Background
Malaria caused by
Plasmodium falciparum continues to cause over a half million deaths each year, with children being disproportionately affected [
1]. Children suffer the greatest morbidity and mortality from malaria since immunity to malaria takes years to develop, increasing with age and exposure [
2,
3]. One manifestation of acquired immunity to malaria is control of blood stage parasites, resulting in lower parasite densities and lack of febrile symptoms of disease [
4‐
6]. Antibodies have been shown to be an important mediator of this blood stage immunity [
7‐
10].
Effective B cell and antibody responses to
Plasmodium infection generally develop only after years of repeated exposure, likely due to immune immaturity of the host and antigenic variation of parasites [
8‐
12]. Another hypothesis for the slow development of immunity is that
Plasmodium infection may interfere with B cell development and maintenance of memory responses [
13‐
17]. After initial maturation in the bone marrow, B cells pass through a series of developmental differentiation stages, many of which can be detected in the peripheral blood. Transitional B cells emerge from the bone marrow and mature into naïve B cells prior to antigen exposure. After antigen exposure, B cells in secondary lymphoid organs differentiate into class-switched classical memory B cells (MBCs) non-class switched ‘innate-like’ MBCs and antibody-secreting plasmablasts/plasma cells [
18]; these cells can be detected in blood as they migrate to other secondary lymphoid organs and tissues. Exposure to
Plasmodium alters the distribution of these B cell sub-sets, and has been associated with an expansion of ‘atypical’ MBCs in individuals living in malaria-endemic areas [
13‐
15,
19]. Atypical MBCs are class-switched but lack the classical MBC marker CD27, and unlike classical MBCs, do not appear to readily produce antibodies [
13,
20,
21]. This functional difference has led to the hypothesis that atypical MBCs may be ‘exhausted’ and may interfere with development of effective immunity [
13,
21]. On the other hand, higher circulating proportions of atypical MBCs and immunity to malaria are both associated with increasing age and
P. falciparum exposure [
13,
14,
22‐
24]. Thus, the relationship between atypical MBCs and immunity to malaria remains unclear.
B cell sub-sets generated during malaria episodes may indicate which B cells are associated with developing immunity. Various studies have described multiple B cell sub-sets in people exposed to varying levels of malaria [
11,
13,
14,
20‐
23,
25,
26], but the kinetics of B cell responses following malaria have not been well described in humans. One study tracked the kinetics transitional B cells following malaria and found that the relative proportion of these cells increased following malaria [
19]. Studies of experimental infection of mice with
Plasmodium chabaudi have found that newly differentiated plasmablasts only circulate in the blood for a short time following primary or secondary infection while other sub-sets such as transitional, naïve B cells and MBCs fluctuate greatly but remain readily detectable in the peripheral blood [
26]. These findings suggest that there are likely to be dynamic changes in the composition of the B cell pool both during and following acute malaria in humans, and that these changes may be reflected in the peripheral blood. Here, the kinetics of six distinct sub-sets of B cells were evaluated during and after treatment for symptomatic malaria, and sub-set proportions were evaluated for associations with measures of immunity to malaria.
Discussion
This study investigated the kinetics of different B cell sub-sets at the time of and following symptomatic malaria and then evaluated whether proportions of these sub-sets were associated with measures of immunity to malaria. The proportions of atypical MBCs and transitional B cells both increased during the 28 days following symptomatic malaria, whereas naïve B cell proportions declined slowly and circulating PCs proportions declined rapidly following treatment of malaria, increasing again in participants who had a second episode of malaria. Notably, children with evidence of greater immunity to malaria had higher proportions of atypical MBCs and lower proportions of transitional B cells following malaria, and these associations remained stable over the 28 days following malaria. These findings suggest that children with higher proportions of these cells during and following malaria have improved immune responses to P. falciparum infection.
People living in malaria-endemic regions have increased proportions of atypical MBCs, with these proportions increasing with age and with cumulative
Plasmodium exposure [
11,
13,
14]. Atypical MBCs have been characterized as having upregulated inhibitory pathways compared to classical MBCs [
13,
20,
21]. Despite one report suggesting serum antibodies may be produced by atypical MBCs [
22], other studies have reported that atypical MBCs have a poor capacity for differentiation and antibody production [
13,
20,
21]. With no direct evidence for an alternative function for atypical MBCs, these cells have been hypothesized to represent a dysfunctional or ‘exhausted’ phenotype. However, atypical MBC proportion, like immunity to symptomatic malaria, is associated with age and transmission intensity [
20,
24]. This study was not able to evaluate a causal relationship between atypical MBCs and immunity to malaria, but it demonstrates that the accumulation of atypical MBCs in the peripheral blood during and after malaria is associated with measures of clinical immunity. This association did not appear to merely be an artifact of age, since all participants in this study were nearly the same age, or of varied
P. falciparum exposure, as results were unchanged in a multivariate analysis including estimates of exposure. One possibility is that the ability to sustain asymptomatic infections, which tend to be of longer duration than treated symptomatic malaria infections, leads to higher proportions of atypical MBCs due to chronic antigen exposure. Alternatively, it may be that higher parasite densities seen in less immune individuals drive down the proportion of atypical MBCs, possibly as a result of apoptosis or homing of atypical MBCs to tissues. It is also possible that atypical MBCs contribute, in an as-yet undefined manner, to anti-malarial immunity. Further studies are necessary to determine if atypical MBCs are causally associated with immunity to malaria.
Increases in transitional B cell proportions following malaria have been previously noted, and this expansion was hypothesized to be due to disruption of B cell homeostasis [
19]. This study replicates the finding of increased proportions of circulating transitional B cell following malaria. In addition, this study found participants with greater parasite density at the time of malaria had higher proportions of transitional B cells. It is possible that this association could be driven by greater
P. falciparum induced, non-specific, polyclonal B cell activation [
39] and apoptosis [
40] in individuals with higher parasite burdens, leading to a homeostatic expansion of transitional B cells. It is also possible that transitional B cells function in an immunoregulatory capacity [
41], similar to IL-10 producing T cells, in the context of malaria [
42]. Transitional B cells from healthy US adults have been shown to secrete IL-10, a potent immunoregulatory cytokine, following CpG stimulation, and regulate T cell responses in vitro through IL-10 secretion [
41,
43,
44]. Further studies would be necessary in order to determine whether higher proportions of transitional B cells causally interfere with immunity to malaria, are a result of inadequate immune control of parasites, or whether expansion of this sub-set is associated with increased parasite densities for other reasons.
Outside of vaccine trials, plasmablasts/plasma cells can be difficult to characterize due to their paucity in the peripheral blood and highly synchronous migration to the bone marrow. Development and migration of PCs has been characterized in experimental infection of mice with
P. chabaudi [
9,
45,
46], but is not well described in humans. Although difficult to assess, these cells provide a snapshot of effector B cell responses formed in response to recent exposure. This study demonstrates that the day a person presents with symptomatic malaria may be an optimum time to characterize plasmablasts/plasma cells in peripheral blood. Given the findings, it may be possible in future studies to assess responding plasmablasts formed in response to acute infection and provide insight into emerging B cell responses to natural infection.
Some unique strengths of this study are the detailed longitudinal follow up of participants, allowing characterization of relevant clinical outcomes and the measurement of numerous B cell sub-type proportions at multiple timepoints during and following an episode of symptomatic malaria. By evaluating proportions of these sub-types at the time of malaria and following treatment, the potential confounding effects of the duration of time since malaria, which can affect the proportions of different cell types, was limited. Similarly, by evaluating children in a very narrow age range, the potential confounding effect of age, which is associated with immunity to malaria and proportions of B cell sub-types, was reduced [
11,
13]. The extensive longitudinal data available on study participants allowed the evaluation and adjustment for the individual’s varied exposure to
P. falciparum, an important consideration when evaluating immunity.
A limitation of this study is the inability to determine whether associations between clinical phenotypes and proportions of B cell sub-sets represent a direct effect of these cell types. In addition, evaluation was limited to B cell sub-sets trafficking through peripheral blood which, while clearly an important compartment for interaction with a blood stage pathogen, may not represent proportions in other important compartments such as secondary lymphoid organs or peripheral tissues. There is evidence using experimental mouse infection with
P. chabaudi that memory B cell responses circulating in the blood reflected the general composition of B cells in peripheral tissues [
26]. Since peripheral lymphoid tissues are not easily sampled, understanding circulating responses may provide the closest accessible insights to general B cell responses and homeostasis in humans. Another caveat to this study is that increased or decreased B cell sub-set proportions could be a result of perturbations in another B cell sub-set, e.g., observed decreases in circulating naïve B cell proportions could be the result of increases in transitional cell proportions. Consistent with methods from several earlier prior studies [
11,
13,
14,
22,
47], the design of this study only allowed for measurement of relative proportions and not absolute B cell counts; future studies may want to account for changes in absolute numbers of cells as well as proportions. Additional studies focusing on the mechanisms of action of atypical MBCs and transitional B cells are also needed to provide insight into the causal relationships underlying the associations observed.
Authors’ contributions
All authors contributed to the study design (RS, PJ, JT, HMZ, MM, EA, MK, GD, MEF, EMR, CJD, and BG), implementation (RS, IS, SW, FN, PJ, GD, MEF, and BG), and/or processing of the data collected (RS, IS, PJ, and BG), as well as in the preparation of this manuscript. All authors read and approved the final manuscript.