Epidemiological observations in malaria endemic areas have long suggested a deficiency in the generation and maintenance of B cell memory to Plasmodium falciparum (Pf) in individuals chronically reinfected with the parasite. Recently, a functionally and phenotypically distinct population of FCRL4+ hyporesponsive memory B cells (MBCs) was reported to be expanded in HIV-infected individuals with high viral loads. In this study, we provide evidence that a phenotypically similar atypical MBC population is significantly expanded in Pf-exposed Malian adults and children as young as 2 years of age as compared with healthy U.S. adult controls. The number of these atypical MBCs was higher in children with chronic asymptomatic Pf infections compared with uninfected children, suggesting that the chronic presence of the parasite may drive expansion of these distinct MBCs. This is the first description of an atypical MBC phenotype associated with malaria. Understanding the origin and function of these MBCs could be important in informing the design of malaria vaccines.

The most deadly of malaria parasites, Plasmodium falciparum (Pf),3 imposes an enormous disease burden on much of the world’s population (1). Malaria kills nearly one million children each year in Africa alone, and there is little doubt that a malaria vaccine would play a central role in preventing these deaths. At present, the human immune response to this pathogen is poorly understood, and the development of a highly effective vaccine would, in all likelihood, benefit from a better understanding of the interactions of this complex pathogen with the human immune system (2). Abs have been shown to have a crucial role in controlling the blood stage of Pf infection. The transfer of Abs from malaria-immune adults to children with malaria resulted in a significant decrease in parasite levels and disease (3). However, epidemiological observations have long suggested that clinical immunity to malaria is slow to develop and short-lived (2). IgG Ab responses specific for Pf Ags are often inconsistently generated, unexpectedly short-lived, and fail to consistently boost upon reinfection (4). Earlier analyses of Pf-specific memory B cells (MBCs) in adults and children in Kenya provided evidence that not all exposures to malaria result in the generation of MBCs (5). Recently, an increase in the total number of MBCs and a decrease in the number of naive B cells in peripheral blood were reported for children with acute malaria (6). Collectively, these observations suggest a deficiency in the generation and long-term maintenance of MBCs specific for Pf. An understanding of the cellular mechanisms responsible for this deficiency would inform the design of malaria vaccines that go beyond the traditional empiric approach and address Pf-specific modulation of the immune response.

Immunological memory, the ability to respond more rapidly and robustly to re-exposure to an Ag, is a hallmark of adaptive immunity. For Ab responses, memory is encoded, in part, in long-lived MBCs (7). Most MBCs in humans express CD27, a member of the TNF family, and have somatically mutated V genes and switched Ig isotypes. Ehrhardt et al. (8) described a morphologically and functionally distinct human MBC population in tonsil defined by the expression of FCRL4, a member of a recently identified family of FcR-like proteins. Because the intracellular domain of FCRL4 contains three immunoreceptor-based inhibition motifs, it is a potential B cell inhibitory receptor, and recent studies in vitro showed that a chimeric protein containing the intracellular domain of FCRL4 and the extracellular domain of FcγRIIB blocked B cell activation when coligated to the BCR (9). The expression of the classical marker for human MBCs, CD27, is much reduced on these FCRL4+ MBCs, but these B cells have undergone isotype switching and somatic hypermutation. FCRL4+ MBCs were found almost exclusively in lymphoid tissues near epithelial surfaces. This MBC population expressed the activation markers CD69, CD80, and CD86 and was functionally distinct from CD27+ FCRL4 MBCs, because FCRL4+ MBCs proliferated and secreted high levels of Igs in response to cytokines and CD40L, but failed to proliferate in response to BCR cross-linking or treatment with Staphylococcus aureus Cowan (SAC). Recent transcriptome analyses of FCRL4+ and FCRL4 MBCs showed that these two populations differentially express genes in several categories, including cell cycle regulators, adhesion molecules, homing receptors, and signal transduction intermediates (10). Although a distinct function has not yet been attributed to FCRL4+ MBCs in vivo, their exclusive location in epithelial-associated lymphoid tissues and their activated phenotype suggest that they may play a role in musosal defense against invading pathogens.

Recently, Moir et al. (11) showed that in the peripheral blood of HIV patients with high viremia, an atypical population of memory B cells (CD20high/CD27/CD21low) with increased expression of FCRL4 was greatly expanded, representing on average 19% of total peripheral blood B cells, compared with less than 4% in healthy individuals. These atypical MBCs in HIV-infected individuals had undergone somatic hypermutation and class switching albeit to lower levels as compared with CD27+ MBCs. Compared with naive B cells and classical MBCs, the atypical MBCs in the peripheral blood of HIV-infected individuals proliferated less to BCR cross-linking and/or CD40L and the TLR agonist, CpG, and showed a decreased ability to differentiate into Ab-secreting cells in response to CpG and SAC. The atypical MBCs in HIV-viremic individuals expressed relatively high levels of inhibitory receptors and a profile of homing receptors similar to that described for tissue-based FCRL4+ MBCs (10, 11) and for exhausted CD8+ T cells during chronic viral infection (12). Because of the overall hyporesponsiveness of these atypical MBCs, their altered expression of inhibitory and homing receptors that together are signatures for virus-induced exhaustion of T cells (12, 13, 14), Moir et al. (11) coined these atypical MBCs “exhausted MBCs.” HIV-specific MBCs were found to be increased in the exhausted MBC compartment as compared with the classical MBC compartment; in contrast, influenza-specific MBCs were more prevalent in the classical MBC compartment. Importantly, exhausted MBCs were found in normal levels in peripheral blood of individuals treated to reduce viremia. These authors proposed that chronic HIV stimulation of B cells may lead to their premature exhaustion, contributing to the poor Ab responses in HIV-infected individuals.

In this study, we report that B cells phenotypically and functionally similar to exhausted MBCs are expanded in individuals chronically exposed to Pf. We propose that the high prevalence of MBCs of this atypical population may contribute to the delayed acquisition and short-lived nature of malaria B cell immunity.

Details of the study site and cohort have been published elsewhere (15). In May 2006, just before the malaria season, 225 individuals between the ages of 2 and 25 years were enrolled in an observational cohort study in the rural village of Kambila, Mali. Individuals were invited to participate after random selection from an age-stratified census of 1500 inhabitants. Enrollment exclusion criteria were hemoglobin <7 g/dL, fever ≥37.5°C, acute systemic illness, use of antimalarial or immunosuppressive medications in the past 30 days, or pregnancy. Of note, asymptomatic Pf parasitemia at enrollment was not exclusionary, and was not treated with antimalarial drugs. For this analysis, an age-stratified subset (n = 87) was randomly selected from the study cohort. Pf transmission is intense at this site and typically begins in June, peaks in October, and ends in December (16). During the 8-mo study period that pertains to this analysis (May–December 2006), subjects were instructed to report symptoms of malaria at the village health center, staffed 24 h per day by a study physician (i.e., passive malaria surveillance). From those with signs or symptoms of malaria, thick blood smears were stained with Giemsa and counted against 300 leukocytes. Slide-positive patients were treated with a standard 3-day course of artesunate plus amodiaquine. Children with severe malaria were referred to the District Hospital after an initial parenteral dose of quinine. For data analysis, malaria was defined as an axillary temperature ≥37.5°C, Pf asexual parasitemia ≥5000/μl, and a nonfocal physical examination by the study physician. For each study participant, the malaria incidence and the time to the first malaria episode were determined. PBMCs were isolated from venous blood collected before (May 2006) and at the peak of malaria transmission (October 2006) and stored in liquid nitrogen. Stool and urine were examined at enrollment for the presence of helminth infections. Hemoglobin was typed by HPLC (D-10 instrument; Bio-Rad). Peripheral blood samples from 10 healthy, anonymous, adult blood bank donors in the United States were also analyzed. Travel history for these individuals was not available, but prior exposure to Pf is unlikely.

This study was approved by the Ethics Committee at the Faculty of Medicine, Pharmacy and Odonto-Stomatology, and the Institutional Review Board at the National Institute of Allergy and Infectious Diseases. Informed consent was obtained from adult participants, and from the parents or guardians of participating children.

Thick blood smears were stained with Giemsa and counted against 300 leukocytes. Pf densities were recorded as the number of asexual parasites/μl whole blood, based on an average leukocyte count of 7500/μl. Each smear was evaluated separately by two expert microscopists blinded to the clinical status of study participants. Any discrepancies were resolved by a third expert microscopist.

At enrollment, duplicate stool samples were examined for Schistosoma mansoni eggs and other intestinal helminths using the semiquantitative Kato-Katz method. To detect Schistosoma hematobium eggs, 10 ml of urine was poured over Whatman filter paper. One or two drops of ninhydrine were placed on the filter and left to air dry. After drying, the filter was dampened with tap water and helminth eggs detected by microscopy.

Data were analyzed using STATA software (StataCorp LP, 2007, Release 10.0). The nonparametric Wilcoxon rank-sum and Wilcoxon matched pairs tests were used to compare continuous variables between unpaired and paired groups, respectively. Age-adjusted Cox regression, logistic regression, and Poisson regression models were used to assess the association between the percentage of atypical MBC and malaria risk. For all tests, two-tailed p values were considered significant if ≤0.05.

PBMCs were obtained by density-gradient centrifugation. All phenotypic analyses were performed using mouse mAbs specific for human B cell markers conjugated to fluorophores. The source of mAbs specific for the following markers conjugated to fluorophores is as follows: PECy7-CD19, PE-CD20, allophycocyanin-CD10, allophycocyanin-CD27, and PE-IgG, BD Biosciences; FITC-CD21, Beckman Coulter; PE-CD85j, PE-CXCR3, PE-CCR6, PE-CCR7, PE-CXCR4, and PE-CXCR5, R&D Systems; and PE-CD11c, Invitrogen. The FCRL4-specific mAb was provided by M. Cooper (Emory University School of Medicine, Atlanta, GA) (8). PE-conjugated rabbit Abs specific for mouse IgG2a were purchased (Invitrogen) and used to detect the mouse FCRL4-specific mAb. A four-color two-stain strategy was used to identify B cell subpopulations (stain 1: FITC-CD21, PE-CD20, PECy7-CD19, and allophycocyanin-CD10; stain 2: FITC-CD21, PE-IgG, PECy7-CD19, and allophycocyanin-CD27). Using this strategy, we report naive B cells as the number of CD19+CD21+CD27 cells from which was subtracted the number of CD19+CD21+ cells that were CD10+, namely immature cells. We determined that naive germinal center cells, CD19+CD21+CD10+CD27+, constitute ∼2% of CD19+ cells (S. Moir, unpublished observations), and this 2% was added to the number of naive B cells. Plasma cells/blasts are reported as the number of CD19+CD21CD20CD10 cells. Immature B cells are reported as the number of CD19+CD10+ cells. Classical MBCs are reported as the number of CD19+CD27+CD21+ cells. Atypical MBCs are reported as the number of CD19+CD21CD27 cells minus the number of CD19+CD21CD10+ immature B cells. Activated MBCs are reported as the number of CD19+CD21CD27+ cells minus the number of CD19+CD21CD10CD20 plasma cells. FACS analyses were performed on a FACSCalibur flow cytometer (BD Biosciences) using FlowJo software (Tree Star).

Mature (CD10) B cells were isolated from PBMCs by negative magnetic bead-based selection using a B cell enrichment mixture supplemented with tetrameric CD10-specific mAb (StemCell Technologies). Mature B cells were separated into CD19+, CD27+/CD21high, and CD27/CD21low fractions using a two-step magnetic bead selection process, as detailed elsewhere (11). The subpopulations were cultured, as previously described (17), in complete medium alone or complete medium plus a mixture of polyclonal activators, which included 2.5 μg/ml CpG oligonucleotide ODN-2006 (18) from Eurofins MWG/Operon, protein A from SAC at 1/10,000 dilution from Sigma-Aldrich, pokeweed mitogen at 1/100,000 dilution from Sigma-Aldrich, and IL-10 at 25 ng/ml from BD Biosciences. Cells were kept at 37°C in a 5% CO2 incubator for 5 days, washed twice with complete medium warmed to 37°C, counted, and distributed onto 96-well ELISPOT plates coated with human IgG-specific goat Abs to detect all IgG-secreting cells, as described (19).

To determine the impact of Pf infection on the generation and maintenance of B cell memory, we initiated a longitudinal study of children and adults in Mali in May 2006, just before the 6-mo malaria transmission season, as described earlier (15). Baseline characteristics of the study subjects are shown in Table I. Several of the variables listed are known to be associated with decreased risk of malaria, including age (20), RBC polymorphisms (21), and asymptomatic parasitemia (22). Of the variables listed, we determined that only three were associated with decreased risk of clinical malaria in our Mali cohort, namely greater age, sickle cell trait (hemoglobin AS), and asymptomatic Pf parasitemia at the time of enrollment, just before the malaria season (15). Asymptomatic parasitemia at the time of enrollment implies that the individual has been stably infected with Pf parasites for at least 6 mo because there is little or no malaria transmission during the 6-mo dry season before enrollment (16). Malaria outcomes by age group are shown in Table II and, as noted, risk of clinical malaria decreased with age, as measured by either malaria incidence or time to the first malaria episode.

Table I.

Baseline characteristics by age group

Age Group, YearsAll (n = 87)
2–4 (n = 35)5–7 (n = 11)8–10 (n = 25)18–25 (n = 16)
Gender, % female (no.) 71.4 (25) 54.6 (6) 32.0 (8) 50.0 (8) 54.0 (47) 
Ethnicity, % (no.)      
 Bambara 62.9 (22) 27.3 (3) 56.0 (14) 68.8 (11) 57.5 (50) 
 Sarakole 31.4 (11) 72.7 (8) 28.0 (7) 25.0 (4) 34.5 (30) 
 Fulani 2.9 (1) 0.0 (0) 12.0 (3) 6.3 (1) 5.7 (5) 
 Malinke 2.9 (1) 0.0 (0) 4.0 (1) 0.0 (0) 2.3 (2) 
Hemoglobin AS, % (no.)a 15.2 (5) 0.0 (0) 4.0 (1) 12.5 (2) 9.4 (8) 
P. falciparum smear positive at enrollment, % (no.)b 8.6 (3) 18.2 (2) 16.0 (4) 12.5 (2) 12.6 (11) 
Parasitemia if smear positive at enrollment, parasites/μl (geometric mean (95% CI)) 952 (33–27,878) 2,862 (155–52,838) 426 (80–2,265) 491 (0.31–765,982) 770 (356–1,664) 
GI helminth, % positive at enrollment (no.)c 12.9 (4) 18.2 (2) 9.1 (2) 0 (0) 10.5 (8) 
Urine schistosomiasis, % positive at enrollment (no.)d 0 (0) 0 (0) 8.3 (2) 41.7 (5) 9.5 (7) 
Distance lived from clinic, meters (mean ± SD) 397.4 (±125) 334 (±64) 385 (±85) 386 (±74) 384 (±99) 
Bed net use, % (no.)e 29.0 (9) 30.0 (7) 16.7 (4) 33.3 (5) 28.7 (25) 
Age Group, YearsAll (n = 87)
2–4 (n = 35)5–7 (n = 11)8–10 (n = 25)18–25 (n = 16)
Gender, % female (no.) 71.4 (25) 54.6 (6) 32.0 (8) 50.0 (8) 54.0 (47) 
Ethnicity, % (no.)      
 Bambara 62.9 (22) 27.3 (3) 56.0 (14) 68.8 (11) 57.5 (50) 
 Sarakole 31.4 (11) 72.7 (8) 28.0 (7) 25.0 (4) 34.5 (30) 
 Fulani 2.9 (1) 0.0 (0) 12.0 (3) 6.3 (1) 5.7 (5) 
 Malinke 2.9 (1) 0.0 (0) 4.0 (1) 0.0 (0) 2.3 (2) 
Hemoglobin AS, % (no.)a 15.2 (5) 0.0 (0) 4.0 (1) 12.5 (2) 9.4 (8) 
P. falciparum smear positive at enrollment, % (no.)b 8.6 (3) 18.2 (2) 16.0 (4) 12.5 (2) 12.6 (11) 
Parasitemia if smear positive at enrollment, parasites/μl (geometric mean (95% CI)) 952 (33–27,878) 2,862 (155–52,838) 426 (80–2,265) 491 (0.31–765,982) 770 (356–1,664) 
GI helminth, % positive at enrollment (no.)c 12.9 (4) 18.2 (2) 9.1 (2) 0 (0) 10.5 (8) 
Urine schistosomiasis, % positive at enrollment (no.)d 0 (0) 0 (0) 8.3 (2) 41.7 (5) 9.5 (7) 
Distance lived from clinic, meters (mean ± SD) 397.4 (±125) 334 (±64) 385 (±85) 386 (±74) 384 (±99) 
Bed net use, % (no.)e 29.0 (9) 30.0 (7) 16.7 (4) 33.3 (5) 28.7 (25) 
a

Data available for 85 subjects.

b

All subjects were asymptomatic at enrollment.

c

Data available for 76 subjects; GI = gastrointestinal.

d

Data available for 74 subjects.

e

Nightly bednet use self-reported at the end of the malaria season.

Table II.

Malaria outcomes over the 8-mo study period, by age group

Age Group, YearsAll (n = 87)
2–4 (n = 35)5–7 (n = 11)8–10 (n = 25)18–25 (n = 16)
Malaria incidence, mean (±SD), no. of episodesa 2.11 (±1.25) 2.09 (±0.83) 1.00 (±1.22) 0.12 (±0.40) 1.44 (±1.32) 
Severe malaria, no. of episodesb 
At least one malaria episode, % (no.) 88.6 (31) 100.0 (11) 56.0 (14) 18.8 (3) 67.8 (59) 
Time to first malaria episode, days since enrollment (median)c 108 128 155 242 136 
Parasitemia at first malaria episode, parasites/μl (geometric mean (95% CI)) 40,457 (28,026–58,404) 10,810 (1,293–90,407) 11,697 (4,618–29,630) 6,949 (5,329–9,062) 20,586 (13,157–32,211) 
Age Group, YearsAll (n = 87)
2–4 (n = 35)5–7 (n = 11)8–10 (n = 25)18–25 (n = 16)
Malaria incidence, mean (±SD), no. of episodesa 2.11 (±1.25) 2.09 (±0.83) 1.00 (±1.22) 0.12 (±0.40) 1.44 (±1.32) 
Severe malaria, no. of episodesb 
At least one malaria episode, % (no.) 88.6 (31) 100.0 (11) 56.0 (14) 18.8 (3) 67.8 (59) 
Time to first malaria episode, days since enrollment (median)c 108 128 155 242 136 
Parasitemia at first malaria episode, parasites/μl (geometric mean (95% CI)) 40,457 (28,026–58,404) 10,810 (1,293–90,407) 11,697 (4,618–29,630) 6,949 (5,329–9,062) 20,586 (13,157–32,211) 
a

Malaria episode defined as temperature > 37.5°C, asexual parasitemia > 5000/μl, and nonfocal physical examination.

b

World Health Organization definition of severe malaria (1).

c

All study participants were enrolled during a 2-wk period just prior to the 6-mo malaria season.

The B cells in the peripheral blood of the volunteers at the peak of the malaria transmission season in October were characterized by flow cytometry using a panel of fluorophore-conjugated Abs specific for CD19, CD27, CD21, CD10, and CD20 that allowed the identification of naive B cells (CD19+CD27CD21+CD10), plasma cells or plasma blasts (CD19+CD27+CD21CD20), immature B cells (CD19+CD10+), classical MBCs (CD19+CD27+CD21+CD10), atypical MBCs (CD19+CD27CD21CD20+CD10), and activated MBCs (CD19+CD27+CD21CD20+CD10). Fig. 1 shows the gating strategy used to identify these B cell subpopulations and a representative example of an individual from the United States as compared with an individual in Mali.

FIGURE 1.

Flow cytometry gating strategies for B cell phenotyping. FACS plots of B cell subsets of a representative malaria naive U.S. and a Pf-exposed Malian volunteer. Within the CD19+ gate, the B cell subpopulations are defined as follows: activated memory (orange) (CD27+CD21CD20+CD10), classical memory (red) (CD27+CD21+), atypical memory (blue) (CD27CD21CD20+CD10), naive (green) (CD27CD21+CD10), immature (purple) (CD10+), and plasma cells or plasma blasts (pink) (CD27+CD21CD20). A four-color two-stain strategy was used as detailed in Materials and Methods to quantify the number of B cells in each subpopulation.

FIGURE 1.

Flow cytometry gating strategies for B cell phenotyping. FACS plots of B cell subsets of a representative malaria naive U.S. and a Pf-exposed Malian volunteer. Within the CD19+ gate, the B cell subpopulations are defined as follows: activated memory (orange) (CD27+CD21CD20+CD10), classical memory (red) (CD27+CD21+), atypical memory (blue) (CD27CD21CD20+CD10), naive (green) (CD27CD21+CD10), immature (purple) (CD10+), and plasma cells or plasma blasts (pink) (CD27+CD21CD20). A four-color two-stain strategy was used as detailed in Materials and Methods to quantify the number of B cells in each subpopulation.

Close modal

A comparison of the proportion of B cells in each subpopulation in the peripheral blood of 10 healthy U.S. adults with 87 Malian adults and children is given (Figs. 2, A–F) as a percentage of total CD19+ B cells. The percentage of CD19+ B cells per PBMC did not vary significantly between individuals before and at the peak of the malaria transmission season (before season, mean 12.16% (95% confidence interval (CI), 11.36–12.97) vs peak season, mean 11.71% (95% CI, 10.86–12.57); p = 0.451). The relative proportions of all the B cell subpopulations analyzed per total B cells for each age group are also shown as stacked plots in Fig. 2,G. The percentage of naive B cells (Fig. 2,A) was similar in both U.S. and Malian adults and children, as was the percentage of plasma cells/plasma blasts (Fig. 2,B). The proportion of immature B cells appeared to be lower in Malian adults as compared with either U.S. volunteers or Malian children (Fig. 2,C), although a more complete phenotypic analysis will be necessary to better characterize these cells. As compared with U.S. adults, Malian adults had a similar percentage of classical MBCs (Fig. 2,D). Malian children in both the 2- to 7-year and 8- to 10-year age groups had a smaller percentage of classical MBCs as compared with either U.S. or Malian adults, most likely an age-related phenomenon unrelated to malaria. Strikingly, the percentage of atypical MBCs was significantly higher in Malian adults and children compared with U.S. adults, with a trend of increasing atypical MBCs with increasing age among the Malian donors (Malian adults, mean 15.5% (95% CI, 9.7–21.2) vs U.S. adults, mean 1.6% (95% CI, 1.0–2.2), p < 0.001; Malian children, mean 9.8% (95% CI, 8.2–11.3), p < 0.001, vs U.S. adults) (Fig. 2,E). The percentage of B cells with an activated MBC phenotype was significantly higher in Malian adults as compared with Malian children and U.S. adults (Malian adults, mean 3.7% (95% CI, 2.5–5.0) vs U.S. adults, mean 1.3% (95% CI, 0.6–2.1), p = 0.001; Malian children, mean 1.9% (95% CI, 1.4–2.4), p < 0.001 vs Malian adults) (Fig. 2 F). As with the atypical MBCs, there was a trend of increased activated MBCs with increased age among the Malian donors. For both atypical MBCs and activated MBCs, the largest differences were between U.S. and Malian adults. We verified that the expansion of atypical MBCs in this population was not an artifact of freezing and thawing by analyzing fresh PBMC from 16 adults at the same study site (mean percentage of atypical MBCs from fresh PBMC, 14.1% (95% Cl, 8.1–20.1)).

FIGURE 2.

Atypical MBCs are significantly increased in Malian as compared with U.S. volunteers. A–F, The percentage of B cell subsets was determined by flow cytometry with phenotypic analysis of subsets, as defined in Fig. 1 and detailed in Materials and Methods. B cell subpopulations are expressed as a percentage of total CD19+ B cells for U.S. adults (n = 10), Malian adults (n = 14), and Malian children aged 8–10 (n = 25) and 2–7 (n = 36) years. The Wilcoxon rank-sum test was used to compare continuous variables between groups. G, The relative proportions of all the B cell subpopulations as analyzed per total CD19+ B cells for each age group are given in stacked plots.

FIGURE 2.

Atypical MBCs are significantly increased in Malian as compared with U.S. volunteers. A–F, The percentage of B cell subsets was determined by flow cytometry with phenotypic analysis of subsets, as defined in Fig. 1 and detailed in Materials and Methods. B cell subpopulations are expressed as a percentage of total CD19+ B cells for U.S. adults (n = 10), Malian adults (n = 14), and Malian children aged 8–10 (n = 25) and 2–7 (n = 36) years. The Wilcoxon rank-sum test was used to compare continuous variables between groups. G, The relative proportions of all the B cell subpopulations as analyzed per total CD19+ B cells for each age group are given in stacked plots.

Close modal

To assess class-switching in the MBC subpopulations, we analyzed cell surface IgG expression of atypical and classical MBCs in our Malian cohort. Overall, the pattern of IgG expression was similar for classical and atypical MBCs (Fig. 3), with the proportion of IgG+ MBCs increasing with age in both classical and atypical MBCs. The proportion of atypical MBCs that were IgG+ was somewhat higher than the proportion of classical MBCs that were IgG+. We conclude that the istotype-switching history is similar in the different MBC subpopulations and that the high proportion of IgG+ atypical MBCs indicates that it is likely these cells have undergone somatic hypermutation.

FIGURE 3.

The IgG expression of atypical and classical MBCs is similar. Stacked plots showing the proportion of IgG+ and IgG MBCs for both the classical MBC subpopulation and the atypical MBC subpopulation. The phenotyping was as in Fig. 1, detailed in Materials and Methods.

FIGURE 3.

The IgG expression of atypical and classical MBCs is similar. Stacked plots showing the proportion of IgG+ and IgG MBCs for both the classical MBC subpopulation and the atypical MBC subpopulation. The phenotyping was as in Fig. 1, detailed in Materials and Methods.

Close modal

The atypical MBCs, classical MBCs, and naive B cells from peripheral blood of a randomly selected subset of Malian adults (n = 6) and children aged 2–4 years (n = 6) were further analyzed to determine the expression level of several inhibitory and homing receptors that are characteristic of both tissue-based (10) and exhausted MBCs (11) (Fig. 4). FCRL4, the cell surface marker that defines tissue-based MBCs and is a characteristic of exhausted MBCs, was expressed at significantly higher levels on atypical MBCs compared with classic MBCs and naive B cells. The expression pattern of inhibitory and homing receptors on atypical MBCs was similar for Malian children and adults, and comparable to that observed in HIV viremic individuals and in FCRL4+ tonsilar MBCs (8, 10, 11). Atypical MBCs showed increased expression of the inhibitory receptors CD85j and CD22. No differences were observed between activated MBCs and atypical MBCs in the expression of the inhibitory receptor LAIR1, although as compared with naive B cells both subpopulations expressed less LAIR1. CD11c and CXCR3 levels were increased on atypical MBCs as compared with either classic MBCs or naive B cells. CXCR4, CD62L, CXCR5, and CCR7 expression was decreased on atypical MBCs, and little difference between subpopulations was observed in the expression of CCR6 (Fig. 4) or CD72 (data not shown). The phenotypes of both the naive B cells and classical MBCs were similar to that described for U.S. individuals.

FIGURE 4.

Inhibitory and tissue-homing receptor expression is increased and lymph node-homing receptor expression is decreased on atypical MBCs relative to classical MBCs. FACS analysis of the expression of inhibitory and homing receptors on naive B cells (green), atypical MBCs (blue), and classical MBCs (red) on a subset of 12 Malian individuals: 6 children aged 2–4 years and 6 adults. For each panel (A–F), the top plots show individual mean fluorescence intensity (MFI) values for each cell subpopulation of each individual, as well as the average MFI and SD. Underneath are histograms of the MFI from a representative individual of each subpopulation. The expression of inhibitory receptors (A–D), tissue-homing receptors (E–H), and lymph node-homing receptors (I–K) is given for classical MBCs, atypical MBCs, and naive B cells (as defined in Fig. 1) using appropriately labeled Abs specific for CD19, CD27, CD21, and the particular inhibitory and homing receptors indicated. The data for different subsets of B cells from individual donors were paired, and the Wilcoxon matched pair test was used for the comparison.

FIGURE 4.

Inhibitory and tissue-homing receptor expression is increased and lymph node-homing receptor expression is decreased on atypical MBCs relative to classical MBCs. FACS analysis of the expression of inhibitory and homing receptors on naive B cells (green), atypical MBCs (blue), and classical MBCs (red) on a subset of 12 Malian individuals: 6 children aged 2–4 years and 6 adults. For each panel (A–F), the top plots show individual mean fluorescence intensity (MFI) values for each cell subpopulation of each individual, as well as the average MFI and SD. Underneath are histograms of the MFI from a representative individual of each subpopulation. The expression of inhibitory receptors (A–D), tissue-homing receptors (E–H), and lymph node-homing receptors (I–K) is given for classical MBCs, atypical MBCs, and naive B cells (as defined in Fig. 1) using appropriately labeled Abs specific for CD19, CD27, CD21, and the particular inhibitory and homing receptors indicated. The data for different subsets of B cells from individual donors were paired, and the Wilcoxon matched pair test was used for the comparison.

Close modal

Exhausted MBCs in HIV-viremic individuals were hyporesponsive in their ability to differentiate into Ab-secreting cells (ASC) in vitro in response to polyclonal stimulation with a combination of CpG and SAC (11). In preliminary studies, we separated peripheral blood B cells from Malian adults into atypical MBCs (CD19+CD27CD21low), classical MBCs (CD19+CD27+CD21high), and naive B cells (CD19+CD27CD21high), as described (11). When stimulated with CpG and SAC, atypical MBCs failed to produce any ASCs (n = 1; classic MBC, 34,900/106 at end of culture (SD ± 2,000); atypical MBC, 0/106; naive B cells, 1300/106 (SD ± 200)). As compared with classical MBCs, atypical MBCs responded poorly to the combination of pokeweed mitogen, SAC, CpG, and IL-10, a combination that is more efficient in inducing the differentiation of MBCs into ASCs (G. E. Weiss, unpublished data) (n = 2; classic MBC, 102,800/106 at end of culture (SD ± 18,200); atypical MBC, 2,600/106 (SD ± 1,200); naive B cells, 8,250/106 (SD ± 2,000)). Naive B cells, as predicted (7), responded weakly to both stimulation mixtures. By this criteria, the atypical MBCs in Malian volunteers are hyporesponsive to stimuli that activate classical MBCs, and thus both phenotypically and functionally resemble exhausted MBCs.

To determine whether the increase in atypical MBCs is related to exposure to Pf, we compared the percentage of atypical MBCs in children with or without asymptomatic Pf parasitemia at the end of the 6-mo dry season, during which little or no parasite transmission occurs (16). Pf parasitemia at this timepoint reflects a chronic infection persisting from the previous year’s transmission season; conversely, aparasitemic individuals at this time have most likely been aparasitemic for months (16). The percentage of asymptomatic parasitemic individuals was similar across all age groups and ranged between 2 and 5%. There was a trend toward a higher percentage of atypical MBCs in children aged 2–10 years with asymptomatic Pf infection (n = 9) as compared with those without infection (n = 62) (Fig. 5,A; with Pf parasitemia, mean 14.7% (95% CI, 2.7–21.0); without Pf parasitemia, mean 9.9% (95% CI, 8.1–11.7); p = 0.055). In multivariate regression analysis that included age as a covariate, this association did not reach statistical significance. Intestinal helminth infection was not associated with a significant change in the percentage of atypical MBCs in children aged 2–10 years (Fig. 5 B); with helminth infection, n = 8, mean 13.4% (95% CI, 5.2–21.6); without helminth infection, n = 56, mean 9.9% (95% CI, 8.1–11.7); p = 0.291, although the sample size may be too small to detect a significant difference. Of note, neither the percentage of atypical MBCs before the malaria season nor the percentage at the peak of the malaria season was associated with risk of clinical malaria, as defined by malaria incidence or time to the first malaria episode (data not shown).

FIGURE 5.

The percentage of atypical MBCs is larger in children with persistent asymptomatic Pf parasitemia as compared with parasite-free children. Shown is the percentage of atypical MBCs per total B cells in children aged 2–10 years (A) with (n = 9) or without (n = 62) Pf parasitemia at the end of the dry season, or B, with (n = 8) or without (n = 56) helminth infection. The Wilcoxon rank-sum test was used to compare continuous variables between groups.

FIGURE 5.

The percentage of atypical MBCs is larger in children with persistent asymptomatic Pf parasitemia as compared with parasite-free children. Shown is the percentage of atypical MBCs per total B cells in children aged 2–10 years (A) with (n = 9) or without (n = 62) Pf parasitemia at the end of the dry season, or B, with (n = 8) or without (n = 56) helminth infection. The Wilcoxon rank-sum test was used to compare continuous variables between groups.

Close modal

The finding of an expanded, atypical MBC subpopulation in Malian individuals is the first description of a phenotypic alteration of MBCs in individuals exposed to Pf. At present, the factors that cause the expansion of the atypical MBCs are not known. In HIV infections, the virus appears to play a role in driving B cells into the exhausted MBC subpopulation. In individuals with untreated HIV infections, the majority of HIV-specific MBCs were present in the exhausted MBC subpopulation (11), and in patients whose viral loads were reduced to levels below detection by antiretroviral therapy, the number of exhausted MBCs decreased to about one-half over a period of 6 mo. However, the exhausted MBC subpopulation in treated patients remained statistically greater than that in healthy donors, presumably due to viral spiking (S. Moir, unpublished observation). Due to the small cell numbers we were able to obtain in our Malian cohort, especially from young children, we were unable to carry out similar assays to directly determine whether Pf-specific MBCs were differentially represented in the atypical MBC population. Finding that persistent Pf infection may be associated with a greater degree of expansion of atypical MBCs suggests that parasite Ags or other parasite products may be responsible for driving the B cells into the atypical MBC subpopulation. Other factors could account for the expanded atypical MBCs in Malian individuals, such as genetic background or environmental factors associated with Pf transmission that were not assessed in our study, for example, malnutrition. Although we did not test for HIV, it is unlikely that HIV is responsible for the expanded atypical MBC compartment in our study population because the prevalence of HIV in Mali is extremely low (1.5%), and based on demographic data we would expect HIV prevalence be lower than the country average in our study population.

The role of atypical MBCs in the context of malaria remains unclear. It has been suggested that FCRL4+ MBCs resident in mucosal lymphoid tissue play a role against invading pathogens, possibly through their influence on other cells, either directly or indirectly through the secretion of cytokines (8, 10). Moir et al. (11) concluded that the HIV-associated exhaustion of B cells may play a role in the diminished HIV-specific Ab responses in infected individuals. By analogy it may be that atypical MBCs in malaria are the end product of a defective pathway that normally functions to yield differentiated MBCs. Indeed, our preliminary results suggested that atypical MBCs in malaria-exposed individuals were hyporesponsive to a combination of polyclonal activators that are known to activate classical MBCs. It is possible that atypical MBCs are responsive to other stimuli and contribute to the short-lived nature of the Ab response observed in malaria-endemic areas (2) by giving rise to short-lived plasma cells. We recently observed that Pf-specific Abs rose significantly during the 6-mo malaria transmission season in the children in the cohort analyzed in this study; however, these Abs were short-lived and were undetectable by the beginning of the next malaria transmission season (P. O. Crompton, unpublished observation). This phenomenon could reflect the activation of atypical MBCs to give rise to short-lived plasma cells or the interference of atypical MBCs in the normal generation of long-lived plasma cells in response to Pf infection.

It is also possible that the expansion of atypical MBCs in Pf infection may in some way benefit the host, reflecting the unique relationship between the parasite and the host that allows the asymptomatic persistence of the parasite within an otherwise functional immune system in individuals who have acquired clinical immunity. To this point, we observed that adults, the majority of whom are immune to clinical malaria, showed more atypical MBCs in peripheral blood as compared with children. Given that the human and Plasmodium genomes have coevolved (24), it is possible the Pf has shaped immune mechanisms, which allows chronic and recurrent infections to occur. Presumably, the persistence of the parasite has some benefit to the host, as indicated, for example, by our finding that asymptomatic Pf infection is associated with protection against clinical disease (15). It is possible that atypical MBCs play a beneficial role in protecting the host from clinical disease by modulating immune responses, for example, through the secretion of cytokines to control inflammation.

It will be important to determine whether individuals with expanded atypical MBCs can be effectively vaccinated to produce long-term Pf-specific memory responses. We have observed that compared with U.S. adults, Malian adults appear to respond less well to the same candidate malaria vaccine in Phase I clinical trials as measured by the generation of Ag-specific MBCs (17) (B. Traore, unpublished observations). Future studies will be needed to determine whether the expansion of atypical MBCs represents a protective response or an immune evasion strategy of Pf, and if the latter, whether it can be overcome by vaccination that specifically addresses this mechanism.

We are grateful to the Malian adults and children of Kambila for their always gracious willingness to participate in this study.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by the Intramural Research Program of the National Institutes of Health, National Institute of Allergy and Infectious Diseases.

3

Abbreviations used in this paper: Pf, Plasmodium falciparum; ASC, Ab-secreting cell; CI, confidence interval; MBC, memory B cell; MFI, mean fluorescence intensity; SAC, Staphylococcus aureus Cowan.

1
Snow, R. W., C. A. Guerra, A. M. Noor, H. Y. Myint, S. I. Hay.
2005
. The global distribution of clinical episodes of Plasmodium falciparum malaria.
Nature
434
:
214
-217.
2
Langhorne, J., F. M. Ndungu, A. M. Sponaas, K. Marsh.
2008
. Immunity to malaria: more questions than answers.
Nat. Immunol.
9
:
725
-732.
3
Cohen, S., I. A. McGregor, S. Carrington.
1961
. Gammaglobulin and acquired immunity to human malaria.
Nature
192
:
733
-737.
4
Bejon, P., J. Mwacharo, O. Kai, T. Mwangi, P. Milligan, S. Todryk, S. Keating, T. Lang, B. Lowe, C. Gikonyo, et al
2006
. A phase 2b randomized trial of the candidate malaria vaccines FP9 ME-TRAP and MVA ME-TRAP among children in Kenya.
PLoS Clin. Trials
1
:
e29
5
Dorfman, J. R., P. Bejon, F. M. Ndungu, J. Langhorne, M. M. Kortok, B. S. Lowe, T. W. Mwangi, T. N. Williams, K. Marsh.
2005
. B cell memory to 3 Plasmodium falciparum blood-stage antigens in a malaria-endemic area.
J. Infect. Dis.
191
:
1623
-1630.
6
Asito, A. S., A. M. Moormann, C. Kiprotich, Z. W. Ng‘ang'a, R. Ploutz-Snyder, R. Rochford.
2008
. Alterations on peripheral B cell subsets following an acute uncomplicated clinical malaria infection in children.
Malar. J.
7
:
238
7
Crotty, S., R. Ahmed.
2004
. Immunological memory in humans.
Semin. Immunol.
16
:
197
-203.
8
Ehrhardt, G. R., J. T. Hsu, L. Gartland, C. M. Leu, S. Zhang, R. S. Davis, M. D. Cooper.
2005
. Expression of the immunoregulatory molecule FcRH4 defines a distinctive tissue-based population of memory B cells.
J. Exp. Med.
202
:
783
-791.
9
Ehrhardt, G. R., R. S. Davis, J. T. Hsu, C. M. Leu, A. Ehrhardt, M. D. Cooper.
2003
. The inhibitory potential of Fc receptor homolog 4 on memory B cells.
Proc. Natl. Acad. Sci. USA
100
:
13489
-13494.
10
Ehrhardt, G. R., A. Hijikata, H. Kitamura, O. Ohara, J. Y. Wang, M. D. Cooper.
2008
. Discriminating gene expression profiles of memory B cell subpopulations.
J. Exp. Med.
205
:
1807
-1817.
11
Moir, S., J. Ho, A. Malaspina, W. Wang, A. C. DiPoto, M. A. O'Shea, G. Roby, S. Kottilil, J. Arthos, M. A. Proschan, et al
2008
. Evidence for HIV-associated B cell exhaustion in a dysfunctional memory B cell compartment in HIV-infected viremic individuals.
J. Exp. Med.
205
:
1797
-1805.
12
Wherry, E. J., S. J. Ha, S. M. Kaech, W. N. Haining, S. Sarkar, V. Kalia, S. Subramaniam, J. N. Blattman, D. L. Barber, R. Ahmed.
2007
. Molecular signature of CD8+ T cell exhaustion during chronic viral infection.
Immunity
27
:
670
-684.
13
Day, C. L., D. E. Kaufmann, P. Kiepiela, J. A. Brown, E. S. Moodley, S. Reddy, E. W. Mackey, J. D. Miller, A. J. Leslie, C. DePierres, et al
2006
. PD-1 expression on HIV-specific T cells is associated with T-cell exhaustion and disease progression.
Nature
443
:
350
-354.
14
Trautmann, L., L. Janbazian, N. Chomont, E. A. Said, S. Gimmig, B. Bessette, M. R. Boulassel, E. Delwart, H. Sepulveda, R. S. Balderas, et al
2006
. Up-regulation of PD-1 expression on HIV-specific CD8+ T cells leads to reversible immune dysfunction.
Nat. Med.
12
:
1198
-1202.
15
Crompton, P. D., B. Traore, K. Kayentao, S. Doumbo, A. Ongoiba, S. A. Diakite, M. A. Krause, D. Doumtabe, Y. Kone, G. Weiss, et al
2008
. Sickle cell trait is associated with a delayed onset of malaria: implications for time-to-event analysis in clinical studies of malaria.
J. Infect. Dis.
198
:
1265
-1275.
16
Dicko, A., A. D. Klion, M. A. Thera, I. Sagara, D. Yalcouye, M. B. Niambele, M. Sogoba, G. Dolo, A. Dao, D. A. Diallo, et al
2004
. The etiology of severe anemia in a village and a periurban area in Mali.
Blood
104
:
1198
-1200.
17
Crompton, P. D., M. Mircetic, G. Weiss, A. Baughman, C. Y. Huang, D. Topham, J. Treanor, I. Sanz, F. E. H. Lee, A. P. Durbin, et al
2009
. The TLR9 ligand CpG promotes the acquisition of Plasmodium falciparum-specific memory B cells in malaria-naive individuals.
J. Immunol.
182
:
3318
-3326.
18
Hartmann, G., R. D. Weeratna, Z. K. Ballas, P. Payette, S. Blackwell, I. Suparto, W. L. Rasmussen, M. Waldschmidt, D. Sajuthi, R. H. Purcell, et al
2000
. Delineation of a CpG phosphorothioate oligodeoxynucleotide for activating primate immune responses in vitro and in vivo.
J. Immunol.
164
:
1617
-1624.
19
Crotty, S., R. D. Aubert, J. Glidewell, R. Ahmed.
2004
. Tracking human antigen-specific memory B cells: a sensitive and generalized ELISPOT system.
J. Immunol. Methods
286
:
111
-122.
20
Marsh, K., S. Kinyanjui.
2006
. Immune effector mechanisms in malaria.
Parasite Immunol.
28
:
51
-60.
21
Aidoo, M., D. J. Terlouw, M. S. Kolczak, P. D. McElroy, F. O. ter Kuile, S. Kariuki, B. L. Nahlen, A. A. Lal, V. Udhayakumar.
2002
. Protective effects of the sickle cell gene against malaria morbidity and mortality.
Lancet
359
:
1311
-1312.
22
Males, S., O. Gaye, A. Garcia.
2008
. Long-term asymptomatic carriage of Plasmodium falciparum protects from malaria attacks: a prospective study among Senegalese children.
Clin. Infect. Dis.
46
:
516
-522.
23
UNAIDS Country Responses Mali http://www.unaids.org/en.
24
Mu, J., J. Duan, K. D. Makova, D. A. Joy, C. Q. Huynh, O. H. Branch, W. H. Li, X. Z. Su.
2002
. Chromosome-wide SNPs reveal an ancient origin for Plasmodium falciparum.
Nature
418
:
323
-326.