Malaria infection has long been associated with diminished T cell responses in vitro and more recently in experimental studies in vivo. Suppression of T cell-proliferative responses during malaria has been attributed to macrophages in a variety of murine and human systems. More recently, however, attention has been directed at the role of dendritic cells in this phenomenon, with several studies suggesting that maturation of dendritic cells is inhibited in vitro by the presence of malaria-infected E. In the studies reported here, we have examined the function of dendritic cells taken directly from infected mice. We found that they express high levels of costimulatory proteins and class II MHC, can activate naive T cells to produce IL-2 as efficiently as dendritic cells from uninfected mice, and support high levels of IFN-γ production by naive T cells through an IL-12-dependent mechanism. Dendritic cells from infected mice also support higher levels of TNF-α production by naive T cells. These same dendritic cells present parasite Ag to a malaria-specific T cell hybridoma, a finding that demonstrates that dendritic cells participate in the generation of Ag-specific immunity during infection. Our findings challenge the contention that dendritic cell function is inhibited by malaria infection.

Immune responses to malaria infection mediate both protection and pathology. The outcome of host infection can be an effective immune response that limits or controls the parasite. Alternatively, the production of large amounts of proinflammatory cytokines, or high ratios of pro- to anti- inflammatory cytokines during blood stage infection is thought to result in the morbid conditions associated with malaria, such as cerebral malaria or severe anemia (1, 2, 3). Control over the quality and quantity of the cytokine response begins at the level of the interaction between dendritic cells (DCs) 3 and naive T cells in the host.

Our understanding of DC function during blood stage malaria is limited and conflicting. There has been a long-standing observation that immune responses, and in particular T cell-proliferative responses, are inhibited during acute infection (4, 5, 6, 7). Most experimental evidence supports the idea that macrophages are the host cell responsible for inhibition of T cell proliferation (8, 9, 10). Two recent reports, however, one using Plasmodium falciparum (11) and the other using Plasmodium yoelii (12), have introduced the idea that DCs may participate in the inhibition or failure of T cell responses during malaria infection. Both studies showed that when DCs are matured in vitro in the presence of infected E, they express lower levels of class II MHC and costimulatory proteins and provide poorer stimulation for T cells than control DCs. By contrast, our results (13) as well as those of Seixas et al. (14) found increased levels of class II MHC and costimulatory proteins on DCs during both P. yoelii infection and interaction with P. chabaudi, although neither study tested the ability of DCs to activate naive T cells.

Here we have directly examined the ability of DCs from infected mice to activate naive T cells using a system that allows us to mimic the initial encounter between these two cell types in vivo. We find that DCs from infected mice express high levels of costimulatory proteins and support IL-2 production as well as DCs from uninfected mice. Furthermore, these cells stimulate markedly higher levels of IFN-γ by responding T cells than their counterparts from uninfected mice, through an IL-12-dependent mechanism. Our results indicate that DCs from malaria-infected mice are not impaired in their ability to activate T cells and do not inhibit T cell responses during malaria infection. DCs do appear to have a central role in stimulating high levels of type 1/proinflammatory cytokines during the early stages of murine malaria infection.

B10.D2 and B10.D2-DO11.10 mice, referred to as TCR-transgenic (Tg), were purchased from The Jackson Laboratory (Bar Harbor, ME) and used after a 2-wk acclimation period. B10.D2-DO11.10 mice express a TCR for OVA on CD4 T cells (15). These mice were purchased from The Jackson Laboratory and bred in our laboratory animal facility by crossing heterozygous males with nontransgenic females. Offspring were determined to be heterozygous if >90% of the CD4+ cells in the peripheral blood were positive when stained with anti-Vβ8.1/8.2. Mice were housed in the Painter Center, Colorado State University (Fort Collins, CO), and all experiments were approved by the Institutional Animal Care and Use Committee.

P. yoelii 17X was used as described in Ref.13 . At day 6 postinfection, when the experiments described here were conducted, the average parasitemia is 13% (n = 39). P. yoelii 17X infection is not lethal for this mouse strain, and resolves by ∼day 25.

Abs used for purification and staining were purchased from BD Biosciences (San Jose, CA): anti-Vβ8.1/8.2 (clone MR5-2); anti-CD4 (clone L3T4); anti-CD11c (clone HL3); anti-CD11b (clone M1/70); anti-CD40 (clone 3/23); anti-CD80 (clone 16–10A1); anti-CD86 (clone GL1); anti-CD3 (clone 145-2C11); anti-CD19 (clone 1D3); and anti-class II MHC (clone 2G9). Anti-TCR clonotype Ab (KJ1-26) was purchased from Caltag Laboratories (Burlingame, CA).

DCs were purified from the pooled spleens of three to five mice by positive selection with anti-CD11c. Initially, we used both directly conjugated anti-CD11c beads (clone N418; Miltenyi Biotec, Auburn, CA) as well as biotinylated anti-CD11c (clone HL3) followed by streptavidin coupled to microbeads (Miltenyi Biotec). Cytokine production was similar with either purification, with slightly higher purity and yield using the indirect method. Therefore, this method was used for all experiments in this report. Purification was conducted according to the manufacturer. Of note, the separation buffer was PBS, 0.5% BSA, 0.2 mM EDTA. Once the purified cell populations were obtained, they were subjected to several washes in medium without EDTA to remove this compound. The CD11c (flowthrough) population was then used for positive selection with biotinylated anti-CD11b and streptavidin-conjugated microbeads. The purity of each population was routinely between 75 and 85% determined by flow cytometry. The CD11b+ population contained 30% class II negative, Gr-1+ cells, regardless of the infection status of the host. These cells likely represent neutrophils (which can be seen in cytospin preparations of these cells). Cells were kept in the separation buffer on ice until they were used and were then placed in culture together with T cells and Ag within 8–10 h of sacrifice.

CD4+ T cells were positively selected from the spleens of TCR-Tg mice. Spleen cells were passed first over Sephadex G-10 (16), and the nonbinding cells used for positive selection with anti-CD4 biotin and streptavidin-conjugated microbeads. Purity was generally >90%.

DC subpopulations (CD11c+b and CD11c+b+) were sorted by FACS as follows. CD11c+ cells were enriched as described above. The cells were then stained with anti-CD3 and anti-CD19 conjugated to PE, anti-CD11c conjugated to biotin, followed by streptavidin-PE-CY5 (BD Biosciences) and CD11b conjugated to FITC. Cells were sorted on a MoFlo (DakoCytomation, Fort Collins, CO) with technical assistance kindly provided by DakoCytomation. CD3- and CD19-positive cells were outgated, and CD11c+b and CD11c+b+ were sorted into separate populations. In both uninfected and infected mice, there were 4 times as many CD11c+b+ cells as CD11c+b cells in the population before sorting. DC and T cells were cultured at a 1:1 ratio (105 of each cell type).

Cell cultures were conducted as described in Ref.13 . Briefly, cells were cultured in round-bottom 96-well plates. DCs (3 × 105) and T cells (1–2 × 105) were used in each well, together with 1 mg/ml OVA. Lower concentrations of OVA resulted in proportionally lower production of each cytokine measured. Titration of DC numbers showed that as the numbers were decreased (to 104 DC with 105 T cells), IL-2, IFN-γ, and TNF-α production decreased proportionally in the presence of both uninfected DC (nDC) and infected DC (iDC) (not shown).

The OVA (Sigma-Aldrich, St. Louis, MO; grade V) was not endotoxin free. The addition of OVA to DC cultures without T cells resulted in TNF-α production (but no other cytokine) that was twice the spontaneous release. By contrast, the deliberate addition of LPS to DC cultures resulted in an 8- to 10-fold increase in TNF-α. The magnitude of both effects (OVA and LPS) was similar in DC from infected and uninfected mice.

When macrophages were added to cell cultures to look for inhibitory activity, 3 × 105 macrophages were added to DCs and T cells. All cultures were conducted in triplicate.

IL-2, IFN-γ, IL-10, and IL-12 p40 and p70 cytokine ELISAs were conducted using reagents from BD Biosciences as described (13). Cytokine bead assays to quantify IL-2, IFN-γ, TNF-α, IL-4, and IL-5 were obtained from BD Biosciences and used as described by the manufacturer. Beads were analyzed on a Coulter Epics XL, using FL2 and FL4 for acquisition. Cytokine quantities obtained by ELISA were similar to those obtained with the bead assay.

Whole spleen cells were isolated from either naive or day 6-infected mice. Cells were plated out into 96-well plates and blocked with anti-CD32 for 30 min at 4°C. To each well were added 2 × 106 cells. After the block, the cells were stained with anti-CD3 and anti-CD19 conjugated to biotin, as well as anti-CD11c-APC and either anti-CD80, CD86, CD40, or an isotype control all conjugated to FITC. Streptavidin APC-Cy7 or Alexa Fluor 350 (Molecular Probes, Eugene, OR) was used to gate out the CD3 and CD19+ lymphocytes. The expression of costimulatory proteins on the CD11c+ population was then determined using a Cyan flow cytometer (DakoCytomation).

CD11c+ cells and T cells were purified from infected mice (day 6). These were cocultured for 5 days. Cells were washed, and recultured with DCs from an additional group of infected mice (also at day 6 postinfection). After 2 days of restimulation, a fusion was conducted with the T cell hybridoma partner BWZ (17). This hybridoma carries the LACZ gene under control of the IL-2 promoter and was used with the hope of obtaining hybridomas with responses that could be measured by LACZ activity. IL-2 production proved to be more consistent, however, and that is what is reported. Hybridomas were screened for expression of CD3 and then for their ability to produce IL-2 in response to parasitized E cultured with uninfected spleen cells. These assays were conducted with 3 × 105 APCs, 5 × 105 hybridomas, and 107 E with a 10% parasitemia. Supernatants were harvested after 24 h and assayed for IL-2.

Anti IL-12 studies were conducted to measure the dependence of IFN-γ production on IL-12. The IL-12-neutralizing Ab was obtained from Caltag Laboratories (clone C17.8). The Ab reacts with the p40 subunit of IL-12. Titration of the Ab showed nearly complete inhibition of IL-12-dependent IFN-γ production at 5 μg/ml and inhibition to a slightly lesser amount occurred at a 10-fold lower dilution (data not shown). Cells were isolated from naive or infected mice as described above and cultured with 5, 0.5, or 0 μg/ml anti-IL-12. The 0-μg/ml group contained 5 μg/ml isotype control.

Student’s t test was used to calculate p values and to compare cytokine levels between treatment groups within an individual experiment. Individual experiments in which cytokines were measured using DCs from uninfected and infected mice generally used pooled spleen cells from three to five mice, resulting in two treatment groups. To analyze cytokine levels between multiple experiments, a mixed effects model was used (18). Using this model, we were able to directly compare the differences in cytokine levels between treatments across multiple experiments and obtain p values for these differences. The value of this approach was that we could measure multiple independent infections. All statistical analyses were made using the SAS system software.

The focus of these studies was to determine how DCs guide T cell responses to malaria. Specifically, we wanted to understand how DCs influence the early phases of the immunity, when the T cell response is initiated. In our previous studies (13), we established that TCR-Tg T cells, when transferred to infected mice, responded to Ag (OVA) with high levels of IFN-γ and low levels of IL-2, reflecting the behavior described for malaria-specific T cells at early stages of infection. Therefore, we chose to use this system to dissect APC function in vitro.

To confirm that DCs are the cells primarily responsible for the activation of naive T cells, we purified CD11c+ DCs from uninfected and infected mice (nDCs and iDCs, respectively). We purified these cells directly through positive magnetic selection based on the expression of surface markers, rather than differential adherence and culture methods. This was done because the latter protocols are known to affect the maturation state and function of both DCs and macrophages (19, 20, 21, 22). Macrophage-enriched populations were positively selected from the CD11c-negative cells using anti-CD11b. We then asked which of these populations was able to support the activation of naive T cells.

As expected, the DC population contained all of the T cell accessory capacity for IL-2 and IFN-γ production in both naive and infected mice (Fig. 1). The CD11b+c cells alone supported small amounts of TNF-α, but no IFN-γ or IL-2 from naive T cells. In addition, highly purified B cells from either uninfected or acutely infected mice had minimal capacity to support IL-2 production by naive T cells (data not shown).

FIGURE 1.

DCs contain the stimulatory capacity for naive T cells in the spleen. DCs were purified by positive selection using anti-CD11c-biotin followed by streptavidin-coupled magnetic beads. The macrophage-enriched population (Mac) was purified from the CD11c population using positive selection with anti-CD11b. Cells from uninfected (UN) and infected (I) mice were cultured with TCR-Tg T cells and 1 mg/ml OVA for 24 or 72 h as indicated before harvesting for cytokine measurement. ∗, The infected group is significantly different from the uninfected group harvested at the same time point.

FIGURE 1.

DCs contain the stimulatory capacity for naive T cells in the spleen. DCs were purified by positive selection using anti-CD11c-biotin followed by streptavidin-coupled magnetic beads. The macrophage-enriched population (Mac) was purified from the CD11c population using positive selection with anti-CD11b. Cells from uninfected (UN) and infected (I) mice were cultured with TCR-Tg T cells and 1 mg/ml OVA for 24 or 72 h as indicated before harvesting for cytokine measurement. ∗, The infected group is significantly different from the uninfected group harvested at the same time point.

Close modal

Nominal T cell activation requires the expression of costimulatory proteins on the surface of DCs and presentation of Ag by MHC. With the exception of our previous study (13), the expression of costimulatory proteins on DCs isolated directly from infected mice has not been reported, although expression of these proteins after 24 h of culture was described by Ocana-Morganer et al. (12). We measured the expression of CD80, CD86, and CD40 on nDCs and iDCs. A higher percentage of DCs from infected mice express CD80 and CD40 (Fig. 2). Similarly significant differences can be seen when using mean fluorescence intensity to assess the level of CD80 expression. Expression of CD86 was low and did not differ between nDC and iDC (not shown). We have also found previously that class II expression is strongly up-regulated on all cell types during infection (13). These changes begin as early as day 3 postinfection (not shown).

FIGURE 2.

A higher percentage of DCs from infected mice express CD80 and CD40 (▪). Values were obtained by gating out lymphocytes using anti-CD3 and anti-CD19 and then gating on CD11c+ cells. The percentage of cells expressing the indicated marker when compared with an isotype control is indicated. ∗, Significant difference between iDCs and nDCs.

FIGURE 2.

A higher percentage of DCs from infected mice express CD80 and CD40 (▪). Values were obtained by gating out lymphocytes using anti-CD3 and anti-CD19 and then gating on CD11c+ cells. The percentage of cells expressing the indicated marker when compared with an isotype control is indicated. ∗, Significant difference between iDCs and nDCs.

Close modal

We then compared the ability of iDCs and nDCs to support production of IL-2, type 1 and proinflammatory cytokines (IFN-γ and TNF-α), and type 2 and anti-inflammatory cytokines (IL-4, IL-5, and IL-10) by naive T cells in an extensive series of experiments. The amount of IL-4 and IL-5 detected was at or below the limits of detection (∼15 pg/ml for each) in all assays regardless of the source of DCs. This was true even when cells were stimulated for 5 days and then restimulated with fresh Ag and APC (whole spleen cells) for 2 additional days (not shown). IL-10 was below the limits of detection in three of six assays. In the others, the presence of DCs from infected mice stimulated low amounts of IL-10 in the presence of Ag, and none without Ag added (data not shown).

iDCs supported high levels of IL-2, IFN-γ, and TNF-α. As shown in Table I, cytokine production in this system was dependent on the addition of T cells and Ag, because production of all cytokines was nominal without these elements (data represent one of two similar experiments). A series of eight experiments was performed in which IL-2 production was measured after 24 and 72 h of culture. iDCs and nDCs did not differ in their ability to support IL-2 production when measured at either time point. By contrast, IFN-γ was consistently higher when iDCs were used as APC (p < 0.05). TNF-α also tended to be higher in the presence of iDCs, but this comparison did not achieve significance (p = 0.08). These data are summarized in Table II. Our findings indicate that iDCs are equally effective at T cell activation as nDCs but that they support a higher level of proinflammatory/type 1 cytokines by T cells.

Table I.

IL-2, IFN-γ, and TNF-α production (all in picograms per milliliter) by naive T cells is dependent on the addition of Ag

T Cells++
OVA+
IL-2    
 nDC 1.9 1.9 532.7 
 iDC 1.9 1.9 448.6 
IFN-γ    
 nDC 2.2 2.0 28.1 
 iDC 5.2 5.2 1697.5a 
TNF-α    
 nDC 4.2 4.5 241.6 
 iDC 10.2 9.5 288.1 
T Cells++
OVA+
IL-2    
 nDC 1.9 1.9 532.7 
 iDC 1.9 1.9 448.6 
IFN-γ    
 nDC 2.2 2.0 28.1 
 iDC 5.2 5.2 1697.5a 
TNF-α    
 nDC 4.2 4.5 241.6 
 iDC 10.2 9.5 288.1 
a

The difference between infected and uninfected DC was different using the t test (p < 0.05).

Table II.

Cytokine production by naive T cells in the presence of dendritic cells derived from infected and uninfected mice

Hour HarvestAverage Uninfected:Infected RatioaN
IL-2 24 1.2 8b 
 72 1.1 
IFN-γ 24 0.4 6c 
 72 0.3 5c 
TNF-α 24 0.5 4d 
 72 0.7 
Hour HarvestAverage Uninfected:Infected RatioaN
IL-2 24 1.2 8b 
 72 1.1 
IFN-γ 24 0.4 6c 
 72 0.3 5c 
TNF-α 24 0.5 4d 
 72 0.7 
a

For each experiment, the mean picograms per milliliter of cytokine from cultures with nDCs was divided by mean picograms per milliliter from cultures with iDCs. This ratio from all experiments was averaged.

b

The number of experiments used to contribute to this number.

c

Statistically significant difference between uninfected and infected, p < 0.05 using mixed effects model.

d

DCs from uninfected mice supported higher TNF-α production after 24 h; p = 0.08.

After 5 days of culture, there were equivalent numbers of viable cells between cultures with nDCs and iDCs. Spleen cells from uninfected mice were added as a source of fresh APCs, and cytokine secretion was measured after 24 of culture with OVA. We found that iDCs and nDCs not only supported equivalent amounts of IL-2 but also retained the pattern of elevated IFN-γ and TNF-α secretion (Fig. 3). These findings, together with our observations that iDCs express high levels costimulatory proteins (Fig. 2) as well as class II MHC (13), provide strong support for the contention that DCs from infected mice are fully functional APCs.

FIGURE 3.

T cells activated by iDCs maintain a consistent cytokine profile upon restimulation. TCR-Tg T cells were cultured for 5 days with nDCs or iDCs, harvested, and washed, and 2 × 104 responder cells were restimulated with OVA (1 mg/ml) using spleen cells from uninfected mice as APC. Cell yield from each primary culture was equivalent. Cytokine quantities are indicated in the number of picograms per milliliter. ∗, iDC supported significantly higher cytokine levels than nDC.

FIGURE 3.

T cells activated by iDCs maintain a consistent cytokine profile upon restimulation. TCR-Tg T cells were cultured for 5 days with nDCs or iDCs, harvested, and washed, and 2 × 104 responder cells were restimulated with OVA (1 mg/ml) using spleen cells from uninfected mice as APC. Cell yield from each primary culture was equivalent. Cytokine quantities are indicated in the number of picograms per milliliter. ∗, iDC supported significantly higher cytokine levels than nDC.

Close modal

To rule out the possibility that a small contaminating population of cells copurified with DCs was influencing cytokine production, we sorted purified CD11c+ cells. CD3+ and CD19+ cells were stained with the same fluorochrome and negatively selected during the process. In addition, CD11cb+ cells were negatively selected to remove macrophages and neutrophils. The CD11c+ population was subdivided into CD11c+b and CD11c+b+ populations (Fig. 4 A). These two subpopulations represent the major subsets of murine DCs; the myeloid (CD11c+b+) and the CD11c+b population that contains the lymphoid (CD8+) and plasmacytoid (Gr1+/B220+) subpopulations.

FIGURE 4.

Sorted DC subpopulations from infected mice support high levels of cytokine production. A, DCs from infected (INF) and uninfected (UN) mice were sorted into CD11c+b and CD11c+b+ subpopulations together with negative selection using anti-CD3 and CD19. Each population was >93% pure. B, DCs sorted as shown were cultured with TCR-Tg T cells (105 T cells and 105 DCs) and OVA. The group labeled all CD11c+ consisted of CD11c+b DC and CD11c+b+ DC recombined in their original ratio (1:4). From this combined population, 105 cells were used to stimulate T cells. Supernatants were assayed for the indicated cytokine (shown in picograms per milliliter) after 24 h. Each comparison between triplicate cultures with nDC (N) and iDC (I) was statistically significant.

FIGURE 4.

Sorted DC subpopulations from infected mice support high levels of cytokine production. A, DCs from infected (INF) and uninfected (UN) mice were sorted into CD11c+b and CD11c+b+ subpopulations together with negative selection using anti-CD3 and CD19. Each population was >93% pure. B, DCs sorted as shown were cultured with TCR-Tg T cells (105 T cells and 105 DCs) and OVA. The group labeled all CD11c+ consisted of CD11c+b DC and CD11c+b+ DC recombined in their original ratio (1:4). From this combined population, 105 cells were used to stimulate T cells. Supernatants were assayed for the indicated cytokine (shown in picograms per milliliter) after 24 h. Each comparison between triplicate cultures with nDC (N) and iDC (I) was statistically significant.

Close modal

When sorted DC were cultured with naive T cells and Ag, both subpopulations (CD11c+b and b+) from infected mice produced substantially higher levels of IL-2, IFN-γ, and TNF-α than the same population from uninfected mice (Fig. 4 B), confirming the behavior observed when DC were purified by positive selection with magnetic beads. To make this comparison directly, CD11c+b and b+ cells were mixed together and cultured with T cells and OVA (in the proportions found in the presort population). The results again confirmed that DCs from infected mice are fully functional in their ability to support cytokine production by responding T cells. A second experiment comparing sorted CD11c+b+ nDC and iDC yielded similar results.

Ocana-Morgener et al. (12) showed that DCs from infected mice were refractory to LPS stimulation of IL-12 production. This result presents a paradox for our finding that DCs (which were purified using a similar protocol) support high levels of IFN-γ (Table II). The major stimulus for IFN-γ production is IL-12, although an IL-12-independent pathway for IFN-γ production has been described for P. berghei (23). Therefore, we measured IL-12 p40 and p70. IL-12 p40 was significantly higher in iDC than in nDC in three of three experiments, one of which is shown in Fig. 5,A. IL-12 p70 was undetectable by ELISA at 24 or 72 h (lower limits of detection, ∼20 pg/ml). We then determined whether IFN-γ production in our system was dependent on IL-12 by adding anti-IL-12 to cultures with iDCs plus naive T cells. The anti p40 Ab effectively abrogated the detection of IL-12 p40 (Fig. 4,A) and reduced the production of IFN-γ to <5% of control. Furthermore, IFN-γ production depended on the presence of Ag (Table I), indicating that contaminating NK cells in the DC preparation could not account for the observed IFN-γ. This finding indicates that biologically relevant concentrations of IL-12 were present even though IL-12 p70 was not detectable.

FIGURE 5.

Anti-IL-12 abrogates IFN-γ production in the presence of iDCs. The indicated concentration of anti-IL-12 was added to cultures of nDCs or iDCs with T cells. IL-12 p40 (A) and IFN-γ (B) were measured after 72 h of culture. IFN-γ production in the presence of iDCs and no anti-IL-12 was significantly higher than IFN-γ production in any other treatment group. In the absence of Ag, IFN-γ production was <15 pg/ml in all samples (not shown).

FIGURE 5.

Anti-IL-12 abrogates IFN-γ production in the presence of iDCs. The indicated concentration of anti-IL-12 was added to cultures of nDCs or iDCs with T cells. IL-12 p40 (A) and IFN-γ (B) were measured after 72 h of culture. IFN-γ production in the presence of iDCs and no anti-IL-12 was significantly higher than IFN-γ production in any other treatment group. In the absence of Ag, IFN-γ production was <15 pg/ml in all samples (not shown).

Close modal

To establish that the DCs used in these experiments are capable of presenting parasite Ag and thus are a population of cells relevant to immune responses to P. yoelii infection, we examined their ability to activate a P. yoelii-specific T cell hybridoma. The hybridoma was generated by immunization with iDCs, and then restimulation in vitro as described in Materials and Methods. Fig. 6 A shows that the hybridoma recognizes parasite Ag, and not an autoantigen present on spleen cells or RBC, because it produced IL-2 only when cultured with parasitized E.

FIGURE 6.

A malaria-specific T cell hybridoma recognizes Ag presented by DCs from infected mice. A, The T cell hybridoma PY-I was cultured with spleen cells from uninfected mice. Uninfected (nRBC) or infected (pRBC) E were added to the cultures, and IL-2 production measured after 24 h. B, PY-I was cultured with unfractionated spleen cells from uninfected (UN) or infected (INF) mice, or with DCs from each group. Supernatants were harvested as above. Stim, Stimulation.

FIGURE 6.

A malaria-specific T cell hybridoma recognizes Ag presented by DCs from infected mice. A, The T cell hybridoma PY-I was cultured with spleen cells from uninfected mice. Uninfected (nRBC) or infected (pRBC) E were added to the cultures, and IL-2 production measured after 24 h. B, PY-I was cultured with unfractionated spleen cells from uninfected (UN) or infected (INF) mice, or with DCs from each group. Supernatants were harvested as above. Stim, Stimulation.

Close modal

We then asked whether iDCs present parasite Ag in vivo. The hybridoma produced small quantities of IL-2 when spleen cells from infected mice were used as APC without the addition of exogenous parasite Ag (Fig. 6 B). When DCs were purified directly from these infected spleens and used as APC, high levels of IL-2 were produced. These findings indicate that the population of DCs used in these studies presents parasite Ag in vivo and thereby directly influences parasite-specific T cell activation.

Our experiments revealed that DCs from infected mice are unlikely players in the inhibition of T cell responses described. In previous work, we found that depletion of CD11b+ cells from the spleens of infected mice abrogated IL-2-inhibitory capacity (13). To directly determine how this subset of cells influences cytokine production, we added CD11b+c cells to cultures with naive TCR-Tg T cells and DCs. This population is enriched for macrophages as assessed by cytologic examination and contains no primary T cell-stimulatory capacity (Fig. 7). The addition of these cells from both uninfected and infected mice tended diminish IL-2 production, a finding consistent with the literature (24). The macrophage-enriched population from infected mice, however, suppressed IL-2 production substantially more than those from uninfected mice. This effect was noted whether the DCs were derived from uninfected (Fig. 6) or infected mice (not shown). IL-2 inhibition was statistically significant in four of four experiments. The effect of these cells on IFN-γ production and TNF-α production was variable, leading to either slight enhancement (two experiments) or no difference (one experiment). The degree of parasitemia was similar in all three experiments, as was the degree of purity of the APC subsets.

FIGURE 7.

The macrophage-enriched population of cells inhibits IL-2, but not IFN-γ or TNF-α production. DCs from uninfected mice were isolated and cultured with T cells, OVA, and CD11b+ cells from infected or uninfected mice as indicated for the 24 or 72 h. ∗, The group was significantly different from the none control; ∗∗, Significant difference between the nMac and iMac groups.

FIGURE 7.

The macrophage-enriched population of cells inhibits IL-2, but not IFN-γ or TNF-α production. DCs from uninfected mice were isolated and cultured with T cells, OVA, and CD11b+ cells from infected or uninfected mice as indicated for the 24 or 72 h. ∗, The group was significantly different from the none control; ∗∗, Significant difference between the nMac and iMac groups.

Close modal

The result confirms our previous study and the findings by others that CD11b+ cells from infected mice inhibit T cell-proliferative responses (13, 25). Furthermore, the inhibition appears to be specific for IL-2 production. The finding that these cells neither inhibited nor enhanced the production of IFN-γ or TNF-α suggests that the DC-T cell axis is primarily responsible for the production of proinflammatory cytokines early in malaria infection.

The function of DCs during malaria infection has not been extensively studied, and those studies that are available offer conflicting results. The experiments described here were performed to gain a better understanding of how DCs control subsequent T cell responses during malaria infection. In particular, we addressed two important questions about the interaction between DCs and naive T cells during infection: 1) whether DCs are responsible for diminished T cell responses frequently described in malaria as suggested by recent reports (11, 12); and 2) how DCs might control the exuberant proinflammatory response documented early in both human and murine malaria (26, 27, 28).

The experimental approach that we chose was to isolate DCs from infected mice and study their function after purification by positive selection, without a period of culture. In this way their function in vitro would most closely reflect their function in vivo. The ability of these DCs to activate a homogeneous population of naive T cells was then assessed to recreate the initial encounter between DCs and naive T cells. We find that DCs taken directly from infected mice, which have matured in vivo rather than in vitro, 1) produce biologically relevant levels of IL-12, 2) support proliferation and IL-2 production by naive T cells to the same extent as DCs from uninfected mice, 3) support substantially higher levels of IFN-γ by responding T cells, and 4) in this and as we demonstrated in previous studies (13) DCs from infected mice express greater levels of class II, CD40, and CD80 than DCs from uninfected mice. To establish that the DC population was relevant to malaria infection, we produced a T cell hybridoma specific for a P. yoelii-derived Ag (the Ag has not yet been identified). DCs purified directly from infected mice stimulated high levels of IL-2 production by the hybridoma, without the addition of additional parasite Ag, indicating that the purified DCs present parasite Ag in vivo.

Two recent reports have suggested that human or murine DC grown in GM-CSF and IL-4, when exposed to malaria-infected E in vitro or in vivo and then matured with LPS, have diminished T cell stimulatory capacity (11, 12). In our studies, DC taken directly from infected mice showed no impairment of their ability to activate T cells. Although it is possible that differences in the responding T cell populations may account for some of the discrepancies between this report and previous ones, we favor the idea that the different signals to which the DC are exposed during development explain the results. When DC are matured in vitro with the use of LPS, they mature without T cell ligation of CD40. A recent study by Straw et al. (29) found that even when DCs are exposed to infectious stimuli provided by a protozoan pathogen (Toxoplasma gondii) in vivo, they will not exhibit a mature phenotype in the absence of T cell signaling through CD40-CD40L interactions. Reis e Sousa and colleagues (20, 30) found that CD40 ligation was required for maximal stimulation of cytokine production by DCs, even in the presence of potent microbial stimuli. Thus, the phenotype and responsiveness of DCs grown in vitro without direct T cell signaling may be different from that of DCs grown in vivo. We hypothesize that when T cells are present during DC development and maturation, the signals they provide to DCs overcome the inhibitory effects of infected E. Other factors that could influence DC maturation, such as Ag uptake through Fc or C receptors, are also not recapitulated in in vitro culture systems. Therefore, we suggest that the function of DCs taken directly from an infected environment more closely resembles their phenotype in vivo.

In previous work, we showed that the spleens of infected mice produce a substance that inhibits IL-2 production without direct cell-cell contact (13). The substance was distinct from NO, PGE2, TGFβ, or IL-10, and removing CD11b+ cells from the spleen cell preparation could eliminate inhibition. By blocking indoleamine 2,3-dioxygenase function in vitro, we were also able to rule out the possibility that indoleamine 2,3-dioxygenase activation in macrophages and subsequent tryptophan catabolism plays an important role in IL-2 suppression (our unpublished results). Here, although we have yet to isolate the inhibitory factor involved, we demonstrate that purified CD11b+CD11c cells are key players in this inhibition of IL-2 production, but have little effect on IFN-γ or TNF-α. These in vitro observations are consistent with our in vivo findings that T cells maturing in an infected environment undergo limited division but produce high levels of IFN-γ (13). Although a large body of earlier literature indicates that macrophages from malaria-infected individuals have immunosuppressive properties, there is a surprising paucity of studies in which macrophages are shown to be directly suppressive to T cell proliferation (25, 31). We cannot exclude the possibility that CD11b+Gr-1+ cells, which are likely to be neutrophils, also participate in the suppression of IL-2 production, because these cells can mediate IL-2 suppression through direct cell contact (32) and the production of soluble factors (33). Ongoing studies in our laboratory are addressing this question.

Our results have implications for the progression of immune responses to malaria pathogens, as well as to other Ags encountered during malaria infection, such as those derived from concurrent viral infections or vaccines. During the acute phase of infection, naive T cells recruited to the immune response may be stimulated to produce IFN-γ and TNF-α. Regulation of these cytokines can be crucial in determining the balance between protection and pathologic responses. Although the T cells in our system were responding to unrelated Ag, other studies have established that the presence of a pathogen influences the nature of the T cell response, even when the Ag is not derived from the pathogen. For example, Chen and Jenkins (34) showed that TCR-Tg T cells specific for OVA (similar to those used here but on a BALB/c background) responded with high IL-2 and IFN-γ to both OVA expressed by Escherichia coli, and an OVA-containing protein injected together with E. coli that did not express OVA. The response was qualitatively and quantitatively different when the OVA-containing protein was given without coinjection of E. coli. Thus, the immune response to a pathogen can mold the response to third-party Ags, and the response to third-party Ags can be used to model pathogen-specific immune responses.

Our finding that IL-2 production was inhibited by macrophages from infected mice may mean that the expansion of Th1 cells early in the immune response is limited, allowing for an outgrowth of Th2 cells. The observation that the cytokine response is dominated by type 1/proinflammatory cytokines early in infection, and type 2 cytokines later is consistent with our results (our unpublished observations with P. yoelii and Refs.35 and36). Although such a well-defined switch has not been documented in human malaria, the acquisition of immunity to disease (but not infection) may be the result of a more gradual shift in the ratio of pro- to anti-inflammatory cytokines by T cells (3). A greater understanding of the mechanisms that mediate this shift in mice may help us better understand similar phenomena in human malaria.

We thank Dr. Joe Smith (Seattle Biomedical Research Institute) for his helpful comments on the manuscript and Angie VanderGaw (DakoCytomation) for her assistance with the cell sorting studies.

1

This work was supported by United States Public Health Service Grant R01AI42354 from the National Institutes of Health.

3

Abbreviations used in this paper: DC, dendritic cell; nDC, uninfected DC; iDC, infected DC; Tg, transgenic.

1
Kurtzhals, J. A., V. Adabayeri, B. Q. Goka, B. D. Akanmori, J. O. Oliver-Commey, F. K. Nkrumah, C. Behr, L. Hviid.
1998
. Low plasma concentrations of interleukin 10 in severe malarial anaemia compared with cerebral and uncomplicated malaria.
Lancet
351
:
1768
.
2
Li, C., I. Corraliza, J. Langhorne.
1999
. A defect in interleukin-10 leads to enhanced malarial disease in Plasmodium chabaudi chabaudi infection in mice.
Infect. Immun.
67
:
4435
.
3
Dodoo, D., F. M. Omer, J. Todd, B. D. Akanmori, K. A. Koram, E. M. Riley.
2002
. Absolute levels and ratios of proinflammatory and anti-inflammatory cytokine production in vitro predict clinical immunity to Plasmodium falciparum malaria.
J. Infect. Dis.
185
:
971
.
4
Weinbaum, F. I., J. Weintraub, F. K. Nkrumah, C. B. Evans, R. E. Tigelaar, Y. J. Rosenberg.
1978
. Immunity to Plasmodium berghei yoelii in mice. II. Specific and nonspecific cellular and humoral responses during the course of infection.
J. Immunol.
121
:
629
.
5
Riley, E. M., C. Maclennan, D. Kwiatkowski, B. M. Greenwood.
1989
. Supression of in-vitro lymphoproliferative responses in acute malaria patients can be partially reversed by indomethacin.
Parasite Immunol.
11
:
509
.
6
Schwarzer, E., F. Turrini, D. Ulliers, G. Giribaldi, H. Ginsburg, P. Arese.
1992
. Impairment of macrophage function after ingestion of Plasmodium falciparum-infected erythrocytes or isolated malarial pigment.
J. Exp. Med.
176
:
1033
.
7
Morakote, N., D. E. Justus.
1988
. Immunosuppression in malaria: effect of hemozoin produced by Plasmodium berghei and Plasmodium falciparum.
Intl. Arch. Allergy Appl. Immunol.
86
:
28
.
8
Wangoo, A., N. K. Ganguly, R. C. Mahajan.
1990
. Immunosuppression in murine malaria: suppressor role of macrophages and their products during acute and chronic Plasmodium berghei infection.
APMIS
98
:
407
.
9
Loose, L. D..
1984
. Characterization of macrophage dysfunction in rodent malaria.
J. Leukocyte Biol.
36
:
703
.
10
Warren, H. S., W. P. Weidanz.
1976
. Malarial immunodepression in vitro: adherent spleen cells are functionally defective as accessory cells in the response to horse erythrocytes.
Eur. J. Immunol.
6
:
816
.
11
Urban, B. C., D. J. Ferguson, A. Pain, N. Willcox, M. Plebanski, J. M. Austyn, D. J. Roberts.
1999
. Plasmodium falciparum-infected erythrocytes modulate the maturation of dendritic cells.
Nature
400
:
73
.
12
Ocana-Morgner, C., M. M. Mota, A. Rodriguez.
2003
. Malaria blood stage suppression of liver stage immunity by dendritic cells.
J. Exp. Med.
197
:
143
.
13
Luyendyk, J. L., O. R. Olivas, L. A. Ginger, A. C. Avery.
2002
. Antigen-presenting cell function during Plasmodium yoelii infection.
Infect. Immun.
70
:
2941
.
14
Seixas, E., C. Cross, S. Quin, J. Langhorne.
2001
. Direct activation of dendritic cells by the malaria parasite, Plasmodium chabaudi chabaudi.
Eur. J. Immunol.
31
:
2970
.
15
Murphy, K. M., A. B. Heimberger, D. Y. Loh.
1990
. Induction by antigen of intrathymic apoptosis of CD4+CD8+TCRlo thymocytes in vivo.
Science
250
:
1720
.
16
Coligan, J. E., A. M. Kruisbeek.
1996
. In R. Coico, ed.
Current Protocols in Immunology
Vol. 1.
:
3.8.11
. John Wiley & Sons, Chichester.
17
Sanderson, S., D. J. Campbell, N. Shastri.
1995
. Identification of a CD4+ T cell-stimulating antigen of pathogenic bacteria by expression cloning.
J. Exp. Med.
182
:
1751
.
18
Ott, R. L., M. Longnecker.
2001
.
Statistical Methods and Data Analysis
Duxbury Publishing Company, Pacific Grove, CA.
19
Vremec, D., K. Shortman.
1997
. Dendritic cell subtypes in mouse lymphoid organs.
J. Immunol.
159
:
565
.
20
Schulz, O., A. D. Edwards, M. Schito, J. Aliberti, S. Manickasingham, A. Sher, C. Reis e Sousa.
2000
. CD40 triggering of heterodimeric IL-12 p70 production by dendritic cells in vivo requires a microbial priming signal.
Immunity
13
:
453
.
21
Reis e Sousa, C., S. Hieny, T. Scharton-Kersten, D. Jankovic, H. Charest, R. N. Germain, A. Sher.
1997
. In vivo microbial stimulation induces rapid CD40 ligand-independent production of interleukin 12 by dendritic cells and their redistribution to T cell areas.
J. Exp. Med.
186
:
1819
.
22
Reis e Sousa, C., G. Yap, O. Schulz, N. Rogers, M. Schito, J. Aliberti, S. Hieny, A. Sher.
1999
. Paralysis of dendritic cell IL-12 production by microbial products prevents infection-induced immunopathology.
Immunity
11
:
637
.
23
Feng, C., S. Watanabe, S. Maruyama, G. Suzuki, M. Sata, T. Furuta, S. Kojima, S. Taki, Y. Asano.
1999
. An alternate pathway for type 1 T cell differentiation.
Int. Immunol.
11
:
1185
.
24
Munn, D. H., E. Shafizadeh, J. T. Attwood, I. Bondarev, A. Pashine, A. L. Mellor.
1999
. Inhibition of T cell proliferation by macrophage tryptophan catabolism.
J. Exp. Med.
189
:
1363
.
25
Scorza, T., S. Magez, L. Brys, P. De Baetselier.
1999
. Hemozoin is a key factor in the induction of malaria-associated immunosuppression.
Parasite Immunol.
21
:
545
.
26
De Souza, J. B., K. H. Williamson, T. Otani, J. H. Playfair.
1997
. Early γ interferon responses in lethal and nonlethal murine blood-stage malaria.
Infect. Immun.
65
:
1593
.
27
Othoro, C., A. A. Lal, B. Nahlen, D. Koech, A. S. Orago, V. Udhayakumar.
1999
. A low interleukin-10 tumor necrosis factor-a ratio is associated with malaria anemia in children residing in a holoendemic malaria region in western Kenya.
J. Infect. Dis.
1999
:
279
.
28
Perkins, D. J., J. B. Weinberg, P. G. Kremsner.
2000
. Reduced interleukin-12 and transforming growth factor-β1 in severe childhood malaria: relationship of cytokine balance with disease severity.
J. Infect. Dis.
182
:
988
.
29
Straw, A. D., A. S. MacDonald, E. Y. Denkers, E. J. Pearce.
2003
. CD154 plays a central role in regulating dendritic cell activation during infections that induce Th1 orTh2 responses.
J. Immunol.
170
:
727
.
30
Edwards, A. D., S. Manickasingham, R. Sporri, S. S. Diebold, O. Schulz, A. Sher, T. Kaisho, S. Akira, C. Reis e Sousa.
2002
. Microbial recognition via toll-like receptor-dependent and -independent pathways determines the cytokine response of murine dendritic cell subsets to CD40 triggerin.
J. Immunol.
169
:
3652
.
31
Correa, M., P. R. Narayanan, H. C. Miller.
1980
. Suppressive activity of splenic adherent cells from Plasmodium chabaudi-infected mice.
J. Immunol.
125
:
749
.
32
Mencacci, A., C. Montagnoli, A. Bacci, E. Cenci, L. Pitzurra, A. Spreca, M. Kopf, A. H. Sharpe, L. Romani.
2002
. CD80+Gr-1+ myeloid cells inhibit development of antifungal Th1 immunity in mice with candidiasis.
J. Immunol.
169
:
3180
.
33
Kusmartsev, S. A., Y. Li, S.-H. Chen.
2000
. Gr-1+ myeloid cells derived from tumor-bearing mice inhibit primary T cell activation induced through CD3/CD28 costimulation.
J. Immunol.
165
:
779
.
34
Chen, Z., M. K. Jenkins.
1998
. Revealing the in vivo behavior of CD4+ T cells specific for an antigen expressed in Escherichia coli.
J. Immunol.
160
:
3462
.
35
Langhorne, J., S. Gillard, B. Simon, S. Slade, K. Eichmann.
1989
. Frequencies of CD4+ T cells reactive with Plasmodium chabaudi chabaudi: distinct response kinetics for cells with Th1 and Th2 characteristics during infection.
Int. Immunol.
1
:
416
.
36
Taylor-Robinson, A. W., R. S. Phillips.
1994
. B cells are required for the switch from Th1- to Th2-regulated immune response to Plasmodium chabaudi chabaudi infection.
Infect. Immun.
62
:
2490
.