1US Army Medical Research Unit, and Kenya Medical Research Institute and 2Department of Zoology, Kenyatta University, Nairobi, Kenya; 3Department of Medicine, Uniformed Services University of Health Sciences, Bethesda, Maryland
Received 28 June 2002; revised 30 September 2002; electronically published 24 January 2003.
Severe anemia is one of the most lethal complications of Plasmodium falciparum malaria. Red blood cells (RBCs) from children with severe malarial anemia are deficient in complement regulatory proteins (CR1 and CD55). A case-control, age- and sex-matched study was carried out to determine whether these deficiencies are acquired or inherited and the relative contribution of these complement regulatory protein deficiencies, the immune complex level, and the parasite density to the development of severe malarial anemia. RBC CR1 and CD55 deficiencies resolved after treatment, suggesting that these changes were acquired. Using conditional logistic regression, a decline in CD55 (or CR1) (odds ratio [OR], 4.2; 95% confidence interval [CI], 2.1 8.1; P < .001) and an increase in immune complex level (OR, 2.6; 95% CI, 1.54.8; P = .001) were significantly associated with severe malarial anemia.
|
Financial support: Military Infectious Disease Research Program; US Department of Defense; World Health Organization's Multilateral Initiative on Malaria.
Subjects were recruited under a protocol approved by the Office of the Surgeon General, US Army, and the Kenya Medical Research Institute, Nairobi. Informed consent was obtained from the parents or guardians of all participants in accordance with all applicable guidelines.
The views of the authors do not purport to reflect the position of the US Army or the US Department of Defense.
This article is published with the permission of the director, Kenya Medical Research Institute.
Reprints and correspondence: Dr. José A. Stoute, US Army Medical Research Unit, Unit 64109, APO AE 09831-4109
Each year, at least 1 million African children die of malaria caused by the protozoan parasite Plasmodium falciparum [1]. Most of these deaths occur as a result of complications, such as severe anemia (SA) or cerebral malaria. The pathogenesis of severe malarial anemia is not well understood. The degree of red blood cell (RBC) loss cannot be explained solely by the direct destruction of RBCs by the parasite [2, 3]. Several groups have reported the finding of IgG and complement-coated RBCs in patients with severe malarial anemia [4, 5], suggesting a role for complement-mediated damage of RBCs.
Although complement activation and immune complex (IC) formation are prominent features of malaria infection [6, 7], little work has been done to evaluate the role of complement regulatory proteins and IC in this infection. RBC complement regulatory proteins include complement receptor 1 (CR1; CD35), decay accelerating factor (CD55), and the membrane inhibitor of reactive lysis (CD59). These proteins are important in protecting RBCs from complement-mediated damage and in controlling the complement activation cascade [8]. In a previous study [9], we showed that RBCs of children with severe malarial anemia are deficient in CD35 and CD55. In the study presented here, we recruited a new cohort of volunteers to determine whether RBC complement regulatory protein deficiencies in children with severe malarial anemia are acquired or inherited and to determine the relative contribution of ICs, parasite density, and complement regulatory protein deficiency to the development of SA.
Subjects and methods. The study was case control in design. Patients with severe malarial anemia and control subjects were recruited from the Nyanza Provincial General Hospital (NPGH) in Kisumu, western Kenya. Patients with SA were defined as children with asexual P. falciparum parasitemia, confirmed by a Giemsa-stained blood smear, and hemoglobin (Hgb) level 5 g/dL. Each patient with SA was matched by age (±2 months) and sex to a child with symptomatic uncomplicated malaria, here referred to as "symptomatic control" (SC), with a Giemsa-stained blood smear showing asexual parasitemia plus an axillary temperature 37.5°C or, in absence of the latter, a history of any 2 of the following: nausea and/or vomiting, diarrhea, irritability, or poor feeding. Subjects were excluded from participation if there were evidence of other concomitant infections, chronic illness, or a history of blood transfusion preceding enrollment. Children were followed-up at 
1-month intervals for a period of 36 months after enrollment. EDTA-anticoagulated blood (2.5 mL) was obtained at enrollment and at monthly intervals.
The personnel who performed all assays were unaware of the clinical status of the volunteers and had no access to the clinical files. All procedures and reagents for indirect fluorescent staining were essentially as described elsewhere [9]. The number of molecules of CR1 per RBC was derived from a fluorescence standard curve created using RBCs with known numbers of CR1. RBC CD55 antibody binding capacities (ABCs) were derived from a standard curve using beads of known ABCs (Flow Cytometry Standards).
Because loss of RBC complement regulatory proteins occurs in conditions characterized by increased IC level [10], we determined whether children with severe malarial anemia have high levels of CIC by a C1q solid phase assay. Wells of an Immulon II HB 96-well plate (Thermo Labsystems) were coated with 10 
g/mL C1q (Sigma-Aldrich) in PBS (pH 7.4). After overnight incubation at 4°C, the plates were washed with wash buffer (0.5% Tween 20 in PBS [pH 7.4]) and were blocked for 1 h at room temperature with blocking buffer (PBS, 0.5% boiled casein, 1% Tween, 0.01% Thimerosal, and 20 g/mL phenol red). Aggregated human IgG (AHG) was prepared by heating 6.5 mg/mL of purified human IgG (Sigma) in PBS at 63°C for 30 min followed by fractionation over a Sephacryl S-300 70 × 2.6 cm column (Amersham Pharmacia Biotech). Fractions from the first peak were pooled and dialyzed overnight against 1 L of PBS. After determination of total protein concentration, the AHG was aliquoted and stored at -70°C until used. Serial dilutions of AHG were made in PBS for use as standard. Standard and test plasma samples were diluted 60 times in dilution buffer (PBS, 0.5% boiled casein, 0.5% Tween, 0.01% Thimerosal, and 20 g/mL phenol red), and 100 L added to duplicate wells and incubated for 1 h at room temperature. The wells were emptied and washed 4 times with wash buffer. Horseradish peroxidaseconjugated goat antihuman IgG (Kirkegaard & Perry Labs) was diluted 1 : 5000 in wash buffer containing 0.5% boiled casein, 100 L was added to each well, and the wells were incubated for 1 h at room temperature. After 4 washes, 200 L of ABTS substrate (Kirkegaard & Perry Labs) was added to each well, the wells were incubated for 45 min, and the OD415 nm was measured. IC level was expressed as g AHG equivalents per milliliter.
Statistical analysis was performed using SPSS software package (SPSS). Matched intergroup comparisons of continuous variables and categorical variables were done by use of the paired Student's t test and McNemar's test, respectively. Variables that showed significant or near-significant differences between groups by univariate analysis or that were felt to be involved in the pathogenesis of severe malarial anemia were used in a conditional logistic regression analysis to assess their contribution to the development of SA. Because the loss or gain of 1 unit of CR1, CD55, or IC is unlikely to be of clinical significance, these variables were transformed into more clinically relevant units by dividing them by their respective standard deviations. Likewise, parasite density was log10 transformed, and the number of degrees centigrade above 37.5°C was used for this analysis instead of the recorded axillary temperature. All tests were 2-tailed, with 
= 0.05.
Results. summarizes the demographic and clinical characteristics of the groups. In addition to demographic variables and the level of complement regulatory proteins and ICs, the univariate analysis contained other variables that could influence the risk of SA, such as the use of mosquito nets [11] and the duration of fever at home before arrival at the hospital. The groups were well balanced in terms of age, sex, and ethnic background. There also were no differences in the location or district of origin among the groups (data not shown). Although patients with SA tended to have higher parasite density than those in the SC group, this difference was not statistically significant when the parasite density was log10 transformed to decrease the variance.
| fig.ommitted | Table 1. Demographic, clinical, and red blood cell (RBC) parameters at enrollment. |
There were 13 (22.4%) deaths among 58 patients with SA at enrollment. Two additional deaths occurred in this group following discharge after their initial hospitalization for anemia. There were 2 (3.5%) deaths among 57 patients in the SC group, all of which occurred during or following a hospitalization for malaria.
RBCs of patients with SA were found to have significantly lower levels of CR1 and CD55 than control subjects at enrollment (P < .001; ). After transfusion and treatment for malaria, RBC CR1 and CD55 increased and remained similar to that of control subjects during the remaining 36-month follow-up period.
| fig.ommitted |
Figure 1. Time course of hemoglobin (A), anti-CD55 antibody binding capacity (ABC) (B), and CR1 (C) in children enrolled with severe malarial anemia and control subjects with symptomatic uncomplicated malaria. Children were followed-up after enrollment, and blood samples were taken at |
We found colinearity between CR1 and CD55 ABC. Therefore, only CD55 ABC was used in the statistical analysis. SA was significantly associated with lower CD55 ABC (odds ratio [OR], 4.2, for every decline in CD55 ABC of 1150 [1 SD]; 95% confidence interval [CI], 2.18.1; P < .001) and higher IC level (OR, 2.6, for every increase of 4.6 g AHG eq/mL [1 SD]; 95% CI, 1.54.8; P = .001). There was no association between SA and duration of fever (OR, 0.9 per day of fever; 95% CI, 0.81.2), axillary temperature >37.5°C (OR, 0.9 for every degree centrigrade above 37.5°C; 95% CI, 0.61.4), or log10 parasite density (OR, 0.9 per log increase; 95% CI, 0.61.5; P < .001).
Discussion. A number of observations suggest that destruction of uninfected RBCs makes the most significant contribution to the development of severe malarial anemia. First, the degree of RBC destruction exceeds by far the level of parasitemia. Second, the hemoglobin count may continue to decrease after treatment, which suggests that alternative mechanisms of RBC destruction are at play [2]. In addition, a number of studies also have shown that the life span of uninfected RBCs decreases during malaria infection [12]. Therefore, our studies have focused on identifying the lesions on the RBCs surfaces that could explain their accelerated destruction during malaria infection.
RBCs can be destroyed or removed from circulation by complement-mediated lysis or through phagocytosis of damaged or opsonized RBCs. These 2 processes are not mutually exclusive and can be fueled by the same mechanism (i.e., the presence of IC and activated complement on the RBCs). Complement regulatory proteins, such as CR1 and CD55, protect RBCs and other tissues from complement-mediated damage by removing CICs as well as regulating the complement activation cascade [8]. CR1 also was recently implicated in the pathogenesis of cerebral malaria by mediating the binding of uninfected RBCs to mature trophozoites or schizonts (rosetting) [13]. CD55 is a glycosyl-phosphatidyl-inositol (GPI)anchored protein that serves to inhibit the complement activation cascade by accelerating the decay of C3 convertases.
Motivated by previous findings of RBC complement regulatory protein deficiencies in diseases characterized by hemolytic anemia such as paroxysmal nocturnal hemoglobinuria [8], systemic lupus erythematosus (SLE), and other autoimmune diseases [10], we looked for and found similar deficiencies in children with severe malarial anemia [9]. In the present study, we recruited a new group of participants and followed them to determine whether these complement regulatory protein defects are acquired or inherited. At enrollment, the RBCs of children with severe malarial anemia were again found to be deficient in CR1 and CD55. These deficiencies corrected rapidly following treatment of malaria and blood transfusion. CR1 and CD55 counts remained stable long after 4 months, which is the maximum expected life span of the donor RBCs. In addition to confirming our initial observations, these results indicate that the complement regulatory protein deficiencies observed on enrollment are acquired and not inherited.
Although the etiology of the reductions in complement regulatory proteins is not clear, it is likely that they are linked to the removal of IC by tissue macrophages. C3b-containing ICs bind to CR1 on the RBC surface and are carried to the macrophages of the reticuloendothelial system where they are safely removed and the RBCs recirculated [14]. In this process, CR1 is proteolytically cleaved from the surface of RBCs and hence continuously lost. The mechanism of decline of CD55 is less clear. However, CD55 reduction has been observed in other disorders known to be associated with hemolysis and IC formation such as SLE [15]. Our findings confirm that ICs are present during malaria infection and that their levels are significantly higher in children with severe malarial anemia than in symptomatic uncomplicated malaria. These results contradict the findings in a previous study [6]. However, in that study, the number of cases was fewer and the degree of anemia was less severe than in the children studied here.
As evidenced from this study, many children are able to tolerate high parasitemias without a significant decrease in hematocrit levels. Determining why some children are able to tolerate high parasitemias while others are not is key to understanding the pathogenesis of malaria complications. Our data suggest that a decline in CD55 and CD35, together with an increase in the level of IC, places parasitemic children at further risk for the development of SA. We propose that when RBCs reach a certain level of CR1 and CD55 deficiency, they lose their ability to control complement activation and remove ICs from circulation. C3b and IC then are indiscriminately deposited on RBCs which leads to their destruction by either phagocytosis and or complement-mediated lysis.
Acknowledgments
We thank the children who participated in our studies, as well as their parents, for their willingness to contribute to this research endeavor. We are very grateful to Dr. Jacques Cohen (Hospital R. Debré, Reims, France) for measuring the number of CR1 molecules in RBCs used as standards. In addition, we are indebted to our dedicated staff of clinicians, nurses, drivers and field workers, who made these studies possible. This work is dedicated to the memory of Malachi Odinga Opollo, scientist, colleague, and friend, whose untimely death left a vacuum in our laboratory.
Impact of clopidogrel on postoperative blood loss after non-elective coronary bypass surgery
Obesity and Cardiovascular Disease: Pathophysiology, Evaluation, and Effect of Weight Loss
Randomized Trial of a Nonpolymer-Based Rapamycin-Eluting Stent Versus a Polymer-Based Paclitaxel-Eluting Stent for the Reduction of Late Lumen Loss
Robustness of Late Lumen Loss in Discriminating Drug-Eluting Stents Across Variable Observational and Randomized Trials
Low Thyroid Function Leads to Cardiac Atrophy With Chamber Dilatation, Impaired Myocardial Blood Flow, Loss of Arterioles, and Severe Systolic Dysfunction
Effect of Pravastatin on Rate of Kidney Function Loss in People With or at Risk for Coronary Disease
Physiological Background of the Loss of Fractal Heart Rate Dynamics
Late Loss in Lumen Diameter and Binary Restenosis for Drug-Eluting Stent Comparison
Magnetic Resonance Imaging During Untreated Ventricular Fibrillation Reveals Prompt Right Ventricular Overdistention Without Left Ventricular Volume Loss
Transgenic Mouse Model of Ventricular Preexcitation and Atrioventricular Reentrant Tachycardia Induced by an AMP-Activated Protein Kinase Loss-of-Function Mutation Responsible for Wolff-Parkinson-Whit