Background
Malaria continues to claim the life of millions in the tropics and it is reported that 1.5–2.7 million deaths are observed annually mostly due to
Plasmodium falciparum [
1]. Individuals in the endemic regions become semi-immune as a result of the repeated infection [
2]. Despite being semi-immune, a significant proportion of these individuals develop the severe forms of malaria disease leading to high mortality and morbidity, with severe malaria anaemia (SMA) as one of the leading causes [
3]. However, much remains to be understood of the pathogenesis of SMA.
Central to the proposal to explain the pathogenesis of SMA is the destruction of high numbers of uninfected red blood cells (uRBC) compared with the infected RBC (iRBC) [
4], due to the consistent observation of SMA at relatively low parasite burdens of semi-immune individuals in malaria endemic areas [
5]. Jakeman
et al used a mathematical method to evaluate that with one destroyed iRBC, there is 10 destructed uRBCs [
6]. The phenomenon of high uRBC destruction at low parasitaemia in the semi-immune is still unclear, but phagocytic cells and/or CD4
+ T lymphocytes are thought to play a role [
4]. Also, inadequate reticulocyte response has been proposed as being a contributory factor to the SMA, due to an abnormal bone marrow cellularity reflected by low reticulocyte counts in SMA patient [
7].
Another process that contributes to the destruction of uRBC is the mechanical mechanism, as indicated by the role of auto-antibodies [
8,
9]. Even though elevated anti-erythrocyte ghost antibody levels have been demonstrated to be associated with human malaria infections [
10], its association with anaemia and host genetic factors has not been clarified in the semi-immune. Anti-erythrocyte auto-antibodies reacting with the surface of normal or acetone fixed human erythrocytes have also been reported to occur in
P. falciparum patients' sera [
11,
12] and are thought to be at least in part responsible for the anaemia frequently seen in acutely infected
P. falciparum patients. Using Direct Coombs antiglobulin test, previous studies proposed a relationship of anti-RBC antibodies in the anaemia seen in
P. falciparum infections [
13,
14].
Although the role of auto-immune mechanism in uRBC destruction resulting in anaemia during malaria has been debated for some time, it is still controversial. While some studies have implicated auto-antibodies such as IgM, IgG and IgA classes [
8,
15‐
18], as having specificity toward uninfected and infected RBCs, thus playing an auto-immune mediated mechanism of uRBC destruction, and others do not [
19]. Thus using the rodent model the association between level of auto-antibodies against uRBC ghost and degree of anaemia at low parasite burden in the semi-immune was investigated. Rodent model of SMA as developed by Evans
et al [
4] are uncomplicated by excessive parasite burdens. In contrast, naïve murine malaria infections are hyperparasitaemic, thereby making the associated haemolytic anaemia not be reflective of SMA in the human populations.
Since severe malaria has been found to vary from one individual to another [
20], with the implication of host genetic factors, due to variation in number of infected erythrocytes and spleen size in the naïve murine malaria [
21,
22], the role of strain specificity in auto immune mediated mechanism of uRBC destruction in the different strains of chronic infected mice was also investigated. Studies have shown that there is a differential level of auto-antibodies in other diseases such as auto-immune haemolytic anaemia in mice strains [
23,
24].
Discussion
In the early stage of malarial infection, destruction of iRBCs is the primary cause of the anaemia [
29]. The severity of anaemia with acute
P. falciparum malaria correlates with density of parasitaemia [
30]. However, in the semi-immune studies have observed that malaria anaemia occurs at low parasitaemia [
4,
5], and variation in extent of Hb reduction has also been noted in these anaemic individuals. However, the association of this RBC destruction in the semi-immune mice with an immunologic mechanism
via auto-antibody, and host genetic factors has not been explored. Results from this study shows that auto-antibody may play a role in the destruction of uRBC leading to low Hb in the semi-immune mice at low parasite burden and associated with host genetic factors.
The study here on SMA at low parasitaemia provided a fine opportunity to evaluate extent of uRBC destruction in the semi-immune. The kinetics of blood haemoglobin, reticulocyte levels and parasitaemia showed that Hb improved gradually in Balb/c, even though reticulocyte production in Balb/c was 2–3 times more than the other mice. In as much as inadequate reticulocyte response [
7] and destruction of iRBC cannot be excluded, destruction and elimination of uRBC in chronic infected mice may be a major contributory factor resulting in anaemia as observed in another [
4] and this study. This is demonstrated during the evaluation of Hb reduction per parasitaemia at the final cycle in the semi-immune mice strains and the observation of Hb loss at a much lower parasitaemia during one of the cycles of infection when compared with the first cycle infection. A recent study has shown that actual parasite numbers may be a major factor in evaluating anaemia than percent parasitaemia [
25]. However, in this study only percent parasitaemia was considered, thus further study to estimate the role of actual parasite number in such a study will be interesting. The kinetics and magnitude of reticulocyte production have been observed to be similar in both phenylhydrazine-induced anaemia and
P. berghei ANKA infected mice [
4], suggesting reticulocyte response was adequate. Inadequate reticulocyte response may be a major factor to low Hb in naïve hyperparasitaemic [
31] or acute infections. Another possible mechanism to explain for the observed low Hb during Plasmodium infections is the preference of
P. berghei ANKA for young erythrocytes/reticulocytes [
32,
33]. Thus, at all levels of parasitaemia, more especially when Hb is low, higher proportions of parasitized reticulocytes than parasitized erythrocytes have been shown to occur [
31]. Due to this phenomenon, not enough reticulocytes are able to develop into mature RBC, as both infected and uninfected reticulocytes are cleared [
31], hence the persistent low Hb in the chronic infected mice despite compensatory erythropoiesis response to haemolytic anaemia.
The destruction of uRBC may be auto-immune mediated [
16] due to the high statistical significant anti-RBC ghost antibodies reported in this study and its significant correlation with anaemia. However, it is suspected that the high auto-antibody mediation could be as a result of the RBC destruction. In that sense a lot of antigens are exposed thus enhancing the synthesis of the antibodies especially in Balb/c. The low parasitaemia observed in the Balb/c seems to indicate that its immunity is much more enhanced compared to the other strains. As a result Balb/c is able to control the parasitaemia growth. The high immune status coupled with the high antibody level in Balb/c appeared to be protective but at a cost, resulting in pathology situation of low Hb. This anti-RBC ghost antibody may lead to sensitization of RBC resulting in immune complex formation during malaria infection at the acute anaemia phase of malaria infection, which has been widely proposed as the cause of RBC destruction [
16] and resultant anaemia [
34]. Several additional autoantigens have been implicated in the auto-immune disorders occurring during malaria, including modified antigen-antibody complexes [
18]. Also, these surface-adherent antigen-antibody complexes initiate complement activation [
35,
36] inducing a prehaemolytic or a haemolytic condition, as observed in this study. The entire immune complex may be auto-immune responses leading to elimination of RBCs. The observation of continues fall in Hb after parasite clearance following treatment with antimalarial in this study and others [
7], in addition to IFA and ELISA results further support the fact that auto-immune mechanisms may be involved to some extent in the low Hb observed at relatively low parasitaemia. Similar observation was made to give explanation for the low Hb during babesiosis infections in cows [
28]. In addition to the IFA result in that study [
28], higher anti-erythrocytic auto-antibody to ghost RBC was reported in the naturally infected cows in comparison with the non-infected. A contrasting result was, however, obtained in another study, where lack of association between auto-immune mechanism and RBC in chronic malaria was reported [
19]. It is not clear if the different parasite strain used could result in this difference, thus this needs to be investigated further.
Previous work showed that depletion of macrophage delayed the clearance of uRBCs in mice, suggesting a role of macrophage in the destruction of uRBCs [
4]. At the onset of malaria infection, macrophage activity is crucial to control level of parasitaemia, via eythrophagocytosis, which is enhanced by opsonization with antibodies and other immune reactions like complement [
35,
36]. However, the over activity can result in pathology (such as low Hb) and sometimes death [
37]. Although significantly high anti-RBC autoantibody was observed in the mice strains, which will enhance macrophage activity, it was surprising that comparative Hb drop, was not observed in them as in semi-immune Balb/c. It is possible the macrophage activity may have been impaired or switched off in semi-immune B6, NZW and CBA. The evidence of low Hb drop at relatively higher parasitaemia in these semi-immune strains on one hand and Balb/c on the other could implicate haemozoin; a waste product of haemoglobin may be a contributing factor. In addition to stimulating TNF secretion, it is known to impair macrophage function [
38]. The relative higher percent parasitaemia observed in the other semi-immune mice strains other than Balb/c might produce a higher amount of haemozoin, which may impair macrophage function. In addition, haemozoin is also reported to suppress erythropoiesis [
39], agreeing well with the data in Figure
3 and Table
1, further supporting that macrophage is suppressed by haemozoin in these strains.
Variation in Hb drop and anti-erythrocytic auto-antibody at low parasitaemia in the semi-immune mice give cause to assume more of host genetic factors are at play. It was realized that in some of the strains more of Hb were lost at relatively much lower parasitaemia, and the possibility of their unique genetic background might play a great role in this various responses. How this affect the variation in Hb loss could be point for further research. This observation goes to establish the fact that despite being exposed to similar plasmodium infections at various times to become semi-immune, the individuals respond differently with some able to withstand the parasite pressure by controlling the parasite growth and others not, leading to high parasitaemia with anaemia and eventually died. It is postulated that the unique genetic background may be responsible in determining how individuals under the same level of malaria transmission in endemic areas respond differently to uRBC destruction at low parasitaemia. It is of interest to note that, the results shown here, reveals that the immune status of the semi-immune appears to delay peak parasitaemia when compared with the naïve status [
4,
40], by 2–5 days depending on the mice strain, suggesting the immune system of the semi-immune has been developed to some extent in that regard, during the repeated infections and treatment. Also, more especially in the other strains, absence of parasites at recovery could imply that the considerable effect it (parasites) exert on its host RBC, which eventually lead to similar alterations as seen in oxidatively damaged normal RBC [
41,
42] are no more. Consequently uRBC destruction is minimized.
The rodent model reported in this study is unique as it enables the study and comparison of RBC destruction in different mice strain at the same time. Similar Hb reduction in the semi-immune Balb/c compares with another study [
4], and to the knowledge of the authors those of semi-immune B6, NZW and CBA are the first to be reported here. While, some deaths were observed in this study, none was reported in that by Evans
et al, [
4]. It is not clear if the source of parasite could contribute to this. Also one advantage of the rodent model is that Hb loss at relatively low parasitaemia could be studied, which is similar to humans. However, a disadvantage in the model reported in this study is that Hb loss was just about 50% of baseline, where as Hb values < 50% has been observed in infants [
43]. It is possible the mice in this study might have become adults after several cycles of infection and treatment to generate the semi-immune status.
Finally, a study into the possible candidate gene that might be responsible in eliciting the various responses especially of Balb/c on one hand and others such as CBA, by studying into their F1 cross, will be very informative. This will help in understanding further the role of host genetic factors in auto-immune mediated RBC destruction in malaria anaemia at the molecular level.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
NTH, AY, MY, KH designed the work with GKH. TY and MNS carried out animal experiment and IFA with GKH. MK designed and carried out the ELISA with GKH. GKH drafted the manuscript with MNS, NTH and KH, who were also involved with data analysis as well extensive revision of the manuscript for intellectual content. NTH and KH supervised the work.