specific memory CD8 T cells could provide protective immunity (defined as the absence of blood-stage parasitemia) to P. yoelii sporozoite challenge. To define the degree of protection after optimal immunization, both naive and immune BALB/c mice were challenged with a range of P. yoelii sporozoites (10–10,000). Importantly, 100% of naive mice developed blood-stage parasitemia at challenge doses exceeding 10 P. yoelii sporozoites with 80% of naive mice parasitized at the lowest dose (Fig. 2A). Thus, infection with as few as 10 P. yoelii sporozoites produced a stringent challenge for naive BALB/c mice. Sterile immunity against P. yoelii in immune BALB/c mice depended on challenge dose. No protection was observed in immune mice infected with 10,000 sporozoites; however, >90% protection was seen after challenge with 10 or 50 sporozoites (Fig. 2B). Protection also was observed after challenge with 100 (54%) and 1000 (29%) sporozoites (Fig. 2B). By comparison, we previously reported that 95% of immunized BALB/c mice were protected (from challenge with 1000 P. berghei sporozoites) when P. berghei CS252-specific memory CD8 T cells exceeded 1% of all PBLs (26). Therefore, sterile immunity at similar sporozoite challenge doses in DC-CS + L. monocytogens expressing CS immune BALB/c mice is more difficult to achieve against P. yoelii than P. berghei.

FIGURE 2.
FIGURE 2.

Threshold of P. yoelii CS280-specific CD8 T cells required for sterilizing protection against a P. yoelii sporozoite challenge is >1% of the PBLs. A and C, Naive BALB/c mice were challenged with the indicated number of P. yoelii sporozoites. The number represents percentage of mice that developed blood-stage parasitemia (no. parasitized/no. challenged). B, BALB/c mice were primed with 1 × 106 DC-CS280 and boosted 7 d later with 1 × 107 LM-CS280. Frequency of P. yoelii CS280-specific CD8 T cells in the blood was determined by ICS for IFN-γ. Mice then were challenged with P. yoelii sporozoites. Black bars represent mice that developed blood-stage parasitemia. Numbers within each group are the percentage of mice that were parasitized (no. parasitized/no. challenged). A and B, Data are cumulative results from two independent experiments. D, BALB/c mice were primed with 1 × 106 DC-CS280 and boosted 7 d later with 1 × 104 to 1 × 107 LM-CS280. Frequency of P. yoelii CS280-specific CD8 T cells in the blood was determined by ICS for IFN-γ after day 69. Mice were challenged with 50–100 P. yoelii sporozoites. The numbers above each group are the percentage of mice that were parasitized (no. parasitized/no. challenged). Data (mean ± SEM) are from the indicated number of mice. C and D, Data are cumulative results from six independent experiments.

To determine if we could identify a threshold of circulating P. yoelii CS280-specific memory CD8 T cells required for sterilizing immunity against P. yoelii sporozoite challenge, we immunized BALB/c mice with DC-CS280, then boosted those mice with titrating amounts (104 to 107 CFU) of LM-CS280 to generate a range of P. yoelii CS280-specific memory CD8 T cells (Fig. 2D). We challenged those mice with either 50 or 100 P. yoelii sporozoites (different challenge doses were used in independent experiments) because we observed 50–90% protection in optimally immunized mice at those challenge doses (Fig. 2B). Over the course of six independent experiments, 100% of naive mice developed blood-stage parasitemia after challenge with 50 or 100 P. yoelii sporozoites (Fig. 2C). As the average P. yoelii CS280-specific memory CD8 T cell response increased in the DC-CS280 + LM-CS280 immunized BALB/c mice, so did the frequency of mice that exhibited sterile protection (Fig. 2D). However, despite challenge with 10- to 20-fold fewer P. yoelii than P. berghei sporozoites (26), the group of mice with the largest average P. yoelii CS280-specific memory CD8 T cell response (>1.5% of all PBLs) only exhibited 56% protection (Fig. 2D). Thus, we were unable to define a threshold of P. yoelii CS280-specific memory CD8 T cells in the PBLs that correlated with high-level (>80%) protection. These data suggest that more CS-specific memory CD8 T cells are required for protection against P. yoelii sporozoite challenge than for protection against P. berghei sporozoite challenge. Thus, the number of CS-specific memory CD8 T cells required for high-level sterile immunity against sporozoite challenge is influenced by Plasmodium species.

Host factors profoundly impact protection by CS252-specific memory CD8 T cells against P. berghei

As noted, RAS-immunized C57BL/6 and B10.D2 mice are more difficult to protect against Plasmodium challenge than RAS-immunized BALB/c mice (11, 18). We next asked whether this outcome was due to the specificity or magnitude of the memory CD8 T cell response or whether strain-specific background genes could influence the threshold of Plasmodium-specific memory CD8 T cells required for sterile immunity. To address this question, we employed the P. berghei model with its defined threshold (≥1% of PBLs) of P. berghei CS252-specific memory CD8 T cells required for high-level protective immunity. P. berghei CS252–260 is presented by the H-2Kd MHC class I molecule expressed by BALB/c mice, as well as DBA/2, CB6F1, and B10.D2 mice. Therefore, these mouse strains allow us to compare the relationship between the magnitude of the P. berghei CS252-specific memory CD8 T cell response and protective immunity in the context of different mouse strain-specific background genes. DBA/2 mice have homozygous expression of H-2d MHC but differ in non-MHC-linked genes from BALB/c mice. CB6F1 mice are the F1 progeny between BALB/c and C57BL/6 mice. Therefore, they express MHC class I molecules from both parental strains (H-2b and H-2d), but due to codominant MHC expression, the surface amount of each molecule is decreased by 50% compared with that of the inbred parental strain. Finally, B10.D2 mice have C57BL/10 background genes [highly related to C57BL/6 (33, 34)] but homozygous expression of the H-2d MHC haplotype. Importantly, B10.D2 mice express the same level of surface H-2Kd as BALB/c and DBA/2 mice. After DC-CS252 + LM-CS252 immunization, we observed large P. berghei CS252-specific memory CD8 T cell populations in all three strains of mice in the spleen, liver, and blood (Fig. 3A–D). No substantial differences in phenotype (CD27lo, CD62Llo, KLRG1hi) or function (few IL-2–producing cells) were noted between P. berghei CS252-specific memory CD8 T cells in the different strains of mice (Fig. 3E). Importantly, all three strains of mice had P. berghei CS252-specific memory CD8 T cells representing ≥1% of all PBLs (Fig. 3D). Of note, we observed a significantly larger P. berghei CS252-specific memory CD8 T cell response in the spleen of B10.D2 mice compared with that in the spleen of DBA/2 mice (p < 0.05; Fig. 3B) and a significantly larger P. berghei CS252-specific memory CD8 T cell response in the liver of CB6F1 mice compared with that in the liver of either DBA/2 or B10.D2 mice (p < 0.01; Fig. 3C). These data demonstrate that DC-CS252 + LM-CS252 immunization generated very large P. berghei CS252-specific memory CD8 T cell populations in DBA/2, CB6F1, and B10.D2 mice that were similar in number and phenotype compared with that generated in BALB/c mice (26).

FIGURE 3.
FIGURE 3.

DC-CS252 + LM-CS252 immunization generates a large memory P. berghei CS252-specific CD8 T cell population in DBA/2, CB6F1, and B10.D2 mice. DBA/2, CB6F1, and B10.D2 mice were primed with 1 × 106 DC-CS252 and boosted 7 d later with 1 × 107 LM-CS252. A, Representative contour plots of P. berghei CS252-specific CD8 T cells in the spleen, liver, and blood as determined by ICS for IFN-γ. The numbers are the percentage of CD8 T cells that are P. berghei CS252-specific. B and C, Total number of P. berghei CS252-specific CD8 T cells was determined in the spleen (B) and liver (C). D, Frequency of P. berghei CS252-specific CD8 T cells among total PBLs. Data from B–D (mean ± SD) are from three mice per strain and representative of multiple experiments. C and D, Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparison posttest. E, Phenotypic (CD27, CD43, CD62L, CD127, and KLRG1) and functional (ability of IFN-γ+ cells to coproduce IL-2 or TNF) characteristics of P. berghei CS252-specific CD8 T cells after stimulation with CS252–260. Data (mean ± SD) are from three mice per strain. n.s., not significant.

To determine if host factors influenced the threshold of P. berghei CS252-specific memory CD8 T cells required for sterile immunity, DBA/2, CB6F1, and B10.D2 mice were primed with DC-CS252, then boosted 7 d later with titrating amounts (104 to 107 CFU) of LM-CS252. P. berghei CS252-specific memory CD8 T cells exhibit similar surface expression of CD27, CD62L, and CD127 and IL-2 production regardless of the LM-CS252 booster dose (Fig. 4A–C). Therefore, we ranked all of the mice from each strain by the magnitude of the P. berghei CS252-specific memory CD8 T cell response in the PBLs (Fig. 4D–F).

FIGURE 4.
FIGURE 4.

Host factors profoundly influence the threshold of P. berghei CS252-specific CD8 T cells required for sterile immunity. Mice were primed with 1 × 106 DC-CS252 and boosted 7 d later with 1 × 104 to 1 × 107 LM-CS252. A–C, Cell surface expression of CD27, CD62L, and CD127 and production of IL-2 in memory P. berghei CS252-specific memory CD8 T cells from (A) DBA/2, (B) CB6F1, and (C) B10.D2 mice as determined by ICS for IFN-γ. Data (mean ± SD) are from 3–10 mice per group. D–F, Frequency of P. berghei CS252-specific CD8 T cells was determined in the blood by ICS for IFN-γ. Naive and DC-CS252 + LM-CS252 immune mice were challenged with 1000 P. berghei sporozoites. The number for naive mice represents the percentage of mice that developed blood-stage parasitemia (no. parasitized/no. challenged). Black bars represent mice that developed blood-stage parasitemia. The number within each group is the percentage of mice with >1 or <1% CS252-specific CD8 T cells of PBLs that were parasitized (no. parasitized/no. challenged).

To be consistent with our previous report with BALB/c mice (26), we challenged all three strains of mice with 1000 P. berghei sporozoites. Thus, the only variables were the introduction of different strain-specific background genes and the 50% reduction in H-2Kd in CB6F1 mice. These variables did not affect susceptibility of naive mice because 100% of naive controls for each mouse strain developed blood-stage parasitemia after sporozoite challenge (Fig. 4D–F). Immunized DBA/2 mice containing ≥1% P. berghei CS252-specific memory CD8 T cells of all PBLs exhibited a high degree of protection (83%), whereas protection was substantially less in mice with Ag-specific T cell frequencies below this threshold (Fig. 4D). Thus, in comparison with our previous study with BALB/c mice, DBA/2 host factors minimally affect the threshold of P. berghei CS252-specific memory CD8 T cells required for sterile immunity. In striking contrast, we observed no protection in immunized CB6F1 mice after challenge, even though some mice had P. berghei CS252-specific memory CD8 T cells representing 5–7% of the PBLs (Fig. 4E). There are two potential explanations for this result. First, C57BL/6 host factors in CB6F1 mice could dramatically influence the threshold of CD8 T cells required for protective immunity. Alternatively, the 50% reduction in H-2Kd could increase the threshold of P. berghei CS252-specific memory CD8 T cells required for protecting CB6F1 mice against the P. berghei sporozoite challenge. We were able to address these possibilities with B10.D2 mice, which have C57BL/10 strain-specific background genes but homozygous expression of H-2Kd. However, none of the immunized B10.D2 mice was protected from P. berghei sporozoite challenge (Fig. 4F) despite the presence of P. berghei CS252-specific memory CD8 T cells exceeding 1% of PBLs. Therefore, C57BL/6 and C57BL/10 host factors, but not the level of MHC class I, profoundly affect the threshold of P. berghei CS252-specific memory CD8 T cells required for sterile immunity against P. berghei sporozoite challenge. We also observed substantially reduced protection in immunized C3D2F1 mice, which are the F1 progeny of C3H/HeJ (H-2k) × DBA/2 (H-2d), mice after challenge with 1000 P. berghei sporozoites (Supplemental Fig. 1). These results suggest that C3H/HeJ strain-specific background genes also impact the threshold of P. berghei CS252-specific memory CD8 T cells required for sterile immunity against P. berghei sporozoite challenge. Importantly, in no circumstance did we observe a decrease in the threshold of memory CD8 T cells required for protection compared with the ≥1% of PBLs that we described in P. berghei-infected BALB/c mice (26). These data demonstrate that host factors can profoundly influence the threshold of P. berghei CS252-specific memory CD8 T cells required for sterile immunity.

Multiple booster immunizations are required for CD8 T cell-dependent protection against P. berghei sporozoite challenge in B10.D2 mice

Our inability to protect DC-CS252 + LM-CS252 immunized B10.D2 mice against P. berghei sporozoite challenge (Fig. 4F) led us to question whether P. berghei CS252-specific memory CD8 T cells could ever protect mice with C57BL/10 background genes against P. berghei sporozoite challenge. To address this question, B10.D2 mice were primed with DC-CS252 and boosted twice with LM-CS252 (Fig. 5A) to generate very high numbers of P. berghei CS252-specific memory CD8 T cells (Fig. 5B–E). We also included groups of BALB/c (protected, Ref. 26) and B10.D2 (not protected, Fig. 4F) mice that received only a single LM-CS252 boost for comparison and to control for the length of time between the last exposure to Ag and the time when mice were challenged with P. berghei sporozoites (Fig. 5A). After immunization, there were substantially more P. berghei CS252-specific memory CD8 T cells in the spleen (Fig. 5B) and liver (Fig. 5C) of B10.D2 mice that received the second booster immunization compared with BALB/c and B10.D2 mice that received only one booster immunization. Furthermore, we observed a larger fraction of P. berghei CS252-specific memory CD8 T cells in the B10.D2 mice that received the second booster immunization among the total PBLs (Fig. 5D) and total CD8 T cells in the blood (Fig. 5E) compared with BALB/c and B10.D2 mice that received only one booster immunization. Of note, the average frequency of P. berghei CS252-specific memory CD8 T cells in B10.D2 mice that received the second booster immunization (Group 5) represented 15% of all PBLs (Fig. 5D) and 60% of all CD8 T cells (Fig. 5E). Thus, a second booster immunization generated a large number and frequency of P. berghei CS252-specific memory CD8 T cells in B10.D2 mice.

FIGURE 5.
FIGURE 5.

B10.D2 mice can be protected against P. berghei sporozoite challenge after the induction of an enormous P. berghei CS252-specific CD8 T cell response. A, Schematic of how mice were immunized. Mice were primed with 1 × 106 DC-CS252 on either day 0 or day 76, then boosted 7 d later with 1.5 × 107 LM-CS252. B10.D2 mice in Group 5 received a second boost of 1.5 × 107 LM-CS252 on day 83. B and C, P. berghei CS252-specific CD8 T cells were determined by ICS for IFN-γ in the spleen (B) and liver (C). Data (mean ± SD) are from three mice per group. D and E, Percentage P. berghei CS252-specific CD8 T cells of PBLs (D) or of CD8+ cells (E). Data (mean ± SD) are from mice used in G. B–E, Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparison posttest. F, Naive mice were challenged with 1000 P. berghei sporozoites. The number represents the percentage of mice that developed blood-stage parasitemia (no. parasitized/no. challenged). G, Percentage of P. berghei CS252-specific of PBLs in individual mice. Mice were challenged with 1000 P. berghei sporozoites. Black bars represent mice that developed blood-stage parasitemia. The number in each group equals the percentage of mice that were parasitized (no. parasitized/no. challenged). Data are representative of two independent experiments.

Naive and immunized mice then were challenged with a stringent dose (1000) of P. berghei sporozoites (Fig. 5F). Consistent with our previous challenge studies, 100% of DC-CS252 + LM-CS252 immunized BALB/c mice (Groups 1 and 2) were protected against P. berghei sporozoite challenge at both intervals after the booster immunization, whereas only 1 out of 18 DC-CS252 + LM-CS252 immunized B10.D2 mice (Groups 3 and 4) was protected (Fig. 5G). In stark contrast, >90% of DC-CS252 + LM-CS252 immunized B10.D2 mice given a second LM-CS252 boost (Group 5) were protected against the P. berghei sporozoite challenge (Fig. 5G). Consistent with these results, we also were able to protect CB6F1 mice given additional LM-CS252 booster immunizations (Supplemental Fig. 2). Although there were differences in the expression of several cell surface markers (CD27, CD62L, CD127, and KLRG1) and the ability to produce IL-2 in P. berghei CS252-specific memory CD8 T cells between the different groups of mice (Supplemental Fig. 3), those changes are unlikely to explain the differences in protection (e.g., Groups 1 and 5 exhibit the most divergence in phenotype, yet both groups are still highly protected). These results clearly demonstrate that multiple booster immunizations increase P. berghei CS252-specific memory CD8 T cells to extremely high numbers (∼15% of PBLs/∼60% of all CD8 T cells) that can overcome C57BL/10 and C57BL/6 host background susceptibility factors and confer sterile immunity after P. berghei sporozoite challenge.

Impact of host factors influence protection in a CD8 T cell-extrinsic manner

One possible explanation for the differences in protection between immunized BALB/c and B10.D2 mice could be strain-specific altered functional qualities of P. berghei CS252-specific memory CD8 T cells. These differences could include altered Vβ usage, functional avidity, and/or cytolytic activity. To determine if P. berghei CS252-specific memory CD8 T cells exhibit differences in Vβ usage, we immunized BALB/c and B10.D2 mice with DC-CS252 + LM-CS252 and screened P. berghei CS252-specific memory CD8 T cells for the utilization of multiple TCR Vβ-chains. Our results revealed no striking difference in the Vβ repertoire between the two different mouse strains (Fig. 6A). We next determined if differences in memory CD8 T cell functional avidity explained differences in protection by stimulating P. berghei CS252-specific memory CD8 T cells from either BALB/c or B10.D2 mice with log10 dilutions of CS252–260 peptide, which permits identification of the peptide concentration necessary to elicit a 50% maximal response for the effector cytokines IFN-γ or TNF. These cytokines have been demonstrated to exhibit antiparasitic properties during Plasmodium liver-stage infection (35–39). Despite the stark contrasts in protection between DC-CS252 + LM-CS252 immunized BALB/c and B10.D2 mice (Fig. 5G), no differences in functional avidity were observed for production of either IFN-γ or TNF between P. berghei CS252-specific memory CD8 T cells from these strains (Fig. 6B). Additionally, the per-cell amount (MFI) of IFN-γ or TNF produced by P. berghei CS252-specific memory CD8 T cells after stimulation did not differ between these strains of mice (Fig. 6C). These data suggest that there are no differences in the functional avidity or the amounts of IFN-γ and TNF produced by P. berghei CS252-specific memory CD8 T cells from either BALB/c or B10.D2 mice.

FIGURE 6.
FIGURE 6.

Memory P. berghei CS252-specific CD8 T cells in BALB/c and B10.D2 mice exhibit similar functional characteristics. BALB/c and B10.D2 mice were primed with DC-CS252 and boosted 7 d later with LM-CS252. A, Percentage of P. berghei CS252-specific CD8 T cells (before day 80) that are positive for the indicated Vβ proteins. P. berghei CS252-specific CD8 T cells were identified by gating on CD8+/Th1.2+/Kd-CS252–260+ cells. Data (mean ± SD) are from three mice per strain. B and C, Splenocytes (before day 80) were stimulated with titrating amounts of CS252–260. After stimulation, intracellular IFN-γ and TNF were detected by ICS. B, Percentage IFN-γ+ or TNF+ of P. berghei CS252-specific CD8 T cells, relative to the maximum response. C, MFI of IFN-γ or TNF produced by P. berghei CS252-specific CD8 T cells after stimulation with titrating amounts of CS252–260. B and C, Data (mean ± SD) are from three mice per strain and are representative of two independent experiments. D, Percentage of PBLs that are CS252-specific CD8 T cells. E and F, CFSEhi (CS252–260 pulsed) and CFSElo (CS252–260 unpulsed) cells were injected i.v. into DC-CS252 + LM-CS252 immunized BALB/c and B10.D2 mice (same mice as in D). CFSE-labeled cells were analyzed in the blood 2 and 4 h after injection and in the spleen 4 h after injection. E, Representative histograms of CFSE-transferred cells in the spleen 4 h after injection. F, Percentage specific killing of CS252–260 pulsed cells. D–F, Data (mean ± SEM) are from six mice per strain from two independent experiments. A–F, Data were analyzed by two-tailed t test. There were no significant differences between any data sets.

To determine if mouse strain-specific cytolytic activity could account for differences in protection, we transferred Thy1.1+ splenocytes pulsed with CS252–260 (CFSEhi) or without CS252–260 (CFSElo) into DC-CS252 + LM-CS252 immunized BALB/c and B10.D2 mice containing similar frequencies of P. berghei CS252-specific memory CD8 T cells in the blood (Fig. 6D). After transfer, we observed similar rates of in vivo killing of CS252–260 pulsed cells in the blood and spleen between BALB/c and B10.D2 mice (Fig. 6E, 6F). These data demonstrate that there are no differences in the in vivo target cell recognition or the cytolytic activity of P. berghei CS252-specific memory CD8 T cells in BALB/c or B10.D2 mice. Collectively, these results suggest that the difference in protection against P. berghei sporozoite challenge between DC-CS252 + LM-CS252 immunized BALB/c and B10.D2 mice is unlikely due to intrinsic characteristics of P. berghei CS252-specific memory CD8 T cells.

Because there were no differences in the functional characteristics of P. berghei CS252-specific memory CD8 T cells in BALB/c and B10.D2 mice, this suggested that the host factors affecting the protective threshold were CD8 T cell-extrinsic. To address this possibility, we made reciprocal bone marrow chimeras between BALB/c and B10.D2 mice (B10.D2 → BALB/c and BALB/c → B10.D2). Chimeric mice were rested for 8–10 wk to allow for reconstitution of the lymphoid compartment (data not shown). Mice then were immunized with DC-CS252 and boosted 7 d later with LM-CS252. Consistent with our previous experiments, the DC-CS252 + LM-CS252 immunization generated a robust P. berghei CS252-specific memory CD8 T cell response in the spleen, liver, and blood in both groups of chimeric mice (Fig. 7A–C). Importantly, 86–99% of P. berghei CS252-specific memory CD8 T cells were bone marrow donor-derived in both groups of mice (Fig. 7D). We then challenged naive BALB/c and B10.D2 mice and DC-CS252 + LM-CS252 immunized chimeric mice with a stringent dose (1000) of P. berghei sporozoites (Fig. 7E). Strikingly, we observed significantly better protection in immune B10.D2 → BALB/c mice compared with that in BALB/c → B10.D2 mice (p = 0.0005; Fig. 7F). These data suggest that mouse strain-specific background genes expressed in nonhematopoietic cells strongly impact P. berghei CS252-specific memory CD8 T cell-mediated protection against P. berghei sporozoite challenge. Collectively, these data demonstrate that the impact of host factors on the threshold of P. berghei CS252-specific memory CD8 T cells required for sterile immunity is CD8 T cell-extrinsic.

FIGURE 7.
FIGURE 7.

Host factors in nonhematopoietic cells influence Plasmodium liver-stage amplification, which impacts memory P. berghei CS252-specific CD8 T cell-mediated protection. BALB/c and B10.D2 mice were lethally irradiated. After irradiation, 0.6 × 107 to 1 × 107 donor cells were transferred into the indicated irradiated mice. Mice were rested for 8–10 wk, then primed with DC-CS252 and boosted 7 d later with LM-CS252. A–C, P. berghei CS252-specific CD8 T cells were determined by ICS for IFN-γ in the spleen (A), liver (B), and blood (C). Data (mean ± SEM) are from six mice per group from two independent experiments. Mice were analyzed >75 d after DC-CS252 immunization. D, Donor B10.D2 (Thy1.2+) and BALB/c (Thy1.1+) CS252-specific CD8 T cells were determined in recipient BALB/c (Thy1.1+) and B10.D2 (Thy1.2+) mice, respectively, >68 d after DC-CS252 immunization. Data (mean ± SEM) are from 13 mice per group from two experiments. E, Naive mice were challenged with 1000 P. berghei sporozoites. The number represents the percentage of mice that developed blood-stage parasitemia (no. parasitized/no. challenged). F, Frequency of P. berghei CS252-specific CD8 T cells of total PBLs. Each bar represents an individual mouse. Mice in both groups are ranked according to the magnitude of the P. berghei CS252-specific CD8 T cell response. DC-CS252 + LM-CS252 immunized mice were challenged with 1000 P. berghei sporozoites. Black bars represent mice that developed blood-stage parasitemia. The number within each group is the percentage of mice that were parasitized (no. parasitized/no. challenged). Data are cumulative results from two independent experiments. Protection results were analyzed by Fisher’s exact test. G, Naive BALB/c and B10.D2 mice were infected i.v. with 1000 P. berghei or P. yoelii sporozoites. P. berghei and P. yoelii 18S rRNA copy numbers were determined in the liver 20 and 40 h after infection. H, Fold increase of P. berghei and P. yoelii 18S rRNA copy numbers from 20 to 40 h. G and H, Data (mean ± SEM) are from 7–10 mice per group from two independent experiments. Data were analyzed by one-way ANOVA followed by Tukey’s multiple comparison posttest. I, Naive BALB/c and B10.D2 mice were infected i.v. with 1000 and 250 P. berghei sporozoites, respectively. P. berghei 18S rRNA copy numbers were determined in the liver 40 h after infection. Data (mean ± SEM) are from five mice per group from two independent experiments and were analyzed by two-tailed t test. J, Naive and DC-CS252 + LM-CS252 immunized BALB/c and B10.D2 mice were challenged with 1000 and 250 P. berghei sporozoites, respectively. The number represents the percentage of mice that developed blood-stage parasitemia (no. parasitized/no. challenged). Protection results were analyzed by Fisher’s exact test.

Threshold of memory CS-specific CD8 T cells required for protection correlates with the rate of Plasmodium amplification in the liver

We have demonstrated that it is more difficult to protect CS-immunized BALB/c mice against P. yoelii sporozoite challenge (Fig. 2) compared with P. berghei sporozoite challenge (26). Additionally, the protection results in the reciprocal bone marrow chimeric mice indicate that host factors in nonhematopoietic cells influence the threshold of CD8 T cells required for protection. These results could relate to differences in the parasite burden in the liver between P. berghei- and P. yoelii-infected mice as well as differences in the parasite burden between the different mouse strains infected with P. berghei. To address this possibility, we infected naive BALB/c and B10.D2 mice with 1000 P. berghei sporozoites and naive BALB/c mice with 1000 P. yoelii sporozoites. We harvested livers from the infected mice 20 and 40 h later and quantified P. berghei and P. yoelii 18S rRNA copy numbers using quantitative real-time PCR (28). Interestingly, there were similar amounts of P. berghei and P. yoelii in all three groups of mice at 20 h after infection (Fig. 7G). These results suggest that, despite differences in the P. berghei and P. yoelii sporozoite ID50 (9), the early parasite burden in the liver is similar when BALB/c mice are infected with the same number of sporozoites. Whereas Plasmodium 18S rRNA copy numbers increased in all three groups at 40 h (Fig. 7G), there was a significantly higher parasite burden in the livers of B10.D2 mice infected with P. berghei and BALB/c mice infected with P. yoelii compared with that in BALB/c mice infected with P. berghei (p < 0.001; Fig. 7G). Specifically, we observed a 34-fold increase in Plasmodium 18S rRNA copy numbers in BALB/c mice infected with P. berghei between 20 and 40 h, whereas parasite burden in P. berghei-infected B10.D2 mice increased on average 212-fold (p < 0.001) and that in P. yoelii-infected BALB/c mice increased 438-fold (p < 0.001) in the same time interval (Fig. 7H). These results suggest that higher parasite burden and/or accelerated Plasmodium amplification impacts CS-specific CD8 T cell-dependent protection against sporozoite challenge. Because infection with 1000 P. berghei sporozoites results in ∼5-fold increase in liver parasite burden at 40 h in B10.D2 mice compared with that in BALB/c mice (Fig. 7H), we infected naive B10.D2 mice with 4-fold fewer P. berghei sporozoites than BALB/c mice (250 and 1000 sporozoites, respectively), which resulted in similar parasite burdens in the liver 40 h after infection (Fig. 7I). To determine if similar parasite burdens in BALB/c and B10.D2 mice now resulted in similar levels of sterile immunity, we challenged DC-CS252 + LM-CS252 immunized BALB/c and B10.D2 mice with 1000 and 250 P. berghei sporozoites, respectively. Interestingly, despite similar parasite burdens in naive mice (Fig. 7I), we still observed significantly reduced protection in the DC-CS252 + LM-CS252 immune B10.D2 mice compared with that in BALB/c mice (p = 0.0031; Fig. 7J). Similarly, when we challenged DC-CS280 + LM-CS280 immune BALB/c mice with 10- to 20-fold fewer P. yoelii sporozoites (50–100), which likely results in reduced parasite burden compared with 1000 P. yoelii sporozoites, we still observed substantially reduced protection (<40%; Fig. 2D) compared with that in DC-CS252 + LM-CS252 immune BALB/c mice challenged with 1000 P. berghei sporozoites (Fig. 7J). Thus, failures of CS-specific memory CD8 T cells in providing sterilizing immunity correlate with faster parasite replication in the liver as opposed to parasite burden. Collectively, our results suggest that the rate of Plasmodium amplification in the liver, a property dependent on species of Plasmodium and strain-specific background genes in nonhematopoietic cells, will strongly dictate the threshold of CS-specific memory CD8 T cells required for sterile immunity.

Discussion

The 1% of PBLs threshold required for sterile immunity against sporozoite challenge defined in our initial report was 10- to 100-fold more memory CD8 T cells than were required for protection against a bacterial or viral challenge (26). This high threshold for protection by memory CD8 T cells was defined using just one mouse strain and one Plasmodium species, which raised the possibility that it may have been an overestimation, based on those specific host–parasite combinations. In contrast to this notion, here we show that varying the Plasmodium species or host factors uniformly increased the threshold for memory CD8 T cells required for protection. Thus, the 1% P. berghei CS252-specific memory CD8 T cells of PBLs threshold defined for protection against P. berghei in BALB/c mice actually may represent the low end of the spectrum required for sterile immunity against Plasmodium infection.

We have shown previously that it is more difficult to protect P. yoelii RAS-immunized mice than P. berghei RAS-immunized mice after challenge with an equivalent number of sporozoites (12). Protection after RAS immunization is generally CD8 T cell-dependent; however, Plasmodium-specific CD4 T cells, Abs, and NK cells also can contribute to protective immunity (11, 13, 19–23). Therefore, it is possible that differences in CD8 T cell, CD4 T cell, Ab, and NK cell responses to RAS immunization in addition to differences in ID50 (9) may underlie the relative difficulty in protection against P. yoelii sporozoite challenge compared with P. berghei sporozoite challenge. Importantly, we were able to rule out the contribution of CD4 T cells, Abs, and NK cells using our CD8 T cell-dependent immunization approach where we demonstrate that more Plasmodium CS-specific memory CD8 T cells were required for protection against P. yoelii sporozoite challenge compared with P. berghei sporozoite challenge. One potential explanation for the differences in thresholds of CS-specific memory CD8 T cells required for protection against P. berghei and P. yoelii sporozoite challenge is the difference in infectivity between these parasite species (9). This possibility was supported by substantially reduced protection in CS-immunized BALB/c mice challenged with 1000 P. yoelii sporozoites compared with that in BALB/c mice challenged with 1000 P. berghei sporozoites. Surprisingly, parasite burden was similar at 20 h after infection with an equivalent number of P. berghei and P. yoelii sporozoites in BALB/c mice. However, over the next 20 h P. yoelii amplified to a much higher level in the livers of BALB/c mice than did P. berghei. These data argue that P. berghei and P. yoelii sporozoites have similar capacities to infect hepatocytes but exhibit markedly different efficiencies of amplification in hepatocytes, which impacts progression toward blood-stage parasitemia and the threshold of memory CD8 T cells required for protection.

There are also data that suggest that mouse strain-specific background genes may impact RAS-induced protective immunity in mice. P. yoelii RAS-immunized B10.D2 mice were more difficult to protect than P. yoelii RAS-immune BALB/c mice, which share the same MHC haplotype but differ in host factors (11, 18). However, these studies remained difficult to interpret because the magnitude or specific nature of the immune response to RAS in these mouse strains was not characterized. Furthermore, RAS-immunized mice in most previous studies were challenged with sporozoites just 2 wk after the last booster immunization, and thus these data do not address protection by memory CD8 T cell responses (11, 18). In contrast, our ability to focus only on Plasmodium CS-specific memory CD8 T cells convincingly demonstrates that mouse strain-specific background genes expressed by nonhematopoietic cells determine P. berghei CS252-specific memory CD8 T cell-dependent protection against sporozoite challenge.

In a previous study, we examined the role of multiple CD8 T cell effector mechanisms involved in protection against P. berghei and P. yoelii sporozoite challenges (39). In that report, it was shown that the cytokines IFN-γ and TNF were important but not essential memory CD8 T cell effector molecules involved in mediating protection against sporozoite challenge. Consistent with this, administration of CpG oligonucleotides, a TLR agonist that induces inflammation, prevented P. berghei sporozoites from progressing to blood-stage parasitemia when administered up to 24 h after sporozoite infection of naive BALB/c mice. However, CpG-induced inflammation prevented P. yoelii sporozoites from progressing to blood-stage parasitemia only when administered during the first 12 h after infection of naive BALB/c mice (39). Interestingly, this window of susceptibility to inflammatory cytokines correlates with the reduced rates of Plasmodium amplification in P. berghei-infected BALB/c mice compared with that in P. yoelii-infected BALB/c mice. For example, both P. berghei and P. yoelii parasite burdens were similar at 20 h after infection, the period where both species are sensitive to TLR agonist-induced inflammation. P. berghei, which undergoes relatively modest accumulation during the next 20 h, remains sensitive to later induction of inflammatory cytokines. In contrast, the extensive amplification of P. yoelii during this later interval is associated with resistance to control by inflammatory cytokines. Thus, it is possible that sterilizing immunity by P. yoelii CS280-specific memory CD8 T cells requires interaction with the parasite-infected hepatocytes during the initial 24 h after infection during the window of cytokine-dependent inhibition. Therefore, a higher threshold of P. yoelii CS280-specific memory CD8 T cells may be required to provide protection during this extremely short window of opportunity. Perhaps the CD8 T cell-extrinsic effects of host factors in C57BL/6 and C57BL/10 mice also contribute to an environment where Plasmodium species are only susceptible to inflammation during a short interval after sporozoite infection. This possibility is consistent with both the higher rate of P. berghei amplification between 20 and 40 h after infection in B10.D2 mice compared with that in BALB/c mice and the higher threshold of CD8 T cells required for protection as well as significantly reduced protection in immunized B10.D2 mice even at similar parasite burdens compared with those in BALB/c mice.

Differences in the rate of Plasmodium amplification in the liver and protective immunity may be explained by mono- or polygenic alterations. It will be interesting to determine whether similar or distinct genetic factor(s) explain why P. berghei replicates faster in B10.D2 mice compared with BALB/c mice and why P. yoelii replicates faster than P. berghei in BALB/c mice. Through genetic mapping, it will be possible to identify the gene(s) that dictates Plasmodium amplification in the liver and protective immunity. Indeed, the highly quantitative nature of our model and the distinct differences between resistant and susceptible combinations of mouse strains and Plasmodium species should make genetic mapping a possibility. This may, in turn, lead to a new class of targets for antimalarial drugs.

The data in this report have implications for the vaccination of genetically diverse humans where individuals likely will exhibit differences in parasite biology in the liver and the threshold of memory CD8 T cells required for protection. Although these results and conclusions have been drawn from the mouse model of malaria, it is reasonable to assume that the various Plasmodium species (and substrains) causing malaria will behave differently in individual humans. This is critically important because there are multiple Plasmodium species capable of infecting humans that exhibit different clinical/pathological outcomes (40). Furthermore, these different Plasmodium species are often endemic in the same regions of the world, and thus vaccination would require dealing with multiple species at the same time. Importantly, memory CD8 T cell requirements for sterile immunity against one Plasmodium species may not translate to others.

In conclusion, our data demonstrate that interactions between different Plasmodium species and mammalian hosts impact the threshold of Plasmodium-specific memory CD8 T cells required for sterilizing immunity. This is in part a result of differences in the rate Plasmodium amplification in the liver. These data also suggest that the high threshold (>1% of PBLs) of P. berghei CS252-specific memory CD8 T cells required for protective immunity in BALB/c mice may define the low end of the spectrum for memory CD8 T cell-mediated protection. Thus, it is likely that vaccines inducing large populations of memory CD8 T cells will have the best chances for protecting humans against malaria.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Jeffrey Nolz and Vladimir Badovinac for critically reviewing the manuscript, Jemmie Hoang, Lisa Hancox, Lecia Epping, Brendan Dunphy, and Lyric Bartholomay for providing P. berghei-infected mosquitoes, and the New York University Insectary for providing P. yoelii-infected mosquitoes.

Footnotes

  • This work was supported by National Institutes of Health Grants 1-F32-AI084329 (to N.W.S.), 5-T32-AI0726024 (to N.S.B.), and AI085515 (to J.T.H.) and the Department of Microbiology and Carver College of Medicine, University of Iowa.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    CS
    circumsporozoite protein
    DC
    dendritic cell
    ICS
    intracellular cytokine staining
    LM-CS252–260
    Listeria monocytogenes expressing CS252–260
    LM-CS280–288
    Listeria monocytogenes expressing CS280–288
    MFI
    geometric mean fluorescence intensity
    RAS
    radiation-attenuated sporozoite.

  • Received January 24, 2011.
  • Accepted March 14, 2011.

References

  1. ↵
    1. World Health Organization
    . 2009. World Malaria Report 2008. World Health Organization, Geneva, Switzerland.
  2. ↵
    1. Nussenzweig R. S.,
    2. J. Vanderberg,
    3. H. Most,
    4. C. Orton
    . 1967. Protective immunity produced by the injection of x-irradiated sporozoites of plasmodium berghei. Nature 216: 160–162.
    OpenUrlCrossRefPubMed
    1. Jobe O.,
    2. J. Lumsden,
    3. A. K. Mueller,
    4. J. Williams,
    5. H. Silva-Rivera,
    6. S. H. Kappe,
    7. R. J. Schwenk,
    8. K. Matuschewski,
    9. U. Krzych
    . 2007. Genetically attenuated Plasmodium berghei liver stages induce sterile protracted protection that is mediated by major histocompatibility complex Class I-dependent interferon-gamma-producing CD8+ T cells. J. Infect. Dis. 196: 599–607.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Tarun A. S.,
    2. R. F. Dumpit,
    3. N. Camargo,
    4. M. Labaied,
    5. P. Liu,
    6. A. Takagi,
    7. R. Wang,
    8. S. H. Kappe
    . 2007. Protracted sterile protection with Plasmodium yoelii pre-erythrocytic genetically attenuated parasite malaria vaccines is independent of significant liver-stage persistence and is mediated by CD8+ T cells. J. Infect. Dis. 196: 608–616.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Li S.,
    2. M. Rodrigues,
    3. D. Rodriguez,
    4. J. R. Rodriguez,
    5. M. Esteban,
    6. P. Palese,
    7. R. S. Nussenzweig,
    8. F. Zavala
    . 1993. Priming with recombinant influenza virus followed by administration of recombinant vaccinia virus induces CD8+ T-cell-mediated protective immunity against malaria. Proc. Natl. Acad. Sci. USA 90: 5214–5218.
    OpenUrlAbstract/FREE Full Text
    1. Schneider J.,
    2. S. C. Gilbert,
    3. T. J. Blanchard,
    4. T. Hanke,
    5. K. J. Robson,
    6. C. M. Hannan,
    7. M. Becker,
    8. R. Sinden,
    9. G. L. Smith,
    10. A. V. Hill
    . 1998. Enhanced immunogenicity for CD8+ T cell induction and complete protective efficacy of malaria DNA vaccination by boosting with modified vaccinia virus Ankara. Nat. Med. 4: 397–402.
    OpenUrlCrossRefPubMed
    1. Sedegah M.,
    2. G. T. Brice,
    3. W. O. Rogers,
    4. D. L. Doolan,
    5. Y. Charoenvit,
    6. T. R. Jones,
    7. V. F. Majam,
    8. A. Belmonte,
    9. M. Lu,
    10. M. Belmonte,
    11. et al
    . 2002. Persistence of protective immunity to malaria induced by DNA priming and poxvirus boosting: characterization of effector and memory CD8(+)-T-cell populations. Infect. Immun. 70: 3493–3499.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Tao D.,
    2. G. Barba-Spaeth,
    3. U. Rai,
    4. V. Nussenzweig,
    5. C. M. Rice,
    6. R. S. Nussenzweig
    . 2005. Yellow fever 17D as a vaccine vector for microbial CTL epitopes: protection in a rodent malaria model. J. Exp. Med. 201: 201–209.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Sedegah M.,
    2. W. W. Weiss,
    3. S. L. Hoffman
    . 2007. Cross-protection between attenuated Plasmodium berghei and P. yoelii sporozoites. Parasite Immunol. 29: 559–565.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Hoffman S. L.,
    2. L. M. Goh,
    3. T. C. Luke,
    4. I. Schneider,
    5. T. P. Le,
    6. D. L. Doolan,
    7. J. Sacci,
    8. P. de la Vega,
    9. M. Dowler,
    10. C. Paul,
    11. et al
    . 2002. Protection of humans against malaria by immunization with radiation-attenuated Plasmodium falciparum sporozoites. J. Infect. Dis. 185: 1155–1164.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Doolan D. L.,
    2. S. L. Hoffman
    . 2000. The complexity of protective immunity against liver-stage malaria. J. Immunol. 165: 1453–1462.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Schmidt N. W.,
    2. N. S. Butler,
    3. V. P. Badovinac,
    4. J. T. Harty
    . 2010. Extreme CD8 T cell requirements for anti-malarial liver-stage immunity following immunization with radiation attenuated sporozoites. PLoS Pathog. 6: e1000998.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Kumar K. A.,
    2. G. Sano,
    3. S. Boscardin,
    4. R. S. Nussenzweig,
    5. M. C. Nussenzweig,
    6. F. Zavala,
    7. V. Nussenzweig
    . 2006. The circumsporozoite protein is an immunodominant protective antigen in irradiated sporozoites. Nature 444: 937–940.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Grüner A. C.,
    2. M. Mauduit,
    3. R. Tewari,
    4. J. F. Romero,
    5. N. Depinay,
    6. M. Kayibanda,
    7. E. Lallemand,
    8. J. M. Chavatte,
    9. A. Crisanti,
    10. P. Sinnis,
    11. et al
    . 2007. Sterile protection against malaria is independent of immune responses to the circumsporozoite protein. PLoS ONE 2: e1371.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Sher A.,
    2. R. L. Coffman
    . 1992. Regulation of immunity to parasites by T cells and T cell-derived cytokines. Annu. Rev. Immunol. 10: 385–409.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Hsieh C. S.,
    2. S. E. Macatonia,
    3. A. O’Garra,
    4. K. M. Murphy
    . 1995. T cell genetic background determines default T helper phenotype development in vitro. J. Exp. Med. 181: 713–721.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Webb J. R.,
    2. S. H. Lee,
    3. S. M. Vidal
    . 2002. Genetic control of innate immune responses against cytomegalovirus: MCMV meets its match. Genes Immun. 3: 250–262.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Weiss W. R.,
    2. M. F. Good,
    3. M. R. Hollingdale,
    4. L. H. Miller,
    5. J. A. Berzofsky
    . 1989. Genetic control of immunity to Plasmodium yoelii sporozoites. J. Immunol. 143: 4263–4266.
    OpenUrlAbstract
  19. ↵
    1. Schofield L.,
    2. J. Villaquiran,
    3. A. Ferreira,
    4. H. Schellekens,
    5. R. Nussenzweig,
    6. V. Nussenzweig
    . 1987. Gamma interferon, CD8+ T cells and antibodies required for immunity to malaria sporozoites. Nature 330: 664–666.
    OpenUrlCrossRefPubMed
    1. Weiss W. R.,
    2. M. Sedegah,
    3. R. L. Beaudoin,
    4. L. H. Miller,
    5. M. F. Good
    . 1988. CD8+ T cells (cytotoxic/suppressors) are required for protection in mice immunized with malaria sporozoites. Proc. Natl. Acad. Sci. USA 85: 573–576.
    OpenUrlAbstract/FREE Full Text
    1. Weiss W. R.,
    2. M. Sedegah,
    3. J. A. Berzofsky,
    4. S. L. Hoffman
    . 1993. The role of CD4+ T cells in immunity to malaria sporozoites. J. Immunol. 151: 2690–2698.
    OpenUrlAbstract
    1. Oliveira G. A.,
    2. K. A. Kumar,
    3. J. M. Calvo-Calle,
    4. C. Othoro,
    5. D. Altszuler,
    6. V. Nussenzweig,
    7. E. H. Nardin
    . 2008. Class II-restricted protective immunity induced by malaria sporozoites. Infect. Immun. 76: 1200–1206.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Doolan D. L.,
    2. S. L. Hoffman
    . 1999. IL-12 and NK cells are required for antigen-specific adaptive immunity against malaria initiated by CD8+ T cells in the Plasmodium yoelii model. J. Immunol. 163: 884–892.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Badovinac V. P.,
    2. K. A. Messingham,
    3. A. Jabbari,
    4. J. S. Haring,
    5. J. T. Harty
    . 2005. Accelerated CD8+ T-cell memory and prime-boost response after dendritic-cell vaccination. Nat. Med. 11: 748–756.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Romero P.,
    2. J. L. Maryanski,
    3. G. Corradin,
    4. R. S. Nussenzweig,
    5. V. Nussenzweig,
    6. F. Zavala
    . 1989. Cloned cytotoxic T cells recognize an epitope in the circumsporozoite protein and protect against malaria. Nature 341: 323–326.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Schmidt N. W.,
    2. R. L. Podyminogin,
    3. N. S. Butler,
    4. V. P. Badovinac,
    5. B. J. Tucker,
    6. K. S. Bahjat,
    7. P. Lauer,
    8. A. Reyes-Sandoval,
    9. C. L. Hutchings,
    10. A. C. Moore,
    11. et al
    . 2008. Memory CD8 T cell responses exceeding a large but definable threshold provide long-term immunity to malaria. Proc. Natl. Acad. Sci. USA 105: 14017–14022.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Schmidt N. W.,
    2. N. S. Butler,
    3. J. T. Harty
    . 2009. CD8 T cell immunity to Plasmodium permits generation of protective antibodies after repeated sporozoite challenge. Vaccine 27: 6103–6106.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Bruña-Romero O.,
    2. J. C. Hafalla,
    3. G. González-Aseguinolaza,
    4. G. Sano,
    5. M. Tsuji,
    6. F. Zavala
    . 2001. Detection of malaria liver-stages in mice infected through the bite of a single Anopheles mosquito using a highly sensitive real-time PCR. Int. J. Parasitol. 31: 1499–1502.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Weiss W. R.,
    2. J. A. Berzofsky,
    3. R. A. Houghten,
    4. M. Sedegah,
    5. M. Hollindale,
    6. S. L. Hoffman
    . 1992. A T cell clone directed at the circumsporozoite protein which protects mice against both Plasmodium yoelii and Plasmodium berghei. J. Immunol. 149: 2103–2109.
    OpenUrlAbstract
  30. ↵
    1. Jabbari A.,
    2. J. T. Harty
    . 2006. Secondary memory CD8+ T cells are more protective but slower to acquire a central-memory phenotype. J. Exp. Med. 203: 919–932.
    OpenUrlAbstract/FREE Full Text
    1. Masopust D.,
    2. S. J. Ha,
    3. V. Vezys,
    4. R. Ahmed
    . 2006. Stimulation history dictates memory CD8 T cell phenotype: implications for prime-boost vaccination. J. Immunol. 177: 831–839.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Wirth T. C.,
    2. H. H. Xue,
    3. D. Rai,
    4. J. T. Sabel,
    5. T. Bair,
    6. J. T. Harty,
    7. V. P. Badovinac
    . 2010. Repetitive antigen stimulation induces stepwise transcriptome diversification but preserves a core signature of memory CD8(+) T cell differentiation. Immunity 33: 128–140.
    OpenUrlCrossRefPubMed
  33. ↵
    The Jackson Laboratory. C57BL/10J. Available at: http://jaxmice.jax.org/strain/000665.html. Accessed: August 4, 2010.
  34. ↵
    The Jackson Laboratory. Inbred Strains of Mice: C57BL. Available at: http://www.informatics.jax.org/external/festing/mouse/docs/C57BL.shtml. Accessed: January 18, 2010.
  35. ↵
    1. Ferreira A.,
    2. L. Schofield,
    3. V. Enea,
    4. H. Schellekens,
    5. P. van der Meide,
    6. W. E. Collins,
    7. R. S. Nussenzweig,
    8. V. Nussenzweig
    . 1986. Inhibition of development of exoerythrocytic forms of malaria parasites by gamma-interferon. Science 232: 881–884.
    OpenUrlAbstract/FREE Full Text
    1. Nussler A.,
    2. S. Pied,
    3. J. Goma,
    4. L. Rénia,
    5. F. Miltgen,
    6. G. E. Grau,
    7. D. Mazier
    . 1991. TNF inhibits malaria hepatic stages in vitro via synthesis of IL-6. Int. Immunol. 3: 317–321.
    OpenUrlAbstract/FREE Full Text
    1. Pied S.,
    2. L. Rénia,
    3. A. Nüssler,
    4. F. Miltgen,
    5. D. Mazier
    . 1991. Inhibitory activity of IL-6 on malaria hepatic stages. Parasite Immunol. 13: 211–217.
    OpenUrlCrossRefPubMed
    1. Nüssler A.,
    2. J. C. Drapier,
    3. L. Rénia,
    4. S. Pied,
    5. F. Miltgen,
    6. M. Gentilini,
    7. D. Mazier
    . 1991. L-arginine-dependent destruction of intrahepatic malaria parasites in response to tumor necrosis factor and/or interleukin 6 stimulation. Eur. J. Immunol. 21: 227–230.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Butler N. S.,
    2. N. W. Schmidt,
    3. J. T. Harty
    . 2010. Differential effector pathways regulate memory CD8 T cell immunity against Plasmodium berghei versus P. yoelii sporozoites. J. Immunol. 184: 2528–2538.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Anstey N. M.,
    2. B. Russell,
    3. T. W. Yeo,
    4. R. N. Price
    . 2009. The pathophysiology of vivax malaria. Trends Parasitol. 25: 220–227.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section