Multiplicity of malaria infection is suggested to be influenced by age, transmission intensity and seasonal variation [3
]. As expected, a general higher parasite density and MOI after the transmission season was observed in this study.
The distribution of msp2
3D7/FC27 allelic families before (40/40 percent) and after (43/44 percent) the transmission season was similar. These figures are very close to those reported in Gambia [28
], Papua New Guinea [9
] and Senegal [29
] but differ from those observed in other areas (differing in malaria transmission intensity) in Senegal [10
]. This suggests that heterogeneity of P. falciparum
infection may differ according to geographical location, transmission intensity and season of sampling.
As has been proposed by previous studies [31
], high parasite densities increase the probability of detecting concurrent clones in an individual. Consistent with this, a positive correlation between MOI and parasite density was observed in this study.
Whether or not MOI is influenced by RBC variants was examined by present study. ABO blood group had no impact on MOI at any time points. This probably reflects the fact that the ABO system does not affect the parasite densities rather the clinical outcome of the disease, where blood group O protects against cerebral malaria [18
Previous studies on the association between sickle cell trait and MOI are contradictory [14
]. No effect of HbAS on MOI was noted in present study. This is in line with reports by Konate et al
] in Senegal but in contrast to what has been shown in Gabon [14
]. The reasons for these discrepancies are probably due to differences in the transmission intensity and in the ages of the study groups.
Alpha-thalassaemic red blood cells have altered membrane properties [33
], which may affect erythrocyte invasion by merozoites [19
]. MSP2 antigen is involved in invasion [19
] and is polymorphic. Hence, changes in membrane of α-thalassaemic RBCs may interrupt invasion by parsites with certain msp2
variants, and that could eventually influence the MOI. In accordance with this, no significant increase in the MOI, after the transmission, was seen in the α-thalassaemic children. This suggests that α-thalassaemic RBCs may protect against invasion by certain parasite strains. However, these data contradicts findings reported by Mockenhaupt et al
], who did not find an influence of α-thalassaemia on the prevalence, density and multiplicity of symptomatic P. falciparum
infection. It is now evident that α-thalassaemia does not affect parasite density in individuals that go on to present clinical malaria [35
]. The reason for these discrepancies might be due to that Mockenhaupt et al
studied symptomatic patients, while present study comprised asymptomatic children. Differences in sample sizes of the two studies could also be a reason for contradictory observations. In any case these data indicate that α-thalassaemic individuals may be less susceptible to infection by certain parasite strains but once they get infected they are not protected against clinical malaria.
Similar to α-thalassaemic, G6PD mutated children did not show a significant increase in the MOI after the transmission season. This may be explained by early phagocytosis of infected G6PD deficient RBCs. Cappadoro et al
] have shown that ring-stage infected G6PD deficient RBCs are 2.3 times more efficiently phagocytosed than normal iRBC. Thus, G6PD mutated individuals probably control the parasite growth better than subjects carrying the normal G6PD gene. This would then lead to lower MOI as a result of lower parasite counts at the end of transmission season. This is supported by the fact that girls with mutated G6PD, in this study, were found to have significantly lower parasite prevalence and density as compared to the girls carrying the normal gene. This phenomenon was reinforced in the case of girls carrying G6PD A- and HbAS combined erythrocyte defects, whereas HbAS alone was not found to influence MOI over the transmission season. No similar observation could be made for boys, for whom the G6PD A- carriage at the hemizygous status is assuredly responsible for enzyme deficiency: the small sample size of the groups of erythrocyte variant (considered alone or in combination) carriers, could have generated this absence of divergent parasite data between groups of males.
Previous studies regarding the variation of MOI over age have suggested that the influence of age on the multiplicity of infection is highly affected by endemicity of malaria [1
]. This is probably a reflection of the development of anti-parasite specific immunity [10
]. Thus, in holo- or hyperendemic area, immunity develops faster and at younger age than in areas with less intense transmission [39
]. Studies have shown an age-dependent MOI in a village with intense perennial malaria transmission but not in areas where malaria is mesoendemic [15
]. In line with this, here we showed that MOI is not influenced by age, at least not in the age range of 2 to 10 years. However, the lack of correlation between MOI and age in present study needs to be confirmed in a study population with broader age range.
Reports regarding the relation of MOI and malaria morbidity are contradictory. A number of studies has suggested that high MOI may confer protection from subsequent clinical malaria [4
]. A few studies also have associated malaria morbidity with high MOI [7
]. In contrast with those, no correlation between MOI and the risk of clinical presentation was noted in this study. This contradiction may be in part due to variable genotyping protocols, differing in sensitivity and accuracy, used in different laboratories [15
]. In this study, sampling was done only at two time points and MOI was not determined at the time of clinical presentation. This might be another possible reason for the lack of correlation between MOI and development of clinical malaria, noted in the present study. Moreover, this result underlines the fact that parasite characteristics are certainly not the only ones involved in malaria morbidity. Host characteristics such as genetic polymorphisms, or transmission intensity could also play a key role.
However, to be able to compare findings by different studies on diversity of P. falciparum infection in relation to other parasitological indexes and/or host factors, performed in different areas, methodology used by these studies should be taken into consideration.
The genotyping results generated by PCR may differ between different laboratories because of applying variable reagents and protocols or even handling [32
]. In addition, the PCR used in this study to determine diversity of parasite population do not discriminate between virulent and avirulent strains. This could perhaps be a general explanation for some of our observations here such as the lack of association between MOI and malaria morbidity, or the variable effects of erythrocyte variants on parasite diversity and clinical pattern of the disease.