A Multifactorial Mechanism in the Superior Antimalarial Activity of α-C-GalCer

We have previously shown that the C-glycoside analog of α-galactosylceramide (α-GalCer), α-C-GalCer, displays a superior inhibitory activity against the liver stages of the rodent malaria parasite Plasmodium yoelii than its parental glycolipid, α-GalCer. In this study, we demonstrate that NK cells, as well as IL-12, are a key contributor for the superior activity displayed by α-C-GalCer. Surprisingly, the diminished production of Th2 cytokines, including IL-4, by α-C-GalCer has no affect on its superior therapeutic activity relative to α-GalCer. Finally, we show that the in vivo administration of α-C-GalCer induces prolonged maturation of dendritic cells (DCs), as well as an enhanced proliferative response of mouse invariant Vα14 (Vα14i) NKT cells, both of which may also contribute to some degree to the superior activity of α-C-GalCer in vivo.


Introduction
Malaria remains one of the most serious and prevalent infectious diseases in the world. Approximately 500 million people acquire the disease annually, which leads to 2-3 million deaths per year. Malaria is caused by species of Plasmodium, an apicomplexan parasite transmitted in nature by mosquito vectors of the genus Anopheles. The life cycle of Plasmodium is the same no matter what the species, consisting of both sexual stages that occur in the mosquito vector and asexual stages that occur in a vertebrate host. The asexual stages can be divided into the pre-erythrocytic stages, which consist primarily of developmental events occurring within infected hepatocytes (the so-called liver stages), and the erythrocytic, or blood, stages, which occur in red blood cells, and which result in the signs and symptoms of malaria [1].
Numerous studies have shown that T cells, including CD8+T cells, CD4+T cells, γδ-T cells, and natural killer (NKT) cells play a key role in the immune response to the pre-erythrocytic stages of Plasmodium [2]. Of recent interest is the role of NKT cells, a unique population of lymphocytes that coexpress markers of NK cells along with a semi-invariant TCR. In mice, the TCR of most NKT cells consists of an invariant Vα14Jα18 α chain paired with a variable set of β chains consisting primarily of Vβ8.2, Vβ7, or Vβ2. This invariant Vα14 TCR (Vα14i) enables NKT cells to recognize the MHC class I-like molecule CD1d, which is capable of presenting hydrophobic molecules such as lipids and hydrophobic peptides to Vα14i NKT cells [3,4].
To date, only a few compounds have been shown to bind CD1d and stimulate Vα14i NKT cells. Of these, the prototype molecule is α-galactosylceramide (α-GalCer), a glycolipid originally derived from a marine sponge extract. Previously, using a murine malaria model, we showed that α-GalCer is able to directly mediate protection to the pre-erythrocytic stages of malaria in a manner dependent on iNKT cells, CD1d, and IFN-γ [5]. More recently, we demonstrated that a structurally similar C-glycoside analog of α-GalCer, α-C-galactosylceramide (α-C-GalCer), is also capable of mediating protection to the pre-erythrocytic stages of malaria in a manner dependent on Vα14i 13  cells, CD1d, and IFN-γ [6]. Moreover, we showed that α-C-GalCer, which differs from α-GalCer only in the identity of the chemical group involved in the glycosidic linkage of the galactose and ceramide moieties of the molecule (Figure 1), exhibits a much more potent and longer lasting antimalarial effect than α-GalCer, an effect which appears to stem from prolonged downstream IFN-γ production by NK cells, and which requires the presence of IL-12 [6]. In a follow-up study, we showed that CD8α+ dendritic cells (DCs) are the cells responsible for producing this IL-12 in response to glycolipid injection, and that depletion of these cells results in attenuated downstream IFN-γ production by NK cells [7]. Our findings were corroborated by a more recent study by Fujii et al., which demonstrated that α-C-GalCer induces higher degree of DC activation than α-GalCer [8].
In the present study, we further examine the physiologic mechanism underlying the superior antimalarial activity of α-C-GalCer. We find that the mechanism involves IL-12, NK cells and DCs, but not the Th2 cytokines IL-4 and IL-10, nor the TCRβ used by the Vα14i NKT cells.

Materials and Methods
was synthesized by Kirin Brewery (Gumma, Japan). The stock solution was dissolved in a 0.5% polysorbate-20 (Nikko Chemical, Tokyo), 0.9% NaCl solution at a concentration of 200 μg/mL, and diluted in PBS to the desired concentration just prior to injection into mice.

Mice.
Six-to eight-week-old female BALB/c and C57BL/ 6 mice were purchased from the National Cancer Institute (Bethesda, MD). IL-12p40-deficient mice of BALB/c and C57BL/6 backgrounds, IL-4-deficient mice of BALB/c background, and IL-10-deficient mice of C57BL/6 background were all purchased from the Jackson Laboratory (Bar Harbor, ME).

Injections.
For glycolipid treatments, mice were injected intraperperitoneally with 1 μg of either α-GalCer or α-C-GalCer. In some experiments, treated and untreated mice were challenged with live P. yoelii sporozoites 3 days later.

Parasites and Their
Use for Challenge. P. yoelii (17XNL strain) was maintained by alternate cyclic passages in Anopheles stephensi mosquitoes and Swiss Webster mice. Sporozoites obtained from dissected salivary glands of infected mosquitoes 2 weeks after their infective blood meal were used for challenge of the mice. Challenge of mice to determine the development of liver-stage malaria infection was performed by an intravenous injection of 10,000 viable sporozoites into the tail vein. The outcome of the challenge was determined 40-42 hours later by measuring the parasite burden in the livers of the mice using a quantitative real-time RT-PCR method [10].

Determination of Malaria Liver Stage Development.
The degree of liver stage development in challenged mice was determined by quantifying the amount of P. yoelii-specific 18S ribosomal RNA (rRNA) molecules in the livers of the Journal of Biomedicine and Biotechnology 3 mice by way of a recently developed real-time RT-PCR technique [10]. 40-42 hours after sporozoite challenge, livers from challenged mice were dissected and homogenized in 4 mL/liver of denaturing solution (4 M guanidium thiocyanate, 25 mM sodium citrate pH 7.0, 0.5% sarcosyl, 0.1 M β-mercaptoethanol) using a PowerGen 125 biohomogenizer (Fisher, Pittsburgh, PA). RNA was then purified from 600 μL aliquots of the homogenates using the method developed by Chomczynski and Sacchi [11]. 2 μg samples of the RNA were then reverse-transcribed, and aliquots of the resulting complementary DNA (cDNA) (133 ng) were used for real-time PCR amplification of P. yoelii 18S rRNA sequences. This amplification was performed in a GeneAmp 5700 Sequence Detection System (PE Applied Biosystems, Foster City, CA). For this purpose, we used primers 5 -GGGGATTGGTTTTGACGTTTTTGCG-3 (54 nM) and 5 -AAGCATTAAATAAAGCGAATACATCCTTAT-3 (60 nM) together with the double-stranded DNA (dsDNA)-specific dye SYBR Green I incorporated into the PCR reaction buffer (PE Biosystems, Foster City, CA) in order to detect the PCR product generated. The temperature profile of the reaction was 95 • C for 10 minutes followed by 40 cycles of denaturation at 95 • C for 15 seconds and annealing/extension at 60 • C for 1 minute.

Isolation of Splenocytes and Intrahepatic
Lymphocytes from Mice. Mouse splenocytes were prepared by gently grinding spleens between the frosted ends of two microscope slides in a petri dish containing 10 mL of DMEM-10 or RPMI-10 medium. The resulting cell suspension was filtered through nylon mesh, pelleted, and incubated with 5 mL of ACK lysis buffer for 5 minutes at room temperature to remove red blood cells. The cells were then washed three times with DMEM-10 or RPMI-10 medium, filtered once more through nylon mesh, and counted using a hemocytometer and tryphan blue (GIBCO, Carlsbad, CA).
Intrahepatic lymphocytes from mice were prepared by grinding livers in a steel mesh screen using a 3 mL syringe pestle, and filtering the resulting homogenate into a 50 mL tube using 45 mL of Hank's Balanced Salt Solution (HBSS) (GIBCO, Carlsbad, CA). After centrifugation of the homogenate, the pellet was resuspended in 45 mL of a 35% Percoll (Amersham Biosciences, Piscataway, NJ) solution (29.1 mL HBSS, 14.2 mL Percoll, 1.6 mL 10X PBS, 200 U heparin), and centrifuged at 500 g for 10 minutes at 20 degrees Celsius. The resulting high-density liver lymphocyte pellet was resuspended in 5 mL of ACK lysis buffer for 5 minutes at room temperature to remove red blood cells, and then washed three times with DMEM-10 or RPMI-10 medium. After the final wash, the cells were filtered through nylon mesh, and counted using a hemocytometer and tryphan blue.

Flow Cytometry Experiments.
In order to measure the degree of DC maturation induced by α-GalCer and α-C-GalCer, freshly isolated splenocytes from BALB/c mice were first incubated for 15 minutes at 4 • C with unlabeled antimouse Fcγ III/II receptor mAb clone, 2.4G2 (PharMingen, San Diego, CA) in staining buffer (PBS containing 1% FBS and 0.1%NaN 3 ) to block Fc receptors. Next, the cells were surface stained with PE-conjugated anti-CD11c mAb clone, HL3, and FITC-conjugated anti-CD86 mAb clone, GL1, FITC-conjugated anti-CD40 mAb clone, HM40-3, or syngeneic MHC-class II-specific FITC-conjugated anti-IA d mAb clone, AMS-32.1 (all PharMingen, San Diego, CA) in staining buffer for 30 minutes at 4 • C. After two washes in staining buffer, the cells were then analyzed using a FACSCalibur instrument (Becton Dickson, San Diego, CA) with CELLQuest software (Becton Dickson, San Diego, CA).
Next, in order to specifically detect Vα14i NKT cells, we first loaded mouse CD1d-IgG 1 dimer X molecules (PharMingen, Sand Diego, CA) with a 10-20-fold molar excess of α-GalCer overnight in 1X PBS at room temperature. After preparation of splenocytes and liver lymphocytes, we incubated the cells with the loaded dimers for 60 minutes in staining buffer at 4 degrees Celsius (after Fc receptor block), using 4 μg of loaded dimer for every 1 × 10 6 cells stained. Next, after washing the cells once with staining buffer, we surface stained the cells with FITC-labeled anti-CD3ε mAb clone, 145-2C11, and PE-labeled antimouse IgG 1 clone, A85-1 (both PharMingen, San Diego, CA) for 30 minutes in staining buffer at 4 degrees Celsius. Finally, after washing the stained cells twice with staining buffer, we analyzed them by FACS as detailed above.
Finally, in order to measure the TCRβ usage of Vα14i NKT cells before and after in vivo α-GalCer or α-C-GalCer stimulation, we first incubated splenocytes and liver lymphocytes with unlabeled antimouse Fcγ III/II receptor mAb clone, 2.4G2 (PharMingen, San Diego, CA) and unlabeled streptavidin (Molecular Probes, Eugene, OR) for 15 minutes in staining buffer at 4 degrees Celsius in order to block Fc receptors and surface biotin molecules. After washing the cells twice with staining buffer, we incubated the cells with α-GalCer-loaded dimers in the manner indicated above. After dimer incubation, we surface stained the cells with FITClabeled anti-CD3ε mAb clone, 145-2C11, and either biotinlabeled anti-Vβ8.1/8.2 clone, MR5-2, biotin-labeled anti-Vβ7 clone, TR310, or biotin-labeled anti-Vβ2, clone B20.6 (all PharMingen, San Diego, CA) for 30 minutes in staining buffer at 4 degrees Celsius, followed by two washes with staining buffer. We then stained the cells with PE-labeled antimouse IgG 1 clone A85-1 and APC-labeled streptavidin (both from PharMingen, San Diego, CA) for 30 minutes in staining buffer at 4 degrees Celsius. Finally, after washing the cells twice with staining buffer, we analyzed them by FACS as detailed above.

IL-12 Is a Key Factor in the Antimalarial Activity of
Both α-C-GalCer and α-GalCer. In a previous study, we showed that α-C-GalCer's ability to stimulate prolonged IFN-γ production is abrogated in the absence of IL-12 [6]. Since IFN-γ is required for α-C-GalCer's anti-malaria effect, we wanted to see if its superior therapeutic activity against malaria liver stages was abrogated in the absence of IL-12 as well. To address this issue, we first treated both WT and IL-12-deficient mice with equal doses of either glycolipid three days before challenge with sporozoites, and then measured malaria liver stage development. As expected, in WT mice α-C-GalCer suppressed liver stage development to a much greater degree than α-GalCer; however, in IL-12-deficient mice the anti-malaria activity of both glycolipids was totally abolished (Figure 2(a)). Thus, IL-12 is a key factor not only driving α-C-GalCer's superior antimalarial effect, but also mediating the antiplasmodial effect of both glycolipids.

α-C-GalCer's Enhanced Antimalarial Response Requires the Presence of NK Cells.
Our finding that α-C-GalCer's superior antimalarial activity requires IL-12 (Figure 2(a)), which is required for optimal IFN-γ production by NK cells following glycolipid administration [6], strongly suggests that NK cells are also required for α-C-GalCer's superior antimalarial activity. To assess the role of NK cells in α-GalCerand α-C-GalCer-mediated protection against malaria, we pretreated mice with anti-asiaoloGM1 antibody, which is known to selectively deplete NK cells [12], or with PBS, and one day later treated the mice with equal doses of α-GalCer or α-C-GalCer, or with nothing. Three days after glycolipid treatment, we challenged the mice with sporozoites and then measured liver stage development as before. We found that in nondepleted PBS-treated control mice, α-C-GalCer exhibited better antimalarial activity than did α-GalCer, as expected. In contrast, we observed an abrogation of α-C-GalCer's superior antimalarial activity in mice depleted of NK cells with anti-asialoGM1 antibody (Figure 2(b)). This result strongly suggests that α-C-GalCer's superior antimalarial effect stems from its ability to stimulate enhanced IL-12 production, which then triggers NK cells to produce more IFN-γ needed to suppress malarial liver stage development.

α-C-GalCer's Enhanced Antimalarial Response Does Not
Involve the Th2 Cytokines IL-4 and IL-10. In an earlier study, we and others showed that mice injected with α-C-GalCer produce a far less IL-4 than did those injected with α-GalCer [6,8]. Given that IL-4 is a Th2 cytokine with known inhibitory activity against Th1 cytokines [13], we were curious to see if α-C-GalCer's enhanced therapeutic activity against diseases ameliorated by Th1 cytokines stems from its ability to stimulate lower amounts of IL-4. To address this issue, we injected both WT mice and mice deficient in IL-4 with equal doses of α-GalCer or α-C-GalCer, or with nothing, and 3 days later challenged them with live P. yoelii sporozoites. Forty-two hours after the challenge, we obtained livers from all the mice and determined the degree of malaria liver stages that developed by way of quantitative real-time RT-PCR. We found that the ability of α-C-GalCer to better inhibit liver stages was the same in both IL-4-deficient mice and WT mice (Figure 3(a)). This result indicates that α-C-GalCer's superior antimalarial activity does not involve its reduced IL-4 production. Another Th2 cytokine known to be produced following α-GalCer injection in mice is IL-10 [14,15]. One of the important effects of IL-10 is the direct inhibition of IL-12 production by APCs, and the consequent downregulation of Th1-type responses [13,16,17]. Given IL-10's production consequent to glycolipid injection in mice, and its inhibitory affect on Th1-type responses, we wanted to see if IL-10 plays a role in α-C-GalCer's enhanced therapeutic activity against diseases ameliorated by Th1 cytokines. To address this issue, we injected IL-10-deficient mice and WT mice with equal doses of α-GalCer or α-C-GalCer, or with nothing, and 3 days later challenged them with live sporozoites for a liver stage protection experiment. As with IL-4 deficient mice, we found that IL-10-deficient mice injected with α-C-GalCer inhibited malarial liver stages better than those injected with α-GalCer-the same as WT mice (Figure 3(b)). Thus, it appears that IL-10 also plays no role in α-C-GalCer's enhanced antimalarial activity relative to α-GalCer.

α-C-GalCer Induces a Prolonged In Vivo Maturation of DCs Compared to α-GalCer.
In a previous study, we showed that CD8α+ DCs are the cells responsible for producing IL-12 in response to α-GalCer and α-C-GalCer injection, and that depletion of these cells results in attenuated downstream IFN-γ production by NK cells [7]. Given the importance of DCs in the in vivo physiological response to α-GalCer and α-C-GalCer, we wanted to see how these two glycolipids affect these cells in vivo in WT mice. To address this issue we looked at the ability of α-GalCer and α-C-GalCer to induce maturation of DCs by following the expression patterns of various surface markers known to be upregulated during DC maturation, namely, MHC class II, CD86, and CD40. More specifically, we injected WT mice with α-GalCer, α-C-GalCer, or nothing, and 2, 6, and 24 hours later obtained splenocytes for FACS analysis of cells coexpressing CD11c and MHC class II, CD86, or CD40. We found that the first marker to show upregulation on CD11c+ DCs after injection of either glycolipid was MHC class II. α-GalCer-treated mice showed increased expression of this marker as soon as 2 hours after injection, while α-C-GalCer-treated mice showed increased expression 6 hours after injection (Figure 4(a)). By 24 hours posttreatment, we saw the highest MHC class II expression on CD11c+ DCs from mice injected with either α-GalCer or α-C-GalCer, with α-GalCer-treated mice expressing slightly more marker than α-C-GalCer-treated mice (Figure 4(a)).
The next marker to show upregulation after glycolipid injection was CD86. In α-GalCer-treated mice, CD86 expression first started showing an increase at 6 hours postinjection; whereas in α-C-GalCer-treated mice this marker did not show increased expression until 24 hours postinjection (Figure 4(b)). Again, as with MHC class II, we observed the highest expression of CD86 24 hours after injection of α-GalCer or α-C-GalCer, with α-GalCer-treated mice expressing more marker than α-C-GalCer-treated mice at this time point (Figure 4(b)).
Finally, the last DC marker to show upregulation following glycolipid injection was CD40. In both α-GalCerand α-C-GalCer-injected mice, increased expression of this marker was only observed 24 hours after treatment. At 2 and 6 hours postglycolipid injection, CD40 expression on DCs was the same as untreated controls (Figure 4(c)). Interestingly, at 24 hours postinjection the expression levels of CD40 on DCs from both α-GalCerand α-C-GalCer-treated mice were more or less the same, in contrast to MHC class II and CD86 (Figure 4(c)). Overall, the upregulation data indicates that α-GalCer induces a faster maturation of CD11c+ DCs than does α-C-GalCer, which appears to induce a more prolonged maturation of this cell type. [18][19][20][21]. This proliferation is accompanied by early downregulation of Vα14i TCRs and NK1.1 on the surface of NKT cells, followed by reappearance of these markers two days later. A finding from earlier studies that α-C-GalCer stimulates less cytokine production by NKT cells than α-GalCer [6,8] suggests that α-C-GalCer is a weaker NKT cell stimulus than α-GalCer. To see if α-C-GalCer also stimulates poorer proliferation of NKT cells when compared to α-GalCer, we injected WT mice with equal doses of either glycolipid, and 5, 24, 48, 72, 120, and 168 hours later obtained splenocytes and liver lymphocytes for FACS analysis of Vα14i NKT cells. To detect Vα14i NKT cells we made use of recently developed mouse CD1d-IgG1 dimers, which, when loaded with α-GalCer, stain Vα14i NKT cells [22].

α-C-GalCer Induces a Slower and Shorter In Vivo Downregulation of Vα14i TCRs and a Greater In Vivo Proliferation of Vα14i NKT Cells Compared to α-GalCer. A number of recent studies have shown that Vα14i NKT cells proliferate in vivo following injection of α-GalCer
We found that Vα14i NKT cells from α-GalCer-treated mice rapidly downregulated their TCRs, becoming undetectable in both spleen and liver by 5 hours, and remaining so at 24 hours ( Figure 5(a)). Not until 48 hours did Vα14i NKT cells from α-GalCer-treated mice become detectable again in Spleen Liver   both spleen and liver. At this time point the cells returned to levels comparable to that detected prior to injection, and remained so at 72, 120, and 168 hours postinjection ( Figure 5(a)).

α-GalCer α-C-GalCer α-GalCer α-C-GalCer
In contrast to α-GalCer, Vα14i NKT cells from mice treated with α-C-GalCer exhibited a slower and shorter TCR downregulation, with small percentages of cells still present at 5 hours in both spleen and liver ( Figure 5(a)). At 24 hours the cells were almost completely undetectable, but started reappearing at 48 hours, although at lower levels than that detected prior to injection (Figure 5(a)). Strikingly, at 72 hours the levels of Vα14i NKT cells in the spleens and livers of α-C-GalCer-treated mice were greatly increased over that observed prior to injection. In the spleen, the percentage of cells was approximately 10 times higher than that observed at the start; similarly, in the liver the percentage was about 3 times higher (Figure 5(a)). At 120 hours, the levels of Vα14i NKT cells started to come down in both the spleen and liver, but still remained higher than that observed prior to injection. By 168 hours, the percentages continued to come down, approaching baseline in the liver, but remaining high in the spleen (Figure 5(a)).
Using the percentages of Vα14i NKT cells we were able to follow the overall proliferative response of the cells stimulated by the glycolipids. We found that the overall pattern of Vα14i NKT cell expansion was the same for both glycolipids, with peak cell numbers occurring 72 hours postinjection followed by a return towards baseline ( Figure 5(b)). Surprisingly, we found that α-C-GalCer stimulated a far greater expansion of Vα14i NKT cells in both spleen and liver than did α-GalCer. The difference was most striking at 72 hours when the numbers of Vα14i NKT cells in the spleens and livers of α-C-GalCer-treated mice were approximately 5 times higher than those in α-GalCer-treated mice ( Figure 5(b)). Overall, this data indicates that despite its poor ability to stimulate cytokine production by NKT cells [6], α-C-GalCer is a better in vivo stimulus for Vα14i NKT cell proliferation than α-GalCer.

TCRβ Usage Does Not Affect In Vivo Vα14i NKT Cell Activation Stimulated by α-GalCer or α-C-GalCer.
Two recent studies show that TCRβ usage by Vα14i NKT cells affects the avidity of Vα14i TCRs for CD1d-glycolipid complexes [22,23]. In particular, they indicate that TCRβ chains encoded by the Vβ8.2 gene segment confer higher avidity to Vα14i TCRs than Vβ7 or Vβ2, and that glycolipid ligands preferentially stimulate these higher avidity cells. Despite these findings, our data showing a complete disappearance of Vα14i NKT cells by 24 hours post-α-GalCer or α-C-GalCer injection ( Figure 5(a)) suggests that in vivo both glycolipids stimulate all Vα14i NKT cells regardless of their TCRβ usage. To confirm this suspicion, we treated WT mice with α-GalCer, α-C-GalCer, or with nothing, and 72 hours later obtained splenocytes and liver lymphocytes for staining with CD1d-IgG1 dimer loaded with α-GalCer. In addition to dimer staining, we also stained the cells for Vβ8.1/8.2, Vβ7, and Vβ2 to see if the distribution Vα14i NKT cells changes after glycolipid-mediated stimulation. More specifically, if α-GalCer or α-C-GalCer preferentially stimulates the higher avidity Vβ8.2-containing NKT cells in vivo, then there should be an increase in the proportions of these cells in the expanded NKT cell population that results after glycolipid-induced proliferation. We found that there was no difference in the proportions of TCRβ usage by Vα14i NKT cells before or after stimulation with α-GalCer or α-C-GalCer ( Figure 6).

Discussion
The current study further defines the physiologic mechanism by which α-C-GalCer exhibits its superior antimalarial activity when compared to its structurally similar analog α-GalCer. As reported previously α-C-GalCer's enhanced antimalarial effect appears to stem from prolonged downstream IFN-γ production by NK cells requiring IL-12 [6]. This requirement for IL-12 is reconfirmed in the present study by our results showing the abrogation of α-C-GalCer's enhanced therapeutic effect against malaria liver stages in mice lacking IL-12 (Figure 2(a)). The current study also definitively defines the important role of NK cells in the mechanism of α-C-GalCer's enhanced antimalarial effect.
Our finding that NK cell depletion abrogates α-C-GalCer's superior antimalarial activity when compared to α-GalCer attests to the key role these cells play (Figure 2(b)). In addition, despite a previously reported finding that α-C-GalCer stimulates less IL-4 than α-GalCer [6,8], the ability of α-C-GalCer to stimulate diminished production of Th2 cytokines like IL-4 and IL-10 has no affect on its superior therapeutic activity relative to α-GalCer ( Figure 3). Taken together, these data, along with our prior results showing that CD8α+ DCs are the cells responsible for producing IL-12 in response to glycolipid injection [7], indicate that α-C-GalCer's superior antimalarial activity stems from its ability to stimulate prolonged IL-12 production by DCs thereby resulting in enhanced downstream IFN-γ production by NK cells, and enhanced protection (Figure 7). Our previous finding that α-C-GalCer stimulates prolonged IL-12 by DCs [7] correlates with our current finding that α-C-GalCer induces prolonged maturation of DCs after injection into mice (Figure 4). The fact that DCs mature at a slower rate after α-C-GalCer administration than after α-GalCer administration implies that the NKT cell inputs into α-C-GalCer-bearing DCs are weaker than those delivered to α-GalCer-bearing DCs. These weaker inputs might stem from a lower affinity interaction between the Vα14i TCR and the CD1d-α-C-GalCer complex. In light of studies demonstrating NKT cell TCR downregulation hours after encounter with α-GalCer-bearing APCs [19][20][21], as well as other studies showing that high affinity interactions result in faster and greater TCR downregulation [24,25], the weaker Vα14i TCR:CD1d-α-C-GalCer interaction would be expected to result in slower and lesser downregulation of Vα14i TCRs on NKT cells than that seen with α-GalCer. Indeed, our experiment looking at the disappearance and reappearance of Vα14i NKT cells after α-GalCer or α-C-GalCer administration demonstrated precisely this phenomenon ( Figure 5). As a consequence of shorter/lesser TCR downregulation, longer contact between NKT cells and α-C-GalCer-bearing DCs probably occurs, which would result in a temporal summation of NKT cell inputs by DCs that might exceed that achieved by α-GalCer-bearing DCs, which receive stronger-but shorter-lived inputs from NKT cells due to the higher affinity interaction between the Vα14i TCR and the CD1d-α-GalCer complex. Since glycolipid-induced IL-12 production by DCs requires contact with NKT cells [26][27][28][29][30][31], α-C-GalCer's ability to stimulate prolonged IL-12 might stem from longer contact time between NKT cells and α-C-GalCer-bearing DCs.
In addition to stimulating a slower and shorter downregulation of TCRs on Vα14i NKT cells, the interaction between CD1d-α-C-GalCer complexes and Vα14i TCRs also stimulates a greater in vivo expansion of Vα14i NKT cells compared to α-GalCer ( Figure 5). This result is surprising because earlier data from us and others demonstrating that α-C-GalCer stimulates less cytokine production by NKT cells than α-GalCer [6,8] implies that α-C-GalCer is a weaker agonist for NKT cells. It is likely that the striking differences we observed in the abilities of α-GalCer and α-C-GalCer to induce cytokine synthesis and cellular expansion by NKT cells are the result of differential TCR signal transduction events, which probably stem from the difference in affinities between the Vα14i TCR:CD1d-α-GalCer interaction and the Vα14i TCR:CD1d-α-C-GalCer interaction.
This situation is analogous to a previous study showing that a partial T cell agonist incapable of stimulating early T cell activation events was capable of stimulating later events, such as proliferation, to the same degree or better as the full agonist [32]. These findings were explained by the so-called kinetic proofreading model for T cell activation, which proposes that while early T cell activation events depend on the affinity of TCR:MHC-peptide interactions, later T cell activation events depend more on the temporal summation of successive TCR signals delivered by MHCpeptide complexes serially engaging different TCRs over time [24,[32][33][34][35][36]. Based on this model, it is possible that α-C-GalCer is a partial NKT cell agonist, whose ability to stimulate enhanced Vα14i NKT cell expansion is due to enhanced temporal summation of APC inputs by NKT cells made possible by slower and shorter TCR downregulation and prolonged contact between NKT cells and α-C-GalCerbearing APCs. While more research is needed to verify this mechanism, it is interesting to note that TCRβ usage does not appear to effect α-C-GalCer's stimulation of NKT cells ( Figure 6) despite two recent studies showing that TCRβ chains encoded by the Vβ8.2 gene segment confer higher avidity to Vα14i TCRs than Vβ7 or Vβ2, and that glycolipid ligands preferentially stimulate these higher avidity cells [22,23].

Conclusions
We have previously shown that the C-glycoside analog of α-GalCer, α-C-GalCer, displays a superior inhibitory activity against the liver stages of the rodent malaria parasite P. yoelii. In the present study, we not only confirmed that IL-12 is a key factor that mediates the antiplasmodial activity of both glycolipids, but also defined the important role of NK cells as mediators of α-C-GalCer's superior effect. We found that the Th2 cytokines IL-4 and IL-10 are not involved. Moreover, we demonstrated a differential activity of α-C-GalCer in its stimulation of DCs as compared to α-GalCer, and correlated this with an enhanced proliferative response of Vα14i NKT cells to α-C-GalCer. Finally, we showed that Vβ usage does not influence the response of Vα14i NKT cells to either glycolipid.