Conceived and designed the experiments: TW KM DM PB DC FV. Performed the experiments: KT GW FO. Analyzed the data: KT FV GW FO. Contributed reagents/materials/analysis tools: GF JS TH GB PA IB. Wrote the paper: DM FV.
The authors have declared that no competing interests exist.
A recently proposed mechanism of protection for haemoglobin C (HbC; β6Glu→Lys) links an abnormal display of
Red blood cell disorders including haemoglobin variants and thalassaemias provide with their unusually high prevalence and distribution in malaria endemic areas
The coexistence of HbC and HbS in a hyperendemic malaria context such as Burkina Faso provided a unique opportunity to explore this question in a comparative approach. In the present study, we compared the humoral response to
Healthy individuals belonging to the Mossi ethnic group of Burkina Faso were recruited after oral informed consent was obtained during the dry seasons 2005/2006 at the “Centre Medical Saint Camille” of the capital city Ouagadougou, and in the villages nearby (Donsin, Kuiti, Nagbagre') of the Kadiogo district, as described in
Geographic Origin | Period of blood collection | Haemoglobin genotype | ||||||
Rural samples (EIR>100) | AA | AC | CC | AS | SC | SS | Total | |
Donsin | Dry season (April 2006 ) | 46 (16.0) | 16 (11.7) | 4 (9.0) | 2 (8.5) | 2 (13.0) | - | 70 |
Kuiti | Dry season (April 2006 ) | 20 (10.7) | 11 (10.2) | 8 (9.2) | 5 (10.2) | 2 (9.0) | 2 (9.5) | 48 |
Nagbagré | Dry season (April 2006 ) | - | - | 2+1 | - | - | - | 3 |
Urban samples (EIR:1-10) | ||||||||
Ouagadougou | Dry season (January–May 2005) | - | 80 (30.0) | 10 (27.4) | 40 (29.5) | 8 (19.2) | 2 (8.0) | 140 |
Ouagadougou | Dry season (January–May 2006) | 86 (29.0) | 26 (24.5) | 3 | 8 (32.5) | 3 (18.3) | - | 126 |
Total | 152 | 133 | 28 | 55 | 15 | 4 | 387 |
Individuals are indicated for each genotype and origin; age is given in brackets.
recruited in December 2004;
age available for 50 subjects;
age available only for 5 subjects;
age available only for 29 subjects;
age available only for 6 subjects;
age available only for 63 subjects;
age available only for 22 subjects;
age available only for 6 subjects.
Antibody responses against the parasite infected-erythrocyte surface antigens were tested using the method with modifications of Williams et al.
Mature trophozoite stage parasitized RBC (pRBC) at between 3–5% parasitaemia were thawed from frozen culture using saline solutions according to a gradient from 12% to 0.9% in a suspension at 50% haematocrit with buffer (0.5% BSA/PBS). 1 µl of serum was pipetted into separate wells of a 96-well U-bottomed plate (Falcon, Becton Dickinson, USA) to which was added 11.5 µl of the infected pRBC cell suspension in 0.5%BSA/PBS and 10 µg/ml of ethidium bromide. The mixture was incubated at room temperature for 30 min, following which the cells were washed three times with 0.5%BSA/PBS, centrifuging at 1000 rpm for 3 min per wash. The cells were then re-suspended in 50 µl 0.5%BSA/PBS containing a 1:50 dilution of sheep anti-human γ chain (FITC) fluorescein isothiocyanate-conjugated antibody (The Binding Site) was added to each well. A further incubation at room temperature in darkness, for 30 min was carried out, after which, following a further series of washes, at least 1000 pRBC were counted on an EPIC/XL flow cytometer (Coulter, UK). The Mean Fluorescence Intensity (MFI) was defined as the difference between the geometric mean of the fluorescence emitted by trophozoite pRBC and the geometric mean of the fluorescence emitted by the non pRBC. Non-specific recognition as measured by European negative controls was subtracted by the MFI of tested individuals.
Enzyme linked immunosorbant assays (ELISA) were performed according to well established protocols
No differences in the Mean Fluorescence Intensity (MFI) according to the haemoglobin genotype were detected when looking at urban and rural samples all together. Further comparisons were carried out separately with the urban and the rural samples due to the different EIR (entomological inoculation rate) and mean age of the two subsets whose characteristics are described in
Prevalence and levels of antibodies tested against all antigens was consistently higher in the villages due to higher exposure to malaria of the rural subset despite the relatively younger age (
Antigen | Prevalence | Pairwise comparisons | ||||
Urban samples (N = 266) | AA vs AC | AC vs CC | AA vs CC | AA vs AS | Overall | |
AMA1 | 78% | P = 0.001 | NS | P = 0.01 | P = 0.001 | P = 0.0001 |
EBA-175 | 73% | P = 0.04 | NS | P = 0.01 | P = 0.03 | P = 0.02 |
MSP-119 | 67% | NS | P = 0.03 | P = 0.01 | P = 0.01 | NS (P = 0.06) |
MSP-2 | 64% | P = 0.001 | NS | P = 0.03 | P = 0.004 | P = 0.0001 |
MSP-3 | 70% | NS | NS | NS | P = 0.02 | NS |
CSP | 65% | P = 0.05 | NS | P = 0.03 | P = 0.001 | P = 0.03 |
PSE | 74% | P = 0.01 | NS | P = 0.04 | P = 0.01 | P = 0.05 |
Rural samples (N = 121) | ||||||
AMA1 | 90% | P = 0.03 | NS | P = 0.03 | NS | NS |
EBA-175 | 76% | NS (P = 0.07) | NS | NS | NS (P = 0.06) | P = 0.05 |
MSP-119 | 70% | NS | NS | NS | NS | NS |
MSP-2 | 82% | NS | NS | NS | NS | NS |
MSP-3 | 77% | NS | NS | NS | NS | NS |
CSP | 68% | NS | NS | NS | NS | NS |
PSE | 78% | NS | NS | NS | NS | NS |
Antigen | AA | AC | AS | CC | SC | SS |
Urban Samples (N = 266) | ||||||
AMA1 | 2.90 (2.78–3.03) | 3.14 (3.06–3.22) | 3.20 (3.10–3.31) | 3.27 (3.16–3.38) | 3.23 (3.05–3.41) | 3.40 (3.35–3.46) |
EBA-175 | 2.75 (2.70–2.81) | 2.82 (2.78–2.87) | 2.84 (2.77–2.92) | 2.91 (2.79–3.04) | 2.82 (2.65–2.99) | 3.15 (3.05–3.25) |
MSP-119 | 2.45 (2.34–2.58) | 2.56 (2.46–2.66) | 2.67 (2.54–2.81) | 2.83 (2.60–3.08) | 2.60 (2.38–2.85) | 2.81 (2.58–3.05) |
MSP-2 | 2.97 (2.84–3.10) | 3.19 (3.13–3.26) | 3.22 (3.12–3.32) | 3.26 (3.15–3.38) | 3.38 (3.36–3.40) | 3.40 (3.35–3.46) |
MSP-3 | 2.44 (2.35–2.53) | 2.50 (2.43–2.58) | 2.60 (2.50–2.71) | 2.55 (2.37–2.74) | 2.66 (2.38–2.97) | 2.52 (2.45–2.61) |
CSP | 2.74 (2.69–2.80) | 2.82 (2.77–2.87) | 2.88 (2.82–2.95) | 2.88 (2.75–3.01) | 2.85 (2.70–3.01) | __ |
PSE | 2.50 (2.44–2.57) | 2.60 (2.56–2.65) | 2.64 (2.56–2.72) | 2.66 (2.55–2.77) | 2.63 (2.46–2.82) | 2.74 (1.26–3.95) |
Rural samples (N = 121) | ||||||
AMA1 | 3.12 (3.04–3.20) | 3.27 (3.18–3.36) | 3.09 (2.54–3.76) | 3.31 (3.23–3.40) | 3.06 (2.59–3.62) | 3.39 |
EBA-175 | 2.82 (2.74–2.90) | 2.94 (2.85–3.03) | 3.03 (2.76–3.34) | 2.95 (2.81–3.08) | 2.70 (2.28–3.20) | 3.16 |
MSP-119 | 2.82 (2.60–3.06) | 2.54 (1.31–4.92) | 1.92 | __ | __ | __ |
MSP-2 | 2.82 (2.70–2.94) | 2.87 (2.75–2.99) | 2.82 (2.37–3.35) | 2.89 (2.57–3.25) | 2.88 (2.41–3.45) | 3.24 |
MSP-3 | 2.57 (2.48–2.67) | 2.71 (2.58–2.84) | 2.52 (2.11–3.01) | 2.69 (2.41–3.01) | 2.86 (1.03–5.97) | 2.55 |
CSP | 2.80 (2.75–2.85) | 2.82 (2.77–2.86) | 2.88 (2.82–2.94) | 2.76 (2.73–2.79) | 2.77 (2.68–2.85) | __ |
PSE | 2.55 (2.46–2.65) | 2.61 (2.49–2.74) | 2.72 (2.55–2.90) | 2.59 (2.48–2.70) | 2.56 (0.83–7.89) | 2.62 |
single individual
The horizontal lines in each box correspond to the median values, the lower edge of each box is the 25% ile and the upper edge is the 75% ile. The whiskers represent the range of the data beyond these percentiles, excluding outliers represented by dots.
Despite substantial evidence of protection against clinical malaria given by the haemoglobin variants HbC and HbS, the precise mechanism(s) are still under debate. A recently proposed mechanism suggested that the protection conferred by HbC may be attributed to the reduction of cytoadherence and impaired rosetting observed in relation to the altered display of PfEMP1 on HbCC erythrocytes
Antigen | Variable | |||
Hb genotype | Age | Gender | Parasite + | |
AMA1 | P = 0.001 | NS | NS | NS |
EBA-175 | NS | NS | NS | NS |
MSP-119 | NS (P = 0.06) | NS | NS | NS |
MSP-2 | P = 0.003 | NS | NS | NS |
MSP-3 | NS (P = 0.08) | NS | NS | NS |
CSP | P = 0.004 | NS | NS | NS |
PSE | P = 0.01 | P = 0.01 | NS | NS |
A4 MFI | NS | NS | NS | NS |
CI MFI | P = 0.05 | NS | NS | NS |
We believe that the discrepancy between results from urban and rural settings could be the consequence of saturated immunity in high transmission contexts. These observations emphasize the need, when studying candidates of genetic resistance/susceptibility to malaria and their underlying hypothesized mechanisms, to take into full account the epidemiological context as a potential confounder.
Although this study can not conclusively validate any of the different functional mechanisms proposed to explain the protection of haemoglobin variants, a consistently enhanced immune reactivity in both HbC and HbS adaptive genotypes suggests the idea of a convergence in terms of their impact on the acquisition of immunity against malaria. It has been already shown that the model initially proposed for explaining G6PD protection, by enhanced phagocytosis of ring- parasitized altered erythrocytes, fits well also in the case of HbS and β-thalassemia
This work is dedicated to the memory of our colleague, Gaia Luoni (1970–2004). We are grateful to the study participants in Burkina Faso for their cooperation, to the laboratory staff at the Centre Medical Saint Camille of Ouagadougou, Burkina Faso and at KEMRI, Kenya for excellent technical support. We thank all the colleagues who kindly provided us with recombinant proteins of