Plasma Kallikrein, Prolylcarboxypeptidase, and Bradykinin Receptors Mediate Experimental Murine Cerebral Malaria

Background: Cerebral malaria (CM) due to Plasmodium falciparum infection is a major cause of death in African children. It manifests with coma and brain swelling due to vasogenic edema. Bradykinin (BK) is a vasoactive peptide liberated from high molecular weight kininogen (HK). Hypothesis: BK is a proximal cause of vasogenic edema in cerebral malaria. This hypothesis was tested using genetic deletion models of proteins of the kallikrein/kinin system (KKS) in mice with CM due to murine malaria Plasmodium berghei ANKA (PbA) infection.

Methods: Murine deletion models of total kininogen (Kng1^-/-), prekallikrein (PK) (Klkb1^-/-), BK B2 receptor (Bdkrb2^-/-), BK B1 receptor (Bdkrb1^-/-), combined B1 and B2 receptors (Bdkrb1^-/-/Bdkrb2^-/-), factor 12 (FXII) (F12^-/-), and prolycarboxypeptidase (Prcp^gt/gt) C57BL/6J mice were used in these investigations. Mice were inoculated intraperitoneally (i.p.) with 10⁶ PbA-infected red blood cells (iRBCs). Infected mice were examined daily for parasitemia and deterioration of neurologic and motor behaviors using a detailed 5-point scoring system (SHIRPA score). On day 5, whole blood from infected and non-infected mice were collected to measure murine CBC with platelet counts and differential WBC counts. Plasma assays for aPTT, PT, HK, PK, FXII and C1 inhibitor (C1INH) activity and antigen levels were determined. Cerebral edema was evaluated by an Evans Blue dye (1%) IV infusion. Plasma HK antigen levels in Kenyan children with cerebral malaria (CM), uncomplicated malaria (UM) and controls also were measured by immunoblot and competitive ELISA.

Results: PbA-infected mice had a significantly lower (P < 0.0001) white blood cell count, lymphocyte count, hematocrit, RBC mean corpuscular volume, and mean corpuscular hemoglobin concentration and platelet count, compared to uninfected mice. The mean activated partial thromboplastin time (aPTT) (P = 0.001) and prothrombin time (PT) (P = 0.008) were prolonged in PbA-infected mice. Plasma HK and PK activity and antigen levels in the PbA-infected mice were significantly lower than values in uninfected mice. All PbA-infected plasma had circulated cHK with 56 and 46 kDa HK. Plasma FXII levels were unchanged between groups. Murine plasmas' C1INH activity were 50% higher in infected mice but only 50% active when compared to their antigen values.

PbA-infected Kng1^-/- mice were protected significantly from neurologic deterioration vs WT mice (P = 0.0063) and had a 58% reduction in brain edema (P = 0.0006). PbA-infected combined Bdkrb1^-/-/Bdkrb2^-/- and Bdkrb2^-/- mice had similar protection from neurologic deterioration (P=0.0031) and brain edema (P = 0.0003). PbA-infected Klkb1^-/- mice also were protected from neurologic deterioration (P = 0.003) and had 64% reduction (P < 0.0001) in brain edema. PbA-infected wild type mice treated with a PK inhibitor (RZLT7824) had significantly reduced the neurologic deterioration (P<0.05) and brain edema (P=0.0116). In contrast, PbA-infected F12^-/- mice only had a 33% reduction in brain edema (P=0.019) and no protection from neurologic deterioration. Prolylcarboxypeptidase is a PK activator in vessel wall independent of FXIIa. PbA-infected Prcp^gt/gt mice were protected significantly from neurologic deterioration (P < 0.0067) with 50% reduction in brain edema (P < 0.0007). PbA-infected wild type mice treated with a PrCP inhibitor also had a significant reduction in neurologic deterioration (P < 0.0095) and brain Evans Blue dye uptake (P = 0.0385).

Twenty human plasmas from CM patient (WHO criteria for diagnosis) were examined. Immunoblot of human plasma with antibody to human D5 showed 56 and 46 kDa cHK in 40% (8/20) of CM vs 18% (3/18) UM patients. HK levels of CM patients at hospital admission (59 ± 4.1 μg/ml; mean ± SEM) were significantly lower (P < 0.001) than at discharge (70 ± 3.6 μg/ml; P < 0.0001) and compared to healthy children and children with non-malaria illness.

Conclusion: These results indicate that in murine and human CM, a high percentage of circulating HK is cleaved (cHK) indicating prior BK release which may mediate brain edema with neurologic deterioration. In murine CM, kininogen, prekallikrein, prolylcarboxypeptidase, or bradykinin receptor deficiency ameliorate the disorder. Factor XII deficiency does not. Pasma kallikrein and propylcarboxypeptidase inhibitors have potential to be adjuncts to current therapy for CM.