Experimental malarial thrombocytopenia can reach life-threatening levels and is believed to be due to Abs targeting platelets for destruction by the reticuloendothelial system. However, we report that Abs account for at most 15% of platelet destruction as Plasmodium berghei-infected B cell-deficient mice exhibited profound thrombocytopenia (83%) as did C57BL/6 controls (98%). Further, no significant increase in Abs bound to intact platelets was observed during infection. P. berghei infection can enhance the activity of anti-platelet Abs as indicated by a significantly (p < 0.005) increased thrombocytopenia on day 4 of infection in mice that were administered a low dose anti-CD41 mAb compared with rat IgG1-injected controls. RAG1−/− and CD4- plus CD8-deficient mice were markedly protected from thrombocytopenia (p < 0.005) and malarial pathogenesis. CD8- or TCRγδ-deficient mice were not protected from thrombocytopenia and CD4-deficient mice were modestly protected. RAG1−/− mice exhibited significantly (p < 0.05) lower levels of plasma TNF, IFN-γ, and IL-12 during infection. IFNγ−/− and IL-12−/− mice exhibited increased survival but similar thrombocytopenia to C57BL/6 controls. Collectively, these data indicate that thrombocytopenia is necessary but not sufficient for malarial pathogenesis and Abs are not the major contributors to malarial thrombocytopenia. Rather, we propose that both CD4+ and CD8+ T cell populations play key roles in malarial thrombocytopenia; a complex bidirectional interaction between cell-mediated immunity and platelets exists during experimental severe malaria that regulates both responses.

Thrombocytopenia is a marked reduction in the number of platelets circulating in the blood that is often elicited by infection with viruses, protozoa, fungi, and bacteria (1). This infectious disease complication increases the propensity for internal as well as dermatologic hemorrhage or purpura. Immune thrombocytopenia purpura is particularly prevalent in children with a peak incidence at age 5 (2). The prevailing view in many of these diseases is that cross-reactive epitopes in the infectious agent elicits anti-platelet Abs, which target the Ab-labeled platelets for destruction by the reticuloendothelial system (3, 4, 5); Abs against several platelet surface proteins are common and include glycoproteins I/IX, Ia/IIa, IIb/IIIa (also termed CD41/CD61 or αIIb/β3), IV, and V (3, 4, 5). Intravenous administration of Ig or anti-IgD are current approaches to ameliorate thrombocytopenia, and these agents are believed to function in part by blocking Fc receptor on the surface of macrophages (6).

Children between the ages of 3 and 5 years are at the greatest risk of developing cerebral malaria, and patients infected with parasites of the genus Plasmodium, the causative agent of malaria, often (>80%) exhibit profound thrombocytopenia (7, 8, 9). Patients presenting with impaired consciousness and obtundation (cerebral malaria) often exhibit retinal hemorrhaging, which is an indicator for a poor prognosis. Petechial hemorrhaging in the brain is often observed upon autopsy in cerebral malaria patients, which relates thrombocytopenia to the genesis of severe malaria (7, 8, 9). Disseminated intravascular coagulation is reported in some patients but not in others (10, 11, 12, 13), suggesting that disseminated intravascular coagulation may function in combination with thrombocytopenia in some cases to mediate hemorrhage. However, the mechanisms for the development of thrombocytopenia in Plasmodium-infected patients are poorly defined. Virtually all malaria patients with thrombocytopenia exhibit platelet-associated Ig that binds to platelet-bound plasmodial Ag (14). The extent of thrombocytopenia correlates with disease severity and IgE levels (15). The most widely accepted mechanism for thrombocytopenia is the development of platelet autoantibodies and subsequent phagocytosis by macrophages (16, 17).

These clinical results provide the rationale for parallel mechanistic studies in animal models. We selected the Plasmodium berghei model of severe malaria because mice infected with this species of Plasmodium exhibit profound thrombocytopenia, impaired consciousness, as well as parasite sequestration (16, 18, 19). Extrapolation of results from this mouse model of severe falciparum malaria to the human condition must be cautious because Plasmodium falciparum does not infect the mouse and there are differences between humans and mice. Nevertheless, the key pathologic features of human malaria (petechial hemorrhage and thrombocytopenia) relevant to this study determining how platelets are consumed during malaria are mimicked in the P. berghei model (16, 19, 20, 21). Several studies have shown that platelets are important in the pathogenesis of experimental severe malaria (ESM)3 (21, 22, 23).

To assess the contributions of Ab vs T cell-dependent Ab-independent mechanisms to malarial thrombocytopenia, we determined whether 1) malarial thrombocytopenia occurs in the absence of B cells and Ab, 2) anti-platelet Abs are elicited by P. berghei infection, 3) experimental malaria exacerbates anti-CD41 mAb-mediated thrombocytopenia, and 4) T cells and selected proinflammatory cytokines are required for malarial thrombocytopenia.

Plasmodium berghei ANKA strain was stored as a frozen stabilate. B cell-deficient (μ-MT−/−), RAG1−/−, CD8−/−, IFNγ−/− (GKO), IL-12β(p40)−/−, BALB/c, and C57BL/6 mice (24, 25, 26, 27) were obtained at 4–5 wk of age from The Jackson Laboratory or bred at La Jolla Bioengineering Institute (La Jolla, CA) from animals obtained from The Jackson Laboratory. All knockout animals were on the C57BL/6 background. Animals were housed in microisolator cages and provided food and water ad libitum. Experimental mice were injected i.p. with 1 × 106P. berghei parasitized erythrocytes when they were between 6 and 10 wk of age. The number of infected erythrocytes among 200 and 1000 erythrocytes in Giemsa-stained thin blood films determined the P. berghei parasitemia. Mice were scored for clinical signs of ESM on day 6, 8, 10, and 12 based on their gripping and righting ability and overall appearance (e.g., breathing and ruffled fur) on a scale of 0 to 5, with 5 reflecting no sign of ESM and a score of 0 was given to mice with no gripping or righting ability. The righting and gripping scores were summed. To kill parasites in the blood, quinine (120 mg/kg) was injected i.p. daily into each mouse. The Institutional Animal Use Committee of La Jolla Bioengineering Institute approved all animal procedures.

Anti-CD41 (αIIb or glycoprotein IIb) mAb (clone MwReg30) was purchased from BD Biosciences. Anti-CD41/CD61 mAb (clone 1B5), provided by Dr. B. Coller (Rockefeller University, New York, NY), is an F(ab′)2 Ab that binds to platelet CD41/CD61 but does not deplete cells through Fc-mediated phagocytosis (28). Rat IgG was purchased from Accurate Chemical and Scientific. Three micrograms of anti-platelet mAb (either clone MwReg30 or clone 1B5) and rat IgG in 100 μl of sterile saline were injected i.p. into individual C57BL/6 mice on day 1 of P. berghei infection.

Anti-CD4 mAb (clone GK1.5), purchased from BD Biosciences, was injected i.p. (0.2 mg) on day 0 of ESM. Anti-TCRγδ (GL3) mAb, provided by Dr. W. Weidanz (University of Wisconsin, Madison, WI), was injected i.p. (1 mg) on day 0 of ESM. Cell depletion was verified by flow cytometry.

Blood (∼400 μl) from the retroorbital plexus from anesthetized mice was collected via heparinized capillary tubes into EDTA collection tubes (BD Biosciences). The blood was centrifuged at 500 × g and the supernatant isolated. The platelet poor plasma (PPP) was aliquoted and stored at −80°C for assessment of cytokines. Plasma for in vitro aggregation assays was used immediately ex vivo. We validated our collection procedure and the levels of fibrinogen were similar to controls. Platelet rich plasma (PRP) was obtained by centrifuging blood at 120 × g and isolating the supernatant. PRP was used within 60 min of isolation. When normalized to the number of platelets per microliter of blood, the PRP contained a similar number of platelets as blood, indicating no coagulation occurred during processing of the blood.

Flow cytometric determination of cell number was performed essentially as previously described (29, 30). For the assessment of platelet number, the tail was pricked with a lancet and 1 μl of blood was immediately transferred to 49 μl of citrate solution (Sigma-Aldrich) to prevent coagulation. Anti-CD41 mAb (clone MwReg30) conjugated with FITC and anti-TER119 mAb (Ly-76; clone Ter119) conjugated with PE-cyanin 5 (PE-Cy5) were added to the blood and incubated together on ice for 15 min. Cells were resuspended in 1 ml of Tyrode’s buffer for flow cytometric analysis.

For verification of cell depletion by mAb treatment or gene knockout, flow cytometric analysis was performed on a single cell suspension of splenocytes as previously described (29, 30). For verification of CD4 or CD8 deficiency, aliquots of splenocytes were fluorescence labeled with anti-CD3 (2C11; biotin-streptavidin-allophycocyanin), anti-CD4 (Rm4.4; FITC), and anti-CD8 (Lyt2; PE-Cy5) mAb; all Abs were purchased from BD Biosciences. For verification of γδ T cell depletion, aliquots of splenocytes were fluorescence labeled with anti-CD3 (2C11; biotin-streptavidin-allophycocyanin), anti-TCRβ (H57; PE-Cy5), and anti-TCRγδ (GL3; FITC).

Cells (splenocytes and platelets) were immediately analyzed on a FACSCalibur (BD Biosciences) and the CellQuest program was used for data acquisition. At least, 5000 cells with size (forward light scatter (FSC)) and granularity (side light scatter (SSC)) of platelets were acquired for subsequent analysis with the Attractors program. For analysis of FSC and SSC of platelets from mice, at least 10,000 CD41+ cells were collected for analysis. At least, 10,000 splenocytes were analyzed for the enumeration of T cell subsets. Spherotech beads, which are easily distinguished from cells based on their unique FSC, SSC, and fluorescence in all three channels, were used to calculate the number of cells per microliter using the formula: [(No. of positive events)/(No. of beads collected) × (bead count/test)/(sample volume in microliters)].

One microliter of blood was obtained from the tail vein and added to 49 μl of citrate buffer. The samples were washed two times with 2 ml of Tyrode’s buffer to remove any circulating Ig that might interfere with the anti-mouse Ig-Alexa Fluor 568 binding to Ab on the surface of platelets. The washed samples were then incubated with anti-CD41-FITC and anti-mouse Ig-Alexa Fluor 568 for 30 min at 4°C. The cells were resuspended in 1 ml of Tyrode’s buffer and immediately analyzed on the FACSCalibur. CD41+ events (20,000) were collected at all time points for analysis.

Plasma (20 μl) from selected source mice were added to 1 μl of PRP from an uninfected C57BL/6 mouse. After a 30 min incubation at room temperature, the number of platelets was assessed in each replicate by flow cytometry as described and compared with the number of platelets in 1 μl of PRP with uninfected plasma added. The reduction in platelet count provides an assessment of the activation potential of the selected plasma.

Three replicates of 25 μl of plasma were analyzed in a modified ELISA using beads (BD Biosciences) for the levels of selected cytokine proteins. The levels of cytokine proteins in plasma were compared with the values of 2-fold serial dilutions of standards starting at 5000 pg/ml to obtain a plasma cytokine concentration. The lower limit of detection for this assay is 20 pg/ml. The cytokines analyzed in this multiplex assay were IFN-γ, TNF-α, IL-6, IL-10, and IL-12p70.

ANOVA with the Statview program (SAS Institute) was performed to statistically compare all measurements with a p value cutoff of 0.05. Survival between different groups was compared with nonparametric log-rank test. The mean and SD of the parasitemia, cytokine levels, and number of platelets per microliter of blood are reported.

C57BL/6 mice infected with P. berghei exhibit marked weight loss, rapid breathing, neurological impairment (such as minimal righting and gripping reflex), obtundation, and then death on about day 6 of infection. BALB/c and IFNγ−/− mice on a C57BL/6 background are resistant to the development of ESM with few mice exhibiting the clinical symptoms of ESM or becoming moribund before day 12 of infection (31, 32). Resistant mice surviving beyond day 12 develop hyperparasitemia and succumb secondarily to anemia 21 days or more after initiation of the infection.

To determine whether polyclonal autoimmune Abs produced during malaria are required for the development of thrombocytopenia during P. berghei infection, groups of five B cell-deficient mice and C57BL/6 controls were injected with 1 × 106 parasitized erythrocytes, and the number of platelets per microliter of blood was assessed by flow cytometry during the course of the infection. The parasitemia in the two groups of animals was similar (Fig. 1,A), indicating that differences in parasite replication did not affect the levels of thrombocytopenia. There was no difference in survival between the two groups of mice with all mice becoming moribund on day 7 of P. berghei infection (Fig. 1,B). The number of platelets per microliter of blood declined markedly (6-fold; p = 0.0002) during the course of P. berghei malaria in B cell-deficient mice declining from 870,974 ± 234,517 platelets/μl on day 0 to 150,108 ± 106,821 platelets/μl on day 6 of infection (Fig. 1,C). The number of platelets per microliter of blood declined markedly in C57BL/6 control mice during the course of P. berghei malaria from 1,086,126 ± 384,928 platelets per microliter on day 0 to 19,118 ± 3722 platelets/μl on day 6 of infection (Fig. 1 C). The extent of thrombocytopenia on day 6 of P. berghei infection was greater in C57BL/6 mice compared with B cell-deficient mice (p = 0.045). Indeed, depletion by experimental malaria of circulating platelets was 98% in C57BL/6 mice and 83% in B cell-deficient mice. This experiment was repeated with similar results: 94% depletion of platelets in C57BL/6 controls and 89% in the B cell-deficient mice infected with P. berghei. Thus, the contribution of anti-platelet Ab to ESM thrombocytopenia is between 5 and 15%.

FIGURE 1.

B cell-deficient mice exhibit marked thrombocytopenia following P. berghei infection. The average P. berghei parasitemia ± SD (A) and cumulative survival (B) in groups of five B cell-deficient mice (▪) and C57BL/6 controls (•) injected i.p. on day 0 with 1 × 106 parasitized erythrocytes. The number of platelets (CD41+ with selected FSC and SSC) (C) and platelet microparticles (CD41+ with lower FSC and SSC) (D) in the two groups of mice were determined during the course of P. berghei infection by flow cytometry.

FIGURE 1.

B cell-deficient mice exhibit marked thrombocytopenia following P. berghei infection. The average P. berghei parasitemia ± SD (A) and cumulative survival (B) in groups of five B cell-deficient mice (▪) and C57BL/6 controls (•) injected i.p. on day 0 with 1 × 106 parasitized erythrocytes. The number of platelets (CD41+ with selected FSC and SSC) (C) and platelet microparticles (CD41+ with lower FSC and SSC) (D) in the two groups of mice were determined during the course of P. berghei infection by flow cytometry.

Close modal

To determine whether there were differences in platelet size in mice with Ab compared with intact controls, we assessed the change in the FSC and SSC of at least 10,000 CD41+ cells (platelets) in the five B cell-deficient mice and C57BL/6 controls. Flow cytometric analysis of CD41+ cells with a selected FSC and SSC (i.e., platelets) in mice infected with P. berghei showed a significant (p < 0.05) increase in the average platelet FSC (crude measure of size) and SSC (crude measure of granularity) in infected B cell-deficient mice (FSC, 25.5 ± 2.8; SSC, 100.8 ± 10.2) compared with uninfected B cell-deficient controls (FSC, 11.7 ± 0.6; SSC, 67.0 ± 2.0). The increase in platelet size was similar to that observed in C57BL/6 mice after infection (FSC, 30.5 ± 1.5; SSC, 111.5 ± 6.6) vs (FSC, 12.8 ± 0.7; SSC, 68 ± 17) in uninfected C57BL/6 mice. In vitro platelet activation assessed by flow cytometry after stimulation with 20 mU thrombin showed a marked decline in platelet number and the formation of platelets with larger FSC than platelets from unstimulated controls. The increase in the physical properties of the platelets during ESM (i.e., FSC and SSC) likely represents platelet activation; platelets from B cell-deficient mice and C57BL/6 controls were activated to a similar extent.

Because platelet particles are proposed to play a major role in malarial pathogenesis (33), we determined whether Ab increases the number of circulating particles during the course of P. berghei malaria. Platelet particles exhibit lower FSC than platelets from uninfected mice while still retaining platelet-specific proteins, such as CD41 (αIIb). The decline in the number of platelet particles detected in B cell-deficient mice (n = 5) and C57BL/6 (n = 5) controls paralleled each other during the course of P. berghei malaria (Fig. 1 D), but there was a modestly greater number of particles in B cell-deficient mice compared with C57BL/6 controls on day 6 of infection. This finding indicates that anti-platelet Ab is not required for platelet particle formation.

To assess whether Ig elicited by infection with P. berghei binds to platelets thereby targeting them for destruction, we fluorescence labeled CD41+ cells/particles (events) with anti-mouse Ig Ab in C57BL/6 mice (n = 5) during the course of P. berghei infection and in uninfected mice (n = 5). On day 3 of P. berghei infection, the percentage of CD41+ events labeled with anti-Ig Ab was significantly increased (p = 0.02) in infected mice when compared with uninfected mice measured at the same time (Fig. 2,A). Although the percentage of CD41+ events that are surface Ig+ increases from 0.2 ± 0.1% on day 0 to 20.9 ± 6.5% on day 5 of infection (Fig. 2,A), the absolute number of CD41+ events that were labeled with anti-Ig was not significantly different (p > 0.5) from controls at any time point due to the thrombocytopenia observed in infected mice (Fig. 2,B). Subdivision of the CD41 events into cells and particles based on their FSC indicated that the CD41+, Ig+ cells (Fig. 2, D and F) exhibited lower FSC than intact platelets from uninfected mice and hence were particles (Fig. 2, C and E). These platelet particles labeled with anti-platelet Ab did not exhibit any significant increase in FSC during the course of infection, whereas CD41+, Ig platelets became significantly (p < 0.001) larger. Few if any particles were detected with lower levels of CD41 on their surface or with lower SSC than conventional platelets. This experiment was repeated with similar results. These findings indicate that anti-platelet Abs are present on the surface of platelet particles but not intact platelets.

FIGURE 2.

Platelet particles rather than platelets are labeled with anti-platelet autoantibody during the course of P. berghei malaria. One microliter of blood was obtained from the tail vein of P. berghei-infected mice (n = 5) and uninfected controls (n = 5) at selected time points after initiation of the infection and was fluorescence labeled with anti-CD41 mAb and anti-mouse Ig. A, The average percentage ± SD of CD41+ events fluorescence labeled with anti-mouse Ig during the course of P. berghei in infected (▪) and uninfected (•) mice. B, The average number ± SD during the course of P. berghei infection of CD41+ events in infected mice (▴), CD41+ Ig+ cells in infected mice (▪) and CD41+ Ig+ cells in uninfected mice (•). CD41+ events were subdivided into CD41+ Ig+ events and CD41+ Ig events in uninfected mice (D) and in infected mice (F) on day 5 of infection. The representative dot plots of FSC and SSC (C and E) of those events are color coded with CD41+ Ig+ events (black dots) and CD41+ Ig events (red dots). Counting beads (blue dots) were used to determine the number of these events. The average FSC of CD41+ Ig+ cells (D) was smaller than intact platelets in uninfected mice, indicating they are particles.

FIGURE 2.

Platelet particles rather than platelets are labeled with anti-platelet autoantibody during the course of P. berghei malaria. One microliter of blood was obtained from the tail vein of P. berghei-infected mice (n = 5) and uninfected controls (n = 5) at selected time points after initiation of the infection and was fluorescence labeled with anti-CD41 mAb and anti-mouse Ig. A, The average percentage ± SD of CD41+ events fluorescence labeled with anti-mouse Ig during the course of P. berghei in infected (▪) and uninfected (•) mice. B, The average number ± SD during the course of P. berghei infection of CD41+ events in infected mice (▴), CD41+ Ig+ cells in infected mice (▪) and CD41+ Ig+ cells in uninfected mice (•). CD41+ events were subdivided into CD41+ Ig+ events and CD41+ Ig events in uninfected mice (D) and in infected mice (F) on day 5 of infection. The representative dot plots of FSC and SSC (C and E) of those events are color coded with CD41+ Ig+ events (black dots) and CD41+ Ig events (red dots). Counting beads (blue dots) were used to determine the number of these events. The average FSC of CD41+ Ig+ cells (D) was smaller than intact platelets in uninfected mice, indicating they are particles.

Close modal

Musaji et al. (34) report that viral infection results in increased thrombocytopenia and increased clearance of platelets through phagocytosis by IFN-γ-activated macrophages if low levels (3 μg) of anti-platelet mAbs are injected upon infection. Injection of this low amount of Ab alone had no effect on platelet numbers in uninfected controls. To assess whether anti-platelet Abs possibly elicited by a previous viral infection exacerbate malarial thrombocytopenia potentially contributing to ESM pathogenesis, we injected a low dose of anti-CD41 mAb (3 μg) i.p. on day 1 of infection and assessed thrombocytopenia during the course of malaria. We selected depleting anti-CD41 mAb (MwReg30) because the CD41/CD61 (αIIb/β3) heterodimer is expressed on the surface of platelets and is often a target for autoantibodies. Upon platelet activation, CD41/CD61 assumes an active conformation, then binds a variety of molecules, including fibrinogen, and is important in clot formation.

Mice injected on day 1 with the anti-CD41 mAb capable of depleting platelets exhibited a significant (p = 0.01) decrease in platelet numbers on day 4 of infection compared with mice injected with a rat IgG1 isotype control. The dose of anti-CD41 mAb was not sufficient to initially deplete platelets because platelet numbers per microliter of blood on day 1 pre-Ig and 2 h post-Ig injection were not significantly different (Fig. 3 A). Injection of higher doses of MwReg30 (100 μg) rapidly depletes platelets within 30 min after i.p. injection (35).

FIGURE 3.

Mice treated on day 1 of infection with a subdepleting dose of anti-CD41 mAb (MWReg30) but not an F(ab′)2 of 1B5 exhibit significantly enhanced platelet depletion on day 4 of P. berghei infection compared with rat IgG controls. A, The average number ± SD of platelets (CD41+ with selected FSC and SSC) in groups of C57BL/6 mice (n = 5) after injection of 3 μg of anti-CD41 (▪) mAb on day 1 of P. berghei infection compared with rat IgG (▴) Ab injected controls. Average number of platelets before (Pre) injection of Ab on day 1 and 2 h after (Post) the injection of Ab is indicated. B, A similar experiment was performed with F(ab′)2 of 1B5 clone (•) of anti-CD41/CD61 mAb.

FIGURE 3.

Mice treated on day 1 of infection with a subdepleting dose of anti-CD41 mAb (MWReg30) but not an F(ab′)2 of 1B5 exhibit significantly enhanced platelet depletion on day 4 of P. berghei infection compared with rat IgG controls. A, The average number ± SD of platelets (CD41+ with selected FSC and SSC) in groups of C57BL/6 mice (n = 5) after injection of 3 μg of anti-CD41 (▪) mAb on day 1 of P. berghei infection compared with rat IgG (▴) Ab injected controls. Average number of platelets before (Pre) injection of Ab on day 1 and 2 h after (Post) the injection of Ab is indicated. B, A similar experiment was performed with F(ab′)2 of 1B5 clone (•) of anti-CD41/CD61 mAb.

Close modal

Anti-CD41/CD61 mAb clone 1B5 is a F(ab′)2 that binds to the surface of platelets but does not result in their depletion (28). Unlike MwReg30, this nondepleting anti-CD41/CD61 (1B5) mAb did not elicit increased thrombocytopenia compared with the isotype control (Fig. 3 B). The experiment with depleting anti-CD41 mAb (MwReg30) was repeated twice with similar results; the 1B5 experiment was repeated once with similar results. Collectively, these results indicate that ESM will accelerate thrombocytopenia if anti-platelet Abs are present through an Fc receptor-dependent mechanism.

Because thrombocytopenia is likely to be mediated through a systemic factor rather than cell-cell contact and to rule out anti-platelet Ab as that factor, we assessed whether plasma from infected B cell-deficient mice and C57BL/6 controls activates platelets in vitro. PPP (20 μl) from either infected or uninfected mice (n = 5) were incubated with PRP from a healthy mouse. We observed a significant increase in the mean FSC (from 9.4 ± 0.4 to 12 ± 2; p < 0.001) and SSC (from 59 ± 3 to 65 ± 3; p = 0.001) of platelets incubated with infected plasma suggesting that a factor produced during infection with P. berghei could mediate platelet activation. A comparable increase in platelet size was observed when platelets were incubated with plasma from B cell-deficient mice (FSC, 12 ± 2; SSC, 71 ± 4; p < 0.001). Further, plasma from P. berghei-infected C57BL/6 mice on day 5 of infection induced a 20% decrease in platelet numbers following incubation with platelets from a healthy mouse compared with plasma from uninfected mice. This experiment was repeated with similar results; however, the magnitude of platelet depletion by the infected plasma varied from experiment to experiment. This result indicates that platelet activation by factors in the plasma of mice with ESM does not require anti-platelet Abs.

To determine whether thrombocytopenia reverses after killing the parasite, we injected i.p. quinine (120 mg/kg) daily into groups of uninfected mice (n = 5), or infected mice (n = 9) beginning on day 5 of ESM. A group of infected mice (n = 7) was injected i.p. with saline vehicle. The parasitemia on day 5 of ESM was similar in both groups of mice before the initiation of quinine therapy (Fig. 4,A). On day 7 onward, the parasitemia was lower in quinine-treated group compared with controls injected with saline (Fig. 4,A). On day 5 of ESM (before the initiation of treatment), both groups of mice exhibited similar degrees of profound thrombocytopenia (Fig. 4 B). The thrombocytopenia remained profound in the saline-injected group and the animals succumbed on day 6 (n = 6) and day 7 (n = 1) of ESM. On day 6 of ESM, the levels of thrombocytopenia were similar in quinine-treated mice and saline-treated controls, but reversed rapidly on day 7 only in the quinine-treated animals and approached uninfected levels by day 8. Only one of the quinine-treated animals succumbed on day 6 of ESM. The number of platelets in blood of uninfected animals treated with quinine remained constant, indicating that quinine at this dose did not have direct effects on thrombocytopenia. Combination therapy of quinine plus trimethoprim sulfamethoxazole to increase the rate of killing of the parasite resulted in toxicity and death of the treated animals. This result indicates that thrombocytopenia reverses once the parasite is killed.

FIGURE 4.

P. berghei-infected mice treated on day 5 of ESM exhibit recovery of platelet numbers. The average P. berghei parasitemia ± SD (A) and platelet numbers ± SD (B) in groups of C57BL/6 mice injected i.p. with 120 mg/kg quinine (▪; n = 9) or with saline vehicle (•; n = 7). A third group was uninfected but injected with same dose of quinine and at the same times as the infected animals (▴; n = 5). Time points of injection of quinine dose (↓) are shown.

FIGURE 4.

P. berghei-infected mice treated on day 5 of ESM exhibit recovery of platelet numbers. The average P. berghei parasitemia ± SD (A) and platelet numbers ± SD (B) in groups of C57BL/6 mice injected i.p. with 120 mg/kg quinine (▪; n = 9) or with saline vehicle (•; n = 7). A third group was uninfected but injected with same dose of quinine and at the same times as the infected animals (▴; n = 5). Time points of injection of quinine dose (↓) are shown.

Close modal

As the lack of B cells and Ab have a modest effect on experimental malarial thrombocytopenia, we determined whether RAG1−/− mice (n = 5) lacking an adaptive immune response were protected from P. berghei thrombocytopenia compared with C57BL/6 controls (n = 5). On days 4 and 6 of P. berghei infection, parasitemia was similar in RAG1−/− mice and C57BL/6 controls, indicating differences in parasite replication did not affect our results (Fig. 5,A). RAG1−/− mice exhibited significantly (p = 0.002) improved survival compared with C57BL/6 controls with five of five RAG1−/− mice surviving beyond day 12 of P. berghei infection (Fig. 5,B). RAG1−/− mice exhibited modest thrombocytopenia; the number of platelets per microliter of blood declined 2-fold (p = 0.03) from 704,179 ± 271,383 on day 0 to 368,041 ± 83,342 on day 6 of infection (Fig. 5 C). The numbers of platelets per microliter of blood were significantly (p < 0.0001) greater in RAG1−/− mice on day 6 of P. berghei malaria than in C57BL/6 controls (20,189 ± 19,075). Thus, depletion by experimental malaria of circulating platelets was 90.3 ± 10.7% in C57BL/6 mice but only 43.7 ± 24.1% in T cell- and B cell-deficient mice.

FIGURE 5.

RAG1-deficient mice exhibit marked protection from thrombocytopenia following P. berghei infection. The average P. berghei parasitemia ± SD (A) and cumulative survival (B) in groups (n = 5) of RAG1-deficient mice (▪) and C57BL/6 controls (•) injected i.p. on day 0 with 1 × 106 parasitized erythrocytes. The number of platelets (CD41+ with selected FSC and SSC) (C) and platelet microparticles (CD41+ with lower FSC and SSC) (D) in the two groups of mice were determined during the course of P. berghei infection by flow cytometry.

FIGURE 5.

RAG1-deficient mice exhibit marked protection from thrombocytopenia following P. berghei infection. The average P. berghei parasitemia ± SD (A) and cumulative survival (B) in groups (n = 5) of RAG1-deficient mice (▪) and C57BL/6 controls (•) injected i.p. on day 0 with 1 × 106 parasitized erythrocytes. The number of platelets (CD41+ with selected FSC and SSC) (C) and platelet microparticles (CD41+ with lower FSC and SSC) (D) in the two groups of mice were determined during the course of P. berghei infection by flow cytometry.

Close modal

The number of platelet particles per microliter of blood declined significantly during the course of P. berghei malaria in both RAG1−/− mice and C57BL/6 controls (Fig. 5 D). Flow cytometric analysis of CD41+ cells revealed that platelets from RAG1−/− mice exhibited significantly lower FSC (26 ± 3) and SSC (107 ± 6) on day 6 of P. berghei infection compared with infected C57BL/6 controls (FSC, 41 ± 2; SSC, 164 ± 17). However, platelets from infected RAG1−/− mice exhibited larger FSC and SSC than platelets from uninfected controls (FSC, 15 ± 0.3; SSC, 83 ± 4), suggesting they were partially activated. These experiments were repeated with similar results.

To define which T cell subset or subsets are required for protection against thrombocytopenia, we assessed the level of thrombocytopenia during the course of ESM in mice lacking selected T cell subsets. Mice treated with anti-CD4 mAb were depleted because few CD3+CD4+ (<0.1%) cells were detected on day 8 of P. berghei infection. An anti-CD4 mAb (Rm4.4) recognizing a distinct epitope from the depleting Ab (GK1.5) was used to ensure we assessed cell depletion rather than Ab masking. As expected, CD8−/− mice exhibited few if any (<0.1%) CD3+CD8+ cells detectable by flow cytometry. Few (<0.1%) if any CD3+ TCRγδ+ T cells were detected by flow cytometry in anti-TCRγδ mAb-injected mice; in addition, few (<0.1%) CD3+ TCRαβ cells were detected, indicating that depletion of γδ T cell population had occurred rather than Ab masking of TCRγδ. Depletion of γδ T cells (data not shown) and CD8+ T cells did not alter the course of thrombocytopenia during ESM (Table I). Anti-CD4 mAb treatment provided some protection from thrombocytopenia particularly on day 4 of ESM, but the combination of CD4+ and CD8+ T cells provided the greatest (p = 0.0002) level of protection. Collectively, these findings indicate that cell mediated immunity comprising both CD4+ and CD8+ T cells function in eliciting malarial thrombocytopenia.

Table I.

Parasitemia and percentage decrease in platelet countsa

Group% ParasitemiaPlatelet No. per Microliter of Blood% Thrombocytopenia
Day 4Day 6Day 0Day 4Day 6Day 4Day 6
Anti-CD4 mAb 8.7 ± 2.1 20.3 ± 6.2 1,830,341 ± 334,083 952,216 ± 309,262 259,007 ± 72,064 52 ± 19b 86 ± 6c 
CD8−/− 11.2 ± 2.2 21.6 ± 2.8 1,673,040 ± 395,512 497,112 ± 138,056 66,267 ± 10,721 67 ± 18 96 ± 1 
Anti-CD4→CD8−/− 6.9 ± 1.7 17.6 ± 2.7 1,574,813 ± 321,182 1,309,523 ± 553,169 417,078 ± 225,058 24 ± 14b 73 ± 15b 
C57BL/6 7.9 ± 3.8 17.0 ± 5.3 1,400,040 ± 207,774 256,428 ± 214,262 99,927 ± 59,028 82 ± 14 92 ± 6 
Group% ParasitemiaPlatelet No. per Microliter of Blood% Thrombocytopenia
Day 4Day 6Day 0Day 4Day 6Day 4Day 6
Anti-CD4 mAb 8.7 ± 2.1 20.3 ± 6.2 1,830,341 ± 334,083 952,216 ± 309,262 259,007 ± 72,064 52 ± 19b 86 ± 6c 
CD8−/− 11.2 ± 2.2 21.6 ± 2.8 1,673,040 ± 395,512 497,112 ± 138,056 66,267 ± 10,721 67 ± 18 96 ± 1 
Anti-CD4→CD8−/− 6.9 ± 1.7 17.6 ± 2.7 1,574,813 ± 321,182 1,309,523 ± 553,169 417,078 ± 225,058 24 ± 14b 73 ± 15b 
C57BL/6 7.9 ± 3.8 17.0 ± 5.3 1,400,040 ± 207,774 256,428 ± 214,262 99,927 ± 59,028 82 ± 14 92 ± 6 
a

Parasitemia and percentage decrease in platelet counts on days 4 and 6 of P. berghei infection in groups (n = 5) of mice lacking selected T cell subsets (column 1) and C57BL/6 controls. Values are mean ± SD.

b

Platelet depletion is significantly less (p < 0.005) than C57BL/6 mice.

c

A value of p = 0.053 for anti-CD4 mAb-injected groups vs C57BL/6 controls on day 6 of ESM.

To determine whether proinflammatory cytokines in the plasma of infected RAG1−/− mice were reduced compared with C57BL/6 controls, we assessed the levels of selected cytokines on day 6 of P. berghei infection in these two groups (n = 5) of mice. IFN-γ, TNF, and IL-10 were assessed because IFN-γ- or TNFR2-deficient mice and IL-10-treated mice are protected from the development of ESM (32, 36, 37). Proinflammatory or type 1 cytokines elicit ESM, whereas type 2 cytokines have no effect or are protective (32, 38). Infected RAG1-deficient mice exhibited significantly reduced levels of proinflammatory IFN-γ, TNF, and IL-12 compared with C57BL/6 controls, but the levels of IL-6 and anti-inflammatory IL-10 were not statistically different (Table II). Both RAG1−/− and C57BL/6 control mice exhibited similar parasitemia on days 4 and 6 of infection, indicating that differences in parasite replication did not account for the results. The RAG1−/− mice exhibited few clinical signs of severe malaria (10 of 10 ± 0), whereas the C57BL/6 mice were clearly ill (4.6 of 10 ± 0.5). This experiment was repeated with similar results. If thrombocytopenia was necessary and sufficient to mediate the genesis of ESM, then the findings that the levels of proinflammatory cytokines IFN-γ, TNF, and IL-12 were significantly lower in RAG1−/− mice and that mice lacking IFN-γ and TNFR2 were protected from ESM suggest that these proinflammatory cytokines may be required to elicit thrombocytopenia during ESM.

Table II.

Levels of selected cytokines in plasmaa

GroupIFN-γTNFIL-6IL-10IL-12
C57BL/6 66 ± 42 63 ± 35 16 ± 7 79 ± 86 25 ± 6 
RAG1−/− 0 ± 0 18 ± 3 10 ± 3 0 ± 0 13 ± 8 
GroupIFN-γTNFIL-6IL-10IL-12
C57BL/6 66 ± 42 63 ± 35 16 ± 7 79 ± 86 25 ± 6 
RAG1−/− 0 ± 0 18 ± 3 10 ± 3 0 ± 0 13 ± 8 
a

Levels of selected cytokines in the plasma of groups (n = 5) of RAG1-deficient mice and C57BL/6 controls on day 6 of P. berghei infection. Average value of cytokine (pg/ml) ± SD is shown. Statistical significance from comparison of infected RAG1−/− mice and C57BL/6 controls was: IFN-γ: p = 0.007; TNF: p = 0.02; IL-6: p = 0.2; IL-10: p = 0.06; IL-12: p = 0.03. In uninfected mice, these cytokines are not detected.

Because BALB/c and IFNγ−/− mice are ESM-resistant strains of mice (31, 32) and RAG1−/− exhibited decreased levels of IFN-γ, we assessed whether the level of thrombocytopenia was similar in these two resistant strains of mice compared with susceptible C57BL/6 controls. Groups of BALB/c and C57BL/6 mice (n = 4) were injected i.p. with 1 × 106 erythrocytes parasitized with P. berghei and the number of platelets per microliter of blood was assessed during the course of the infection. The parasitemiae were similar in ESM-resistant BALB/c and IFNγ−/− mice and ESM-susceptible C57BL/6 controls on days 4 and 6 of infection (Table III), indicating that differences in parasite replication did not affect the thrombocytopenia results. However, the number of platelets per microliter of blood on day 6 of infection was significantly greater in BALB/c mice (157,360 ± 18,042) when compared with C57BL/6 controls (32,600 ± 8074; p = 0.006). As we reported previously (32), IFNγ−/− mice on a C57BL/6 background were significantly (p = 0.005) protected from ESM. The platelet numbers on day 4 of infection were significantly (p = 0.002) lower in ESM-susceptible C57BL/6 mice than in ESM-resistant IFNγ−/−, but the numbers on day 6 of infection were similar (p = 0.6) in both groups of mice. The experiments with ESM-resistant BALB/c and IFNγ−/− mice were repeated with similar results.

Table III.

Parasitemia and percentage decreasea

Group% ParasitemiaPlatelet No. per microliter of blood% Thrombocytopenia
Day 4Day 6Day 0Day 4Day 6Day 4Day 6
BALB/c 2.3 ± 1.0 15.6 ± 4.4 1,141,961 ± 357,327 1,083,449 ± 365,219 183,948 ± 61,470 1 ± 31 83 ± 6b 
C57BL/6 3.1 ± 0.6 12.3 ± 4 858,775 ± 178,610 627,685 ± 207,918 32,600 ± 8,074 24 ± 30 96 ± 1b 
IFN-γ−/− 2.5 ± 11.7 14.5 ± 13.8 938,982 ± 438,288 918,798 ± 319,103 57,351 ± 37,353 0c 91 ± 11 
C57BL/6 3.4 ± 1.2 18.7 ± 3.8 1,483,208 ± 263,129 481,844 ± 278,677 69,424 ± 21,867 65 ± 25c 95 ± 2 
IL-12−/− 2.9 ± 1.0 18.9 ± 5.4 1,008,159 ± 309,032 1,039,573 ± 641,014 39,471 ± 26,955 96 ± 3 
C57BL/6 2.8 ± 0.9 15.1 ± 2.3 1,182,998 ± 532,813 1,053,548 ± 264,494 35,936 ± 15,025 96 ± 2 
Group% ParasitemiaPlatelet No. per microliter of blood% Thrombocytopenia
Day 4Day 6Day 0Day 4Day 6Day 4Day 6
BALB/c 2.3 ± 1.0 15.6 ± 4.4 1,141,961 ± 357,327 1,083,449 ± 365,219 183,948 ± 61,470 1 ± 31 83 ± 6b 
C57BL/6 3.1 ± 0.6 12.3 ± 4 858,775 ± 178,610 627,685 ± 207,918 32,600 ± 8,074 24 ± 30 96 ± 1b 
IFN-γ−/− 2.5 ± 11.7 14.5 ± 13.8 938,982 ± 438,288 918,798 ± 319,103 57,351 ± 37,353 0c 91 ± 11 
C57BL/6 3.4 ± 1.2 18.7 ± 3.8 1,483,208 ± 263,129 481,844 ± 278,677 69,424 ± 21,867 65 ± 25c 95 ± 2 
IL-12−/− 2.9 ± 1.0 18.9 ± 5.4 1,008,159 ± 309,032 1,039,573 ± 641,014 39,471 ± 26,955 96 ± 3 
C57BL/6 2.8 ± 0.9 15.1 ± 2.3 1,182,998 ± 532,813 1,053,548 ± 264,494 35,936 ± 15,025 96 ± 2 
a

Parasitemia and percentage decrease in platelet counts on days 4 and 6 of P. berghei infection in BALB/c (n = 4), IFN-γ−/− (n = 5), IL-12−/− (n = 5) mice and C57BL/6 controls. Values are mean ± SD.

b

Platelet depletion in BALB/c mice on day 6 of infection is significantly less than C57BL/6 mice p = 0.006.

c

Platelet depletion in IFNγ−/− on day-4 of infection is significantly less than control mice p = 0.02

Because the RAG1-deficient mice exhibited lower levels of IL-12, and IL-12 is a proinflammatory cytokine that is often required to elicit production of IFN-γ, it is possible that the absence of this cytokine has a greater effect than the absence of IFN-γ alone. To determine whether 1) IL-12-deficient mice are protected from P. berghei infection or 2) IL-12 is critical for development of malarial thrombocytopenia, we infected groups (n = 5) of IL-12−/− mice and ESM-susceptible C57BL/6 controls with P. berghei. IL-12−/− mice were significantly protected from ESM (p = 0.003) when compared with C57BL/6 controls with one in five IL-12−/− mice surviving beyond day 12 of infection and no C57BL/6 mice (n = 5) surviving. The parasitemia in both groups of mice was similar on days 4 and 6 of infection, indicating that protection cannot be attributed to differences in parasite replication. Despite the increased survival of the IL-12−/− mice, the extent of thrombocytopenia on days 4 and 6 of infection was similar in both groups of mice (Table III). The experiment assessing survival and thrombocytopenia in IL-12−/− mice was repeated and in this experiment we observed a modest and similar level of thrombocytopenia in both IL-12−/− mice and C57BL/6 controls on day 4 of ESM and profound thrombocytopenia in both groups of mice on day 6. The survival experiment alone was repeated two more times with similar results. Collectively, these findings indicate that 1) the proinflammatory cytokines IL-12 and IFN-γ contribute to ESM pathogenesis but are not required for the development of malarial thrombocytopenia and 2) thrombocytopenia alone is not sufficient to mediate disease.

Assessment of thrombocytopenia in ESM-susceptible C57BL/6 mice infected with P. berghei clearly indicates that there is a marked removal of platelets from the blood during ESM and that the thrombocytopenia reaches life-threatening levels, with blood platelet numbers declining between 100- and 1000-fold. Thrombocytopenia of this magnitude results in impaired clot formation and hence a predisposition to hemorrhage. Indeed, mice with ESM exhibit significantly (p < 0.05) greater clotting times (data not shown) and hemorrhaging into vital organs, such as the brain and lung. The mechanisms mediating thrombocytopenia during ESM are poorly defined but may be due to 1) decreased production, 2) increased sequestration, or 3) increased destruction of platelets. There is no apparent defect in platelet production accounting for P. berghei malarial thrombocytopenia because the number of megakaryocytes in bone marrow actually increases at the same time as consumption of radiolabeled platelets is markedly increased (16).

Increased platelet sequestration is a potential mechanism for the removal of platelets from the blood during ESM because firm adherence of platelets to the microvasculature via ICAM-1 and P-selectin has been reported (23, 39, 40). However, it is unlikely that platelet adhesion to an inflamed microvasculature is the major contributor to malarial thrombocytopenia because 1) the level of thrombocytopenia is not altered in infected P-selectin-deficient mice that lack platelet adhesion (23, 41, 42) and 2) platelets adhere to the microvasculature primarily on day 6 of infection whereas marked thrombocytopenia is already observed on day 4 or day 5 of infection. In addition, thrombocytopenia is not ameliorated in IFNγ−/− mice, and increases in ICAM-1 expression during ESM requires IFN-γ (43). The increase in platelet number after initiation of drug therapy on day 5 of ESM is consistent with sequestered platelets being released from the endothelium. However, the increase in number may also reflect the abatement of platelet destruction and increased platelet production levels elicited by ESM (16). Thus, destruction of platelets appears to be the primary mechanism mediating thrombocytopenia. Platelet destruction during ESM could occur via anti-platelet Abs and phagocytosis by macrophages, apoptosis, and platelet Ab independent phagocytosis in spleen.

As proposed for viral infections, and immune thrombocytopenia purpura (2, 14, 34), the mechanism for platelet depletion during malaria is believed to be the production of anti-platelet Abs that bind to platelets and then targets them for destruction by activated macrophages (16). However, our results indicate that thrombocytopenia in B cell-deficient mice infected with P. berghei is marked; the difference in thrombocytopenia between B celldeficient mice and B cell-intact controls on day 6 of P. berghei infection was 98 vs 83%, respectively, suggesting that Ab-mediated platelet depletion accounts for at the most 5–15% of the malarial thrombocytopenia.

Platelets are reportedly coated with Ab during malaria and that the number of platelets with platelet-bound Ab correlated inversely with the platelet count (16). In contrast, our flow cytometry results indicate that there is no significant increase in the number of Ab-coated CD41+ events during the course of P. berghei malaria. Further, these CD41+ events represent particles rather than intact platelets as they exhibit smaller FSC (size) than platelets. Few CD41+ events are detected with SSC much lower than platelets as observed by Piguet et al. (44, 45, 46); this may be due to the processing of platelets to PRP by these researchers, whereas we analyze blood directly.

Our in vitro results with PPP from infected mice indicates that this PPP activates platelets from uninfected mice to increase their FSC and SSC; the number of platelets declines (∼20%) when PPP from infected mice is added compared with PPP from uninfected mice. There was no difference in the platelet response to PPP from B cell-deficient mice and infected C57BL/6 controls, indicating that Ab is not responsible for this in vitro activation of platelets. We are assuming that activation is a step in the process leading to platelet destruction. The marked thrombocytopenia in B cell-deficient mice infected with P. berghei, the lack of detection of Ab bound to platelets during ESM, and the similar effects of PPP from B cell-deficient mice and from C57BL/6 controls on in vitro platelet activation collectively indicate that anti-platelet Ab plays a minimal role in malarial thrombocytopenia. Although the magnitude of thrombocytopenia is markedly different, these results agree with those of Musaji et al. (34), who observe that acute viral thrombocytopenia occurs in the absence of anti-platelet Ab.

Chronic viral or other infections can ultimately elicit anti-platelet Abs whose platelet depleting activity can then be enhanced by T cell-independent, macrophage-dependent phagocytosis resulting from subsequent viral infection; repeated viral infections has therefore been proposed to mediate childhood immune thrombocytopenia purpura (34). Similarly, we observe that P. berghei infection will exacerbate the effects of anti-CD41 mAb Abs if present. The increased platelet depletion by anti-CD41 mAb during ESM is mediated by the Fc receptor because an F(ab′)2 of an anti-CD41 mAb does not elicit increased thrombocytopenia. Experimental malaria is a much stronger stimulus for thrombocytopenia than viral infection because ESM depletes virtually all platelets (98%) from the circulation within 6 days of infection, whereas lactate dehydrogenase-elevating virus and mouse hepatitis virus result in ∼50% loss of platelets (34). By day 6 of ESM, there is no difference between thrombocytopenia in mice treated with depleting anti-platelet clone or with rat IgG, nor did the mice die more rapidly.

One slight internal discrepancy with our data might be: why is malarial thrombocytopenia in B cell-deficient mice ameliorated compared with C57BL/6 controls when no anti-platelet Ab is detected on platelets circulating during P. berghei malaria? One possible explanation is that the cell-mediated immune response during malaria is altered in the absence of B cells (47, 48). Indeed, the marked amelioration of thrombocytopenia in RAG1−/− mice indicates that T cell-mediated immune responses are important in malarial thrombocytopenia. The fact that RAG1−/− mice lack both T and B cells and are resistant to the development of ESM thrombocytopenia suggests that both cell types may collaborate in mediating the destruction of platelets.

There are several possible mechanisms whereby the cell-mediated immune response might be responsible for the destruction of platelets during P. berghei malaria. The first is that factors produced through the cell-mediated immune response induce platelets to undergo apoptosis and form platelet microparticles, as proposed by Piguet et al. (46). They proposed that signaling through CD40L on activated platelets initiates the caspase cascade, resulting in apoptosis of the platelet and the formation of platelet microparticles (46). Indeed, this group reports that caspases are activated in platelets during ESM, and inhibition of the caspases protects against microparticle formation (46, 49). Signaling by TNF does not initiate apoptosis during P. berghei infection because malarial thrombocytopenia and formation of microparticles is similar in TNFR1- and TNFR2-deficient mice (45, 50).

In contrast to the Piguet and colleagues (46, 49) results, we detect no increases in the number of particles in RAG1−/− and C57BL/6 mice by flow cytometry in whole blood during the course of P. berghei malaria. Protection from thrombocytopenia in CD40−/− and CD40L−/− (CD154) mice (44) is mediated by similar mechanism to RAG1-deficient mice, namely, the absence of a cell-mediated immune response rather than the induction of apoptosis in platelets. Our depletion studies indicate that removal of CD8+ T cells or γδ T cells alone provides minimal protection against thrombocytopenia, and depletion of CD4+ T cells provides some protection (∼15%). CD4+ plus CD8+ T cell subset deficiency does markedly protect against thrombocytopenia (∼30%). It is well recognized that the immune response compensates for the lack of a T cell subset during malaria (48); thus, in the single T cell subset depletion studies, CD8+ T cells may compensate for the lack of CD4+ T cells and vice versa.

We have reported that depletion of platelets early during P. berghei infection leads to an altered immune response with significantly decreased levels of proinflammatory cytokines, such as IFN-γ and IL-2, as well as increased levels of IL-10, indicating a role for platelets in the regulation of pathogenic cytokines and cell-mediated immune responses (35). The current results with RAG1−/− mice exhibiting ameliorated malarial thrombocytopenia indicate that the interaction is bidirectional between platelets and the cell-mediated immune response. If thrombocytopenia is sufficient to mediate the genesis of ESM, the decreased levels of proinflammatory TNF, IFN-γ, and IL-12 in the RAG1−/− mice together with the protection against ESM in TNFR2−/−, IFNγ−/−, and IL-12−/− mice would suggest that IL-12 and IFN-γ might be important cytokines in the development of malarial thrombocytopenia. Because mice protected from the development of ESM (IL-12−/− and IFNγ−/− mice) still exhibit profound thrombocytopenia on day 6 of ESM when the C57BL/6 controls are moribund, these findings suggest that either thrombocytopenia is a symptom of ESM but is not essential to its pathogenesis or severe thrombocytopenia is necessary but not sufficient to mediate ESM. However, because profound thrombocytopenia occurs in all moribund mice, and moribund mice exhibit increased clotting times, vascular leak and hemorrhaging into the brain and lung, we propose that thrombocytopenia is necessary for the development of ESM. Hemorrhage often occurs due to thrombocytopenia, linking thrombocytopenia with an important mechanism of pathogenesis. In addition, platelet depletion protected mice and altered the pathogenic immune response, indicating that platelet activation and thrombocytopenia are unlikely to be merely symptoms of ESM.

In conjunction with thrombocytopenia, other factors such as disruption of the vascular wall integrity by activated immune cells and up-regulation of endothelial cell adhesion molecules by proinflammatory cytokines and chemokines may be necessary second signals to cause hemorrhage. Thus, analogous to the two-signal hypothesis for activation of T cells (51, 52), activation of two or more pathogenic mechanisms is required for lethal malaria and thrombocytopenia is likely one of the critical pathogenic mechanisms. This study indicates that the mechanism underlying malaria thrombocytopenia does not involve Abs but rather requires a cell-mediated immune response. Our results indicate that Ab-mediated mechanism of thrombocytopenia in falciparum patients suggested by the earlier studies by Kelton et al. (14) should be revisited to ensure that intact platelets and not platelet particles are indeed labeled with Ab in falciparum patients and that these Abs function to remove platelets from blood. The precise mechanism whereby cell-mediated immunity mediates thrombocytopenia is independent of IL-12 and IFN-γ. Further, these results strongly support a bidirectional link between the coagulation system and cell-mediated immunity whereby platelets alter the proinflammatory cytokine response but their activation can in turn be affected by it.

We thank Dr. Barry Coller for helpful discussions of the data and review of the manuscript.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by National Institutes of Health Grants EB003824 (to J.N.), HL40696 (to J.A.F.), R24 HL643952 (to M.I.), and AI40667 (to H.C.H.).

3

Abbreviations used in this paper: ESM, experimental severe malaria; PPP, platelet poor plasma; PRP, platelet rich plasma; FSC, forward light scatter; SSC, side light scatter.

1
Levi, M., T. T. Keller, E. van Gorp, H. Ten Cate.
2003
. Infection and inflammation and the coagulation system.
Cardiovasc. Res.
60
:
26
-39.
2
Cines, D. B., V. S. Blanchette.
2002
. Immune thrombocytopenic purpura.
N. Engl. J. Med.
346
:
995
-1008.
3
Kunicki, T. J., K. Furihata, D. Nugent.
1988
. Glycoprotein IIb-IIIa as an immunologic target.
Curr. Stud. Hematol. Blood Transfus.
44
-63.
4
Kunicki, T. J., P. J. Newman.
1992
. The molecular immunology of human platelet proteins.
Blood
80
:
1386
-1404.
5
He, R., D. M. Reid, C. E. Jones, N. R. Shulman.
1994
. Spectrum of Ig classes, specificities, and titers of serum antiglycoproteins in chronic idiopathic thrombocytopenic purpura.
Blood
83
:
1024
-1032.
6
Semple, J. W..
2002
. Immune pathophysiology of autoimmune thrombocytopenic purpura.
Blood Rev.
16
:
9
-12.
7
Marsh, K., M. English, J. Crawley, N. Peshu.
1996
. The pathogenesis of severe malaria in African children.
Ann. Trop. Med. Parasitol.
90
:
395
-402.
8
Marsh, K., D. Forster, C. Waruiru, I. Mwangi, M. Winstanley, V. Marsh, C. Newton, P. Winstanley, P. Warn, N. Peshu.
1995
. Indicators of life-threatening malaria in African children.
N. Engl. J. Med.
332
:
1399
-1404.
9
Krogstad, D. J..
1999
. Malaria. R. L. Guerrant, and D. H. Walker, and P. F. Weller, eds.
Tropical Infectious Diseases: Principles, Pathogens, and Practice
736
Churchill Livingstone, Philadelphia.
10
Punyagupta, S., T. Srichaikul, P. Nitiyanant, B. Petchclai.
1974
. Acute pulmonary insufficiency in falciparum malaria: summary of 12 cases with evidence of disseminated intravascular coagulation.
Am. J. Trop. Med. Hyg.
23
:
551
-559.
11
Srichaikul, T., P. Puwasatien, J. Karnjanajetanee, V. A. Bokisch, P. Pawasatien.
1975
. Complement changes and disseminated intravascular coagulation in Plasmodium falciparum malaria.
Lancet
1
:
770
-772.
12
Horstmann, R. D., M. Dietrich.
1985
. Haemostatic alterations in malaria correlate to parasitaemia.
Blut
51
:
329
-335.
13
Mohanty, D., N. Marwaha, K. Ghosh, A. P. Chauhan, S. Shah, S. Sharma, K. C. Das.
1985
. Vascular occlusion and disseminated intravascular coagulation in falciparum malaria.
Br. Med. J.
290
:
115
-116.
14
Kelton, J. G., J. Keystone, J. Moore, G. Denomme, E. Tozman, M. Glynn, P. B. Neame, J. Gauldie, J. Jensen.
1983
. Immune-mediated thrombocytopenia of malaria.
J. Clin. Invest.
71
:
832
-836.
15
Seka-Seka, J., Y. Brouh, A. C. Yapo-Crezoit, N. H. Atseye.
2004
. The role of serum immunoglobulin E in the pathogenesis of Plasmodium falciparum malaria in Ivorian children.
Scand. J. Immunol.
59
:
228
-230.
16
Grau, G. E., P. F. Piguet, D. Gretener, C. Vesin, P. H. Lambert.
1988
. Immunopathology of thrombocytopenia in experimental malaria.
Immunology
65
:
501
-506.
17
Sørensen, P. G., H. Mickley, K. G. Schmidt.
1984
. Malaria-induced immune thrombocytopenia.
Vox Sang.
47
:
68
-72.
18
Hearn, J., N. Rayment, D. N. Landon, D. R. Katz, J. B. de Souza.
2000
. Immunopathology of cerebral malaria: morphological evidence of parasite sequestration in murine brain microvasculature.
Infect. Immun.
68
:
5364
-5376.
19
Mackey, L. J., A. Hochmann, C. H. June, C. E. Contreras, P. H. Lambert.
1980
. Immunopathological aspects of Plasmodium berghei infection in five strains of mice. II. Immunopathology of cerebral and other tissue lesions during the infection.
Clin. Exp. Immunol.
42
:
412
-420.
20
Rest, J. R..
1982
. Cerebral malaria in inbred mice. I. A new model and its pathology.
Trans. R. Soc. Trop. Med. Hyg.
76
:
410
-415.
21
Männel, D. N., G. E. Grau.
1997
. Role of platelet adhesion in homeostasis and immunopathology.
Mol. Pathol.
50
:
175
-185.
22
Grau, G. E., F. Tacchini-Cottier, C. Vesin, G. Milon, J. N. Lou, P. F. Piguet, P. Juillard.
1993
. TNF-induced microvascular pathology: active role for platelets and importance of the LFA-1/ICAM-1 interaction.
Eur. Cytokine Network
4
:
415
-419.
23
Sun, G., W.-L. Chang, J. Li, S. M. Berney, D. Kimpel, H. C. van der Heyde.
2003
. Inhibition of platelet adherence to brain microvasculature protects against severe Plasmodium berghei malaria.
Infect. Immun.
71
:
6553
-6561.
24
Kitamura, D., J. Roes, R. Kuhn, K. Rajewsky.
1991
. A B cell-deficient mouse by targeted disruption of the membrane exon of the immunoglobulin μ chain gene.
Nature
350
:
423
-426.
25
Mombaerts, P., J. Iacomini, R. S. Johnson, K. Herrup, S. Tonegawa, V. E. Papaioannou.
1992
. RAG-1-deficient mice have no mature B and T lymphocytes.
Cell
68
:
869
-877.
26
Dalton, D. K., S. Pitts-Meek, S. Keshav, I. S. Figari, A. Bradley, T. A. Stewart.
1993
. Multiple defects of immune cell function in mice with disrupted interferon-γ genes.
Science
259
:
1739
-1742.
27
Mattner, F., J. Magram, J. Ferrante, P. Launois, K. Di Padova, R. Behin, M. K. Gately, J. A. Louis, G. Alber.
1996
. Genetically resistant mice lacking interleukin-12 are susceptible to infection with Leishmania major and mount a polarized Th2 cell response.
Eur. J. Immunol.
26
:
1553
-1559.
28
Smyth, S. S., E. D. Reis, H. Vaananen, W. Zhang, B. S. Coller.
2001
. Variable protection of β3-integrin-deficient mice from thrombosis initiated by different mechanisms.
Blood
98
:
1055
-1062.
29
Chang, W. L., H. C. van der Heyde, B. S. Klein.
1998
. Flow cytometric quantitation of yeast a novel technique for use in animal model work and in vitro immunologic assays.
J. Immunol. Methods
211
:
51
-63.
30
van der Heyde, H. C., D. D. Manning, W. P. Weidanz.
1993
. Role of CD4+ T cells in the expansion of the CD4, CD8 γδ T cell subset in the spleens of mice during blood-stage malaria.
J. Immunol.
151
:
6311
-6317.
31
Chang, W. L., S. P. Jones, D. J. Lefer, T. Welbourne, G. Sun, L. Yin, H. Suzuki, J. Huang, D. N. Granger, H. C. van der Heyde.
2001
. CD8+ T-cell depletion ameliorates circulatory shock in Plasmodium berghei-infected mice.
Infect. Immun.
69
:
7341
-7348.
32
Yañez, D. M., D. D. Manning, A. J. Cooley, W. P. Weidanz, H. C. van der Heyde.
1996
. Participation of lymphocyte subpopulations in the pathogenesis of experimental murine cerebral malaria.
J. Immunol.
157
:
1620
-1624.
33
Combes, V., N. Coltel, M. Alibert, M. van Eck, C. Raymond, I. Juhan-Vague, G. E. Grau, G. Chimini.
2005
. ABCA1 gene deletion protects against cerebral malaria: potential pathogenic role of microparticles in neuropathology.
Am. J Pathol.
166
:
295
-302.
34
Musaji, A., F. Cormont, G. Thirion, C. L. Cambiaso, J. P. Coutelier.
2004
. Exacerbation of autoantibody-mediated thrombocytopenic purpura by infection with mouse viruses.
Blood
104
:
2102
-2106.
35
van der Heyde, H. C., I. Gramaglia, G. Sun, C. Woods.
2005
. Platelet depletion by anti-CD41 (αIIb) mAb injection early but not late in the course of disease protects against Plasmodium berghei pathogenesis by altering the levels of pathogenic cytokines.
Blood
105
:
1956
-1963.
36
Lucas, R., P. Juillard, E. Decoster, M. Redard, D. Burger, Y. Donati, C. Giroud, C. Monso-Hinard, T. De Kesel, W. A. Buurman, et al
1997
. Crucial role of tumor necrosis factor (TNF) receptor 2 and membrane-bound TNF in experimental cerebral malaria.
Eur. J. Immunol.
27
:
1719
-1725.
37
Kossodo, S., C. Monso, P. Juillard, T. Velu, M. Goldman, G. E. Grau.
1997
. Interleukin-10 modulates susceptibility in experimental cerebral malaria.
Immunology
91
:
536
-540.
38
Jennings, V. M., A. A. Lal, R. L. Hunter.
1998
. Evidence for multiple pathologic and protective mechanisms of murine cerebral malaria.
Infect. Immun.
66
:
5972
-5979.
39
Favre, N., C. Da Laperousaz, B. Ryffel, N. A. Weiss, B. A. Imhof, W. Rudin, R. Lucas, P. F. Piguet.
1999
. Role of ICAM-1 (CD54) in the development of murine cerebral malaria.
Microbes Infect.
1
:
961
-968.
40
Combes, V., A. R. Rosenkranz, M. Redard, G. Pizzolato, H. Lepidi, D. Vestweber, T. N. Mayadas, G. E. Grau.
2004
. Pathogenic role of P-selectin in experimental cerebral malaria: importance of the endothelial compartment.
Am. J. Pathol.
164
:
781
-786.
41
Chang, W. L., J. Li, G. Sun, H. L. Chen, R. D. Specian, S. M. Berney, D. N. Granger, H. C. van der Heyde.
2003
. P-selectin contributes to severe experimental malaria but is not required for leukocyte adhesion to brain microvasculature.
Infect. Immun.
71
:
1911
-1918.
42
Li, J., W. L. Chang, G. Sun, H. L. Chen, R. D. Specian, S. M. Berney, D. Kimpel, D. N. Granger, H. C. van der Heyde.
2003
. Intercellular adhesion molecule 1 is important for the development of severe experimental malaria but is not required for leukocyte adhesion in the brain.
J. Investig. Med.
51
:
128
-140.
43
Bauer, P. R., H. C. van der Heyde, G. Sun, R. D. Specian, D. N. Granger.
2002
. Regulation of endothelial cell adhesion molecule expression in an experimental model of cerebral malaria.
Microcirculation
9
:
463
-470.
44
Piguet, P. F., C. D. Kan, C. Vesin, A. Rochat, Y. Donati, C. Barazzone.
2001
. Role of CD40-CVD40L in mouse severe malaria.
Am. J. Pathol.
159
:
733
-742.
45
Piguet, P. F., C. D. Kan, C. Vesin.
2002
. Role of the tumor necrosis factor receptor 2 (TNFR2) in cerebral malaria in mice.
Lab. Invest.
82
:
1155
-1166.
46
Piguet, P. F., C. D. Kan, C. Vesin.
2002
. Thrombocytopenia in an animal model of malaria is associated with an increased caspase-mediated death of thrombocytes.
Apoptosis
7
:
91
-98.
47
van der Heyde, H. C., M. M. Elloso, W. L. Chang, B. J. Pepper, J. Batchelder, W. P. Weidanz.
1996
. Expansion of the γδ T cell subset in vivo during bloodstage malaria in B cell-deficient mice.
J. Leukocyte Biol.
60
:
221
-229.
48
Weidanz, W. P., J. R. Kemp, J. M. Batchelder, F. K. Cigel, M. Sandor, H. C. van der Heyde.
1999
. Plasticity of immune responses suppressing parasitemia during acute Plasmodium chabaudi malaria.
J. Immunol.
162
:
7383
-7388.
49
Piguet, P. F., C. Vesin.
2002
. Modulation of platelet caspases and life-span by anti-platelet antibodies in mice.
Eur. J. Haematol.
68
:
253
-261.
50
Piguet, P. F., C. Vesin, C. Da Kan.
2002
. Activation of platelet caspases by TNF and its consequences for kinetics.
Cytokine
18
:
222
-230.
51
Madrenas, J., L. A. Chau, J. Smith, J. A. Bluestone, R. N. Germain.
1997
. The efficiency of CD4 recruitment to ligand-engaged TCR controls the agonist/partial agonist properties of peptide-MHC molecule ligands.
J. Exp. Med.
185
:
219
-229.
52
Madrenas, J., R. H. Schwartz, R. N. Germain.
1996
. Interleukin 2 production, not the pattern of early T-cell antigen receptor-dependent tyrosine phosphorylation, controls anergy induction by both agonists and partial agonists.
Proc. Natl. Acad. Sci. USA
93
:
9736
-9741.