Preserved Dendritic Cell HLA-DR Expression and Reduced Regulatory T Cell Activation in Asymptomatic Plasmodium falciparum and P. vivax Infection

Abstract

Clinical illness with Plasmodium falciparum or Plasmodium vivax compromises the function of dendritic cells (DC) and expands regulatory T (Treg) cells. Individuals with asymptomatic parasitemia have clinical immunity, restricting parasite expansion and preventing clinical disease. The role of DC and Treg cells during asymptomatic Plasmodium infection is unclear. During a cross-sectional household survey in Papua, Indonesia, we examined the number and activation of blood plasmacytoid DC (pDC), CD141⁺, and CD1c⁺ myeloid DC (mDC) subsets and Treg cells using flow cytometry in 168 afebrile children (of whom 15 had P. falciparum and 36 had P. vivax infections) and 162 afebrile adults (of whom 20 had P. falciparum and 20 had P. vivax infections), alongside samples from 16 patients hospitalized with uncomplicated malaria. Unlike DC from malaria patients, DC from children and adults with asymptomatic, microscopy-positive P. vivax or P. falciparum infection increased or retained HLA-DR expression. Treg cells in asymptomatic adults and children exhibited reduced activation, suggesting increased immune responsiveness. The pDC and mDC subsets varied according to clinical immunity (asymptomatic or symptomatic Plasmodium infection) and, in asymptomatic infection, according to host age and parasite species. In conclusion, active control of asymptomatic infection was associated with and likely contingent upon functional DC and reduced Treg cell activation.

INTRODUCTION

Plasmodium falciparum and Plasmodium vivax are both major causes of malaria morbidity and mortality in Southeast Asia (1, 2). Clinical disease from these infections, malaria, is characterized by the presence of fever together with detectable Plasmodium parasites in the blood. Asymptomatic parasitemia, the carriage of Plasmodium parasites without clinical disease, is also prevalent in both high- and low-transmission areas, with estimates of 11% in children and adults in Timika, Papua, Indonesia (3) and 47% in Papua New Guinea school children (4). Indeed, in most areas where malaria is endemic, the majority of parasite carriers are asymptomatic (5), and those with gametocytes are a major reservoir for transmission by mosquitoes, contributing to malaria transmission within a population (6, 7).

During acute P. falciparum and P. vivax malaria (8) and during primary prepatent P. falciparum infection (9), blood dendritic cells (DC) are functionally compromised, with an inability to appropriately stimulate cellular immunity. This is paralleled by reduced HLA-DR expression (8, 10). The relative increase in regulatory T (Treg) cells during acute malaria (11, 12) is also thought to inhibit host immunity. Together, Plasmodium modulation of DC and Treg cell function appears to foster an immune suppressive environment in malaria that assists parasite growth and parasite transmission (13). However, in individuals living in areas of malaria endemicity who have asymptomatic parasitemia and clinical immunity, it remains to be determined whether or how DC and Treg cells are modulated by Plasmodium.

Asymptomatic individuals in areas of malaria endemicity have clinical immunity that restricts parasite expansion and prevents clinical disease. Such immunity is not sterilizing, but instead, a state of host tolerance exists whereby parasites persist in the blood without the occurrence of clinical symptoms (14, 15). The development of clinical immunity requires repeated parasite infections (14), and thus, the transmission intensity and parasite diversity in regions of endemicity determine age associations in the time to development of clinical immunity (16).

It is not fully understood which immune cells mediate clinical immunity or whether children and adults respond similarly (14, 15, 17). Research suggests a protective effect both for antibodies (18–21) and effector T cell cytokine responses (particularly gamma interferon [IFN-γ]) (21–24). As the induction and maintenance of both effective B cell (antibody) and T cell (cytokine) responses require functional DC (25) and each may be modulated by Treg cells (26), we sought to examine these cells in both children and adults with patent asymptomatic P. falciparum or P. vivax infection and to compare their responses to those seen with acute uncomplicated P. falciparum or P. vivax malaria. We hypothesized that the appropriate activation of DC and Treg cells would characterize asymptomatic Plasmodium infection.

(This work was presented in part as a poster at the 13th International Symposium on Dendritic Cells [DC2014], Tours, France.)

MATERIALS AND METHODS

Study participants and sample collection.

This study was conducted in Timika, in Papua, Indonesia, which has equal prevalence of P. falciparum and P. vivax, unstable malaria transmission, and asymptomatic parasitemia and acute uncomplicated malaria occurring in both children and adults (3). Timika thus represents a unique location to evaluate DC and Treg cell activation in clinical immunity (asymptomatic infection) and immunopathology (acute malaria) across different ages. As part of a household survey conducted between April and July in 2013, 168 children (2 to 10 years) and 162 adults (16 to 40 years) were evaluated using field-based flow cytometry of peripheral blood. All participants were residents in the Timika district, with similar proportions of Papuans (49% of children and 40% of adults) and non-Papuan migrants (51% of children and 60% of adults) from surrounding provinces. The participants were asymptomatic, with neither fever nor other symptoms of malaria at recruitment or in the 24 h prior to recruitment.

Ten milliliters of venous blood was collected from a single adult per household (i.e., 162 households), and finger prick capillary blood was collected from children into lithium heparin tubes. Among the 168 children (2 to 10 years) who provided blood samples for this immunology study, there was an average of 1.24 children per household. Blood samples were transported to the laboratory at room temperature for microscopic parasite detection and flow cytometry, which were performed within 6 h of blood collection. Thick and thin blood smears stained with 5% Giemsa solution were examined by trained microscopists to identify Plasmodium species and parasitemia. Individuals found to be parasitemic were treated according to standard local antimalarial treatment protocols, comprising dihydroartemisinin-piperaquine for 3 days for each species plus, if not G6PD deficient, a 14-day course of primaquine for P. vivax.

In parallel with the household survey, 14 adults and 2 children being treated for acute uncomplicated malaria in Rumah Sakit Mitra Masyarakat Hospital (RSMM) in Timika were enrolled as a comparator group. Malaria was defined by a temperature of >37°C or history of fever with clinical malaria symptoms, a blood film positive for Plasmodium, and no alternative clinical explanation. Venous blood was collected and processed identically as in the household survey samples.

Flow cytometric analysis.

To examine DC, 100 to 300 μl of whole blood was stained with different flow cytometry panels, comprising lineage markers anti-CD3 antibody (clone HIT3a), anti-CD14 antibody (clone HCD14), anti-CD19 antibody (clone H1H9), and anti-CD56 antibody (clone HCD56) conjugated to Alexa Fluor 488 (AF488), anti-CD1c antibody (clone L161) conjugated to phycoerythrin (PE), anti-CD303 antibody (clone 201A) conjugated to peridinin chlorophyll a protein-cyanine 5.5 (PerCP-Cy5.5), anti-HLA-DR antibody (clone L243) conjugated to PerCP, anti-CD141 antibody (clone M80) conjugated to allophycocyanin (APC), and anti-CD34 antibody (clone 581) conjugated to AF488. Whole blood was stained with antibody master mixes (prepared weekly) for 15 min at room temperature. Red blood cells were lysed with fluorescence-activated cell sorting (FACS) lysing solution (BD Biosciences) and washed with phosphate-buffered saline (PBS). Cells were fixed in 1% (wt/vol) paraformaldehyde in PBS and then stored at 4°C in the dark and read within 3 h of staining.

For the phenotypic evaluation of Treg cells, 50 μl of whole blood was stained in TruCount tubes (BD Biosciences) containing 52,345 beads and using a lyse-no-wash protocol to allow accurate calculation of absolute cell numbers (27). Whole blood was stained with anti-CD45RA antibody (H100) conjugated to fluorescein isothiocyanate (FITC), anti-CD25 antibody (BC 96) conjugated to PE, anti-CD4 antibody (RPA-T4) conjugated to PerCP, and anti-CD127 antibody (A019D5) conjugated to Alexa Fluor 647 (AF647). Red blood cells were lysed by adding 450 μl of FACS lysing solution (BD Biosciences). Samples were read within 1 h of staining.

All samples were acquired on a portable BD Accuri C6 4-color flow cytometer using CFlow Sampler software (BD Biosciences). All antibodies were purchased from BioLegend (San Diego, CA).

Data analysis.

Flow cytometric data were analyzed using FlowJo software (TreeStar, Ashland, OR). The absolute number of DC was calculated by multiplying the proportion of DC events over white blood cell (WBC) events from flow cytometric data, with the absolute number of WBC being derived from automated blood counts (in units of 10⁹ cells per liter), then multiplying by 1,000 to give the number of DC per microliter of blood. Automated blood counts were performed on all blood samples tested for DC using an MS9-5sH cell count analyzer (Melet Schloesing Laboratories, Osny, France). Absolute numbers of Treg cells were calculated using TruCount tubes (BD Biosciences) according to the manufacturer's instructions to give the number of Treg cells per microliter of blood. Samples with less than 80 events in the DC gate (n = 8) or Treg cell gate (n = 1) were excluded. All statistical analyses were performed using GraphPad Prism 6 (GraphPad Software, La Jolla, CA). The Mann-Whitney U test was used for comparison between groups. Spearman's rank correlation coefficient was used to measure statistical associations between cell variables and parasitemia. Exact P values are presented but were not corrected for multiple comparisons.

Ethical approval.

The study was approved by the Human Research Ethics Committees of Gadjah Mada University, Yogyakarta, Indonesia, the Eijkman Institute Research Ethics Commission, Jakarta, Indonesia, and the NT Department of Health and Families and Menzies School of Health Research, Darwin, Australia. Written informed consent was obtained from all participants (or the primary caregiver or relative) prior to blood sampling.

RESULTS

Study participants.

Whole-blood capillary samples were obtained from 168 children, of whom 15 had asymptomatic P. falciparum infection, 36 had asymptomatic P. vivax infection, and 117 had no parasites detectable by microscopy and were considered controls. Venous blood was collected from 176 adults, of whom 20 had asymptomatic P. falciparum infection, 20 had asymptomatic P. vivax infection, 122 were aparasitemic controls, and 14 had acute uncomplicated malaria. Ninety-four percent (153/162) of asymptomatic adults had lived in Timika for more than 2 years. The self-reported durations of residency were similar between the 40 asymptomatic adults, for whom the median residency was 10 years and the interquartile range (IQR) 5 to 15 years, and the 122 aparasitemic control adults, for whom the median residency was 10 years and the IQR 5 to 20 years. Participant characteristics are summarized in Table 1. Groups were for the most part gender balanced. Parasitemia was significantly higher in adults with uncomplicated malaria than in the species-matched adult asymptomatic group. Adults with uncomplicated malaria had significantly lower lymphocyte counts and significantly higher monocyte counts than the adult controls. There were no significant differences in age or lymphocyte or monocyte count between healthy controls and children and adults with asymptomatic parasitemia.

TABLE 1.

Study participants

Characteristica	Value for indicated groupb
	Children				Adults
	Controls	Asymptomatic parasitemia		P. vivax malaria	Controls	Asymptomatic parasitemia		Malaria
		P. falciparum	P. vivax			P. falciparum	P. vivax	P. falciparum	P. vivax
No. of subjects
Total	117	15	36	2	122	20	20	6	8
Female/male	60/57	5/10	14/22	0/2	66/56	14/6	15/5	4/2	5/3
Papuan/non-Papuan	44/73	13/2	25/11	2/0	48/74	11/9	6/14	6/0	6/2
Age (yrs) [median (IQR)]	6 (4–8)	6 (5–8)	6 (3–9)	10	30 (25–35)	31 (25–36)	32 (29–36)	21 (20–24)c	27 (23–36)
Time resident in area (yrs) [median (IQR)]					10 (5–20)	10 (5–20)	10 (2–14)	NA	NA
No. of parasites/μl [median (IQR)]	ND	357 (140–960)	339 (143–1,863)	4,282, 4830	ND	257 (113–2,065)	201 (91–543)	24,725 (10,734–57,274)d	58,341 (10,778–89,678)d
Lymphocyte count (109/liter) [median (IQR)]	3.9 (3.1–4.5)	3.4 (2.5–3.8)	3.7 (3.0–4.4)	2.1, 2.9	2.5 (2.1–3.0)	2.5 (1.9–2.8)	2.5 (2.0–3.7)	1.4 (0.7–1.6)c,d	1.0 (0.7–1.2)c,d
Monocyte count (109/liter) [median (IQR)]	0.4 (0.3–0.6)	0.4 (0.2–0.6)	0.4 (0.3–0.5)	0.4, 1.5	0.5 (0.4–0.6)	0.6 (0.4–0.7)	0.5 (0.4–0.8)	0.7 (0.6–0.7)c	0.9 (0.4–1.4)c

^aIQR, interquartile range.

^bND, parasite not detected by microscopy; NA, not available.

^cSignificantly different from control group (Mann-Whitney U test, P < 0.05).

^dSignificantly different from asymptomatic malaria group (Mann-Whitney U test, P < 0.05).

Flow cytometry was constrained by the amount of blood available for testing, and hence, it was not always possible to conduct both DC and Treg cell stains for each participant. In 12% (91/784) of stains, flow cytometry data were deemed unreliable due to insufficient lysis of red blood cells.

HLA-DR expression of circulating DC during asymptomatic patent Plasmodium infection.

Peripheral blood total DC were identified by flow cytometry as lineage marker-negative HLA-DR⁺ cells (Fig. 1B). The activation/maturation of circulating blood DC was examined by measuring the median fluorescence intensity of HLA-DR expression. There was increased HLA-DR expression on total DC in adults with asymptomatic P. vivax infection (P = 0.05) (Fig. 2A), while the HLA-DR expression in those with asymptomatic P. falciparum infection was comparable to that in controls. In contrast, during acute uncomplicated P. falciparum or P. vivax malaria, DC HLA-DR expression was significantly lower (P = 0.0001 and P < 0.0001, respectively) (Fig. 2A). In children, HLA-DR expression was increased significantly in asymptomatic P. vivax infections (P = 0.02) (Fig. 2D).

FIG 1.

Representative staining of DC and T cells in fresh whole blood from children and adults. (A) DC subsets were identified by gating on lineage-negative WBC (CD3, -14, -19, and -56) and then on expression of CD303, CD1c, and CD141 surface markers to identify CD303⁺ plasmacytoid DC, CD1c⁺ myeloid DC, and CD141⁺ myeloid DC, respectively. (B) Total DC were identified as lineage marker-negative (CD3, -14, -19, -34, and -56) and HLA-DR-positive peripheral blood mononuclear cells. (C) Total Treg cells were identified as CD4⁺ lymphocytes with high expression of CD25 and low expression of CD127 surface markers. Total Treg cells were then subdivided into active and resting Treg cells based on expression of CD45RA.

FIG 2.

HLA-DR median fluorescence intensity (MFI) of total DC was determined in adult controls (n = 48), asymptomatic P. falciparum-infected adults (n = 18), asymptomatic P. vivax-infected adults (n = 19), uncomplicated P. falciparum-infected adults (n = 6), and uncomplicated P. vivax-infected adults (n = 8) (A), as well as in child controls (n = 37), asymptomatic P. falciparum-infected children (n = 10), asymptomatic P. vivax-infected children (n = 9), and uncomplicated P. vivax-infected children (n = 2) (D). In a subset of samples, HLA-DR MFI of CD1c⁺ myeloid DC (B, E) and CD141⁺ myeloid DC (C, F) was determined in adult controls (n = 4), adults with asymptomatic malaria (n = 4), and adults with uncomplicated malaria (n = 3), as well as in child controls (n = 2) and children with asymptomatic malaria (n = 8). Bars and whiskers show median values and interquartile ranges. Mann-Whitney U test was used for comparisons between groups (P < 0.05 was considered significantly different). Data were obtained by analysis of fresh whole blood using the 4-color flow cytometry panels shown in Fig. 1. AS, asymptomatic malaria; UM, uncomplicated malaria; Pf, P. falciparum; Pv, P. vivax; HLA-DR, human leukocyte antigen (MHC class II).

In a limited number of samples, additional staining was possible and HLA-DR expression could be examined on two myeloid DC (mDC) subsets, CD1c⁺ (blood dendritic cell antigen 1 [BDCA-1] positive) mDC (CD1c⁺ mDC) and CD141⁺ (BDCA-3 positive) mDC (CD141⁺ mDC). In both mDC subsets, HLA-DR was preserved during asymptomatic infection, but in 3 adults with uncomplicated malaria, HLA-DR expression appeared reduced (Fig. 2B, C, E, and F). Additional data are required to determine the significance of the apparent reduction.

Age- and species-specific variations in DC subset numbers.

Blood DC subsets showed marked variations according to clinical immunity (asymptomatic or symptomatic Plasmodium infection), host age, and parasite species. Three DC subsets were identified by flow cytometry (Fig. 1A) and are reported as the absolute number of CD303⁺ (BDCA-2 positive) plasmacytoid DC (pDC), CD1c⁺ mDC, and CD141⁺ mDC (see Table S1 in the supplemental material).

Plasmacytoid DC expanded in children with asymptomatic P. vivax infection (P = 0.002, n = 34) and were retained in adults with asymptomatic infection (Fig. 3). The expansion/retention appeared specific to asymptomatic infection, because pDC declined significantly in adults with acute uncomplicated malaria from either P. falciparum or P. vivax (P = 0.015, n = 4) (Fig. 3).

FIG 3.

Top: absolute number of CD303⁺ plasmacytoid DC (A), CD1c⁺ myeloid DC (B), and CD141⁺ myeloid DC (C) in peripheral blood of adult controls (A, n = 48; B, n = 52; C, n = 39), asymptomatic P. falciparum-infected adults (A, n = 17; B, n = 20; C, n = 14), asymptomatic P. vivax-infected adults (A, n = 18; B, n = 20; C, n = 17), and adults with uncomplicated malaria (A, n = 4; B, n = 8; C, n = 7). Bottom: absolute numbers of CD303⁺ plasmacytoid DC (D), CD1c⁺ myeloid DC (E), and CD141⁺ myeloid DC (F) in peripheral blood of child controls (D, n = 74; E, n = 77; F, n = 54), asymptomatic P. falciparum-infected children (D, n = 12; E, n = 10; F, n = 9), and asymptomatic P. vivax-infected children (D, n = 34; E, n = 33; F, n = 30). Bars and whiskers show median values and interquartile ranges. Mann-Whitney U test was used for comparisons between groups (P < 0.05 was considered significantly different). Data were obtained by analysis of fresh whole blood using the 4-color flow cytometry panels shown in Fig. 1A.

CD1c⁺ mDC declined significantly in children with asymptomatic infection with P. falciparum (P = 0.012, n = 10) or P. vivax (P = 0.006, n = 33) (Fig. 3). CD1c⁺ mDC were retained by adults with asymptomatic infection but declined in adults with acute uncomplicated malaria from either P. falciparum or P. vivax (P = 0.001, n = 8) (Fig. 3). Adult malaria patients were evaluated as a single group because the low number of patients with either P. falciparum or P. vivax precluded analysis for each species.

The CD141⁺ mDC subset appeared to be retained by children and adults with asymptomatic infection (Fig. 3).

Reduced Treg cell activation in asymptomatic patent Plasmodium infection.

Regulatory T cells were assessed in 94 children and 149 adults by flow cytometry as CD4⁺ CD25⁺ CD127^low lymphocytes and subdivided into activated and resting Treg cells (aTregs and rTregs, respectively) based on their CD45RA expression (28) (Fig. 1C).

Treg cell numbers were significantly lower in children with asymptomatic P. falciparum infection (P = 0.0004, n = 6), while in children with asymptomatic P. vivax infection, Treg cell numbers were similar to the numbers in aparasitemic controls.

In adults with asymptomatic P. falciparum or P. vivax infection, there was no change in the absolute or relative number of Treg cells (Table 2). In contrast, there was a significant decrease in the absolute number of Treg cells during acute uncomplicated malaria from each species (P < 0.0001) (Table 2)

TABLE 2.

T cell data

Parameter	Value for indicated group [median (IQRa)] or individual valuesb
	Children				Adults
	Controls	Asymptomatic parasitemia		P. vivax malaria	Controls	Asymptomatic parasitemia		Malaria
		P. falciparum	P. vivax			P. falciparum	P. vivax	P. falciparum	P. vivax
No. of subjects	76	6	10	2	103	14	18	6	8
CD4 T cells/μl blood	1,220 (1,034–1,520)	994 (461–1,266)	1,371 (858–1,738)	596, 635	713 (596–909)	836 (548–1,044)	773 (588–941)	201 (117–351)**	243 (170–321)***
CD25+ CD127low CD4+ Treg cells/μl blood	83 (63–110)	47 (34–56)**	86 (61–119)	43, 53	42 (33–53)	42 (31–67)	41 (30–54)	14 (11–23)***	16 (11–21)***
% of Treg cells in CD4 T cells	6.7 (6.0–7.9)	5.3 (4.4–6.7)*	6.5 (6.0–7.4)	7.2, 8.3	5.6 (4.6–6.4)	6.3 (5.2–7.1)	5.4 (4.4–6.5)	6.9 (5.7–9.7)	6.3 (5.6–7.4)
CD45RA− CD25high aTregs/μl blood	9.1 (7.3–12.1)	2.0 (1.3–3.1)***	4.9 (4.0–6.2)***	8.7, 16.9	8.1 (6.3–11.1)	7.1 (2.9–9.2)	6.1 (5.4–7.4)*	3.9 (2.6–5.0)**	4.3 (3.3–7.4)*
% of aTregs in total Treg cells	11.2 (9.2–13.7)	4.1 (3.6–6.1)***	7.0 (4.9–7.6)***	16.5, 39.5	20.7 (16.5–25.8)	11.9 (9.4–18.0)***	14.8 (13.0–18.1)*	25.2 (20.1–33.8)	29.2 (26.3–44.4)**
CD45RA+ CD25+ rTregs/μl blood	49.9 (35.6–67.4)	25.1 (19.2–32.7)*	57.2 (33.3–82.0)	11.0, 37.2	11.8 (6.8–18.7)	11.7 (8.8–21.3)	10.0 (7.2–17.8)	4.4 (3.1–6.0)**	2.9 (1.7–4.1)***
% of rTregs in CD25+ CD127low CD4+ T cells	59.0 (52.3–65.2)	57.3 (52.8–62.5)	65.4 (54.4–72.5)	25.7, 70.2	30.9 (20.8–36.1)	30.9 (23.6–37.8)	28.6 (23.4–38.4)	30.2 (19.1–33.4)	19.4 (14.2–20.5)*
rTreg/aTreg ratio	5.2 (3.9–7.3)	15.0 (9.8–16.5)**	9.1 (7.9–15.9)***	0.6, 4.3	1.3 (0.9–2.0)	2.3 (1.5–3.9)*	1.9 (1.3–2.7)*	1.1 (0.8–1.5)	0.7 (0.4–0.9)**

^aIQR, interquartile range.

^bAsterisks indicate values significantly different from the value for the control group (Mann-Whitney U test) as follows: *, P < 0.05; **, P ≤ 0.001; ***, P ≤ 0.0001.

Importantly, in both adults and children with asymptomatic P. falciparum or P. vivax infection, significantly smaller proportions of circulating Treg cells were activated (CD25^high CD45RA⁻) than in aparasitemic controls, resulting in significantly increased ratios of rTregs to aTregs in asymptomatic Plasmodium infection, while the opposite was observed in acute uncomplicated malaria (Table 2). These observations suggest that reduction in the proportion of activated Treg cells is associated with clinical immunity.

Treg cell and DC activation and parasitemia.

There was no evidence of a relationship between Treg cell activation and parasitemia within each of the asymptomatic adult or asymptomatic child (Spearman's rank correlation coefficient [r_s] = −0.05, P = 0.8, n = 32, and r_s = −0.05, P = 0.8, n = 16, respectively) or adult malaria (r_s = 0.2, P = 0.5, n = 14) groups. There was no overlap in parasitemia between the asymptomatic adults and malaria patients (Table 1). Total DC HLA-DR expression was inversely correlated with parasitemia in adults with asymptomatic P. falciparum infection (r_s = −0.54, P = 0.02, n = 18) and positively correlated in children with asymptomatic P. vivax infection (r_s = 0.80, P = 0.01, n = 9).

DISCUSSION

This study represents the first description of appropriate DC maturation and Treg cell activation in asymptomatic Plasmodium infection, characterized by increased or retained HLA-DR expression on DC and reduced activation of Treg cells in both children and adults with asymptomatic P. falciparum or asymptomatic P. vivax parasitemia. In contrast, in adults with acute malaria, DC have reduced HLA-DR expression and an increased proportion of Treg cells are activated. These data provide evidence of appropriate immune activation in asymptomatic infection and immune dysfunction in acute clinical malaria. The active control of asymptomatic infection is associated with and is likely contingent upon functional DC and reduced Treg cell activation.

The evaluation of adults and children highlighted similarities in DC and Treg cell activation in the two age groups, and notwithstanding some differences depending on parasite species (P. falciparum and P. vivax), our results suggest some common DC- and Treg cell-related immunological mechanisms contribute to the control of Plasmodium parasitemia in children and adults. In malaria, a similar panspecies effect has been noted with DC dysfunction in both acute P. falciparum and P. vivax malaria (8) and Treg cell activation in acute P. falciparum (12) and P. vivax (29, 30) malaria. It remains an open question whether there are age-associated differences in DC and Treg cell activation during acute malaria, because the low number of uncomplicated malaria samples examined (particularly in children) prevented a comprehensive comparison of DC and Treg cell activation between children and adults with acute malaria.

Asymptomatic parasitemia in areas of endemicity is associated with reduced parasite burdens compared to results for those with clinical illness (31). While we hypothesize that activation of DC and reduced activation of Treg cells may contribute to the control of parasitemia during clinical immunity, it is unlikely that low parasitemia per se results in these measures of immune activation. In contrast to the increased/retained HLA-DR expression we found in asymptomatic parasitemia, our observations from experimental P. falciparum blood stage infections of malaria-naive volunteers indicate that DC HLA-DR expression levels on pDC decrease prior to the onset of symptoms (9), and at levels of parasitemia much lower than those seen in patent asymptomatic infections in the current and other studies in areas of malaria endemicity (32, 33). While we speculate that the retention/increase in DC HLA-DR expression, which is required for antigen presentation, and the reduction in Treg cell activation observed in individuals from area of endemicity with asymptomatic parasitemia contribute to their clinical immunity, the relative contribution to reduced parasite growth and/or reduced inflammatory responses remains to be determined. Our DC HLA-DR data support the earlier description of HLA-DR retention by circulating blood DC in Fulani children, which is understood to contribute to protective immunity against malaria (34).

Treg cells are efficient suppressors of immune responses. We observed a significant drop in circulating Treg cell numbers in acute malaria, which coincided with a significant drop in lymphocytes and CD4 T cell counts. Lymphopenia is common in clinical malaria (35, 36), occurring already at subpatent parasitemia during a first P. falciparum infection (32), which suggests that the absence of lymphopenia in asymptomatic Plasmodium carriers is notable and potentially reflects or contributes to clinical immunity. Transient T cell migration from the peripheral blood (37) to tissues such as the spleen (38) or to sites of inflammation during acute malaria contributes to lymphopenia, and T cells reemerge following antimalarial treatment (36). We observed no significant change in the proportion of Treg cells within the CD4 T cell compartment, suggesting that Treg cells leave the peripheral blood together with other CD4 T cells to travel to sites of inflammation (39). However, in clinical malaria, more of the Treg cells remaining in the circulation displayed a recently activated CD25^high CD45RA⁻ phenotype, which represents highly suppressive Treg cells (28). In contrast, in asymptomatic Plasmodium infection, the circulating Treg cells were of a less activated phenotype than those in uninfected controls.

Studies on Treg cell activation in asymptomatic Plasmodium infection are limited. A study from Peru, where malaria is hypoendemic, reports similar CD25⁺ Foxp3⁺ CD127^low Treg cell counts between symptomatic and asymptomatic P. falciparum infections and uninfected controls (40). In contrast, Indonesian school-aged children with asymptomatic patent or subpatent Plasmodium infection show a trend to reduced CD25⁺ Foxp3⁺ CD127^low Treg cell proportions among CD4 T cells compared to results for uninfected children (41), similar to our observation in asymptomatic P. falciparum-infected children. These asymptomatic children, however, display increased Treg cell TNFR2 expression and suppression of Th2 responses (41). We were unfortunately unable to assess TNFR2 expression in this study due to limited blood volumes and limited channels on our field flow cytometer, which precludes direct comparison of these studies.

Self-resolving murine models of malaria consistently show a transient expansion of Treg cells in the spleen shortly after infection with P. yoelli or P. chabaudi, followed by a drop in Treg cell numbers due to apoptosis (42). Continued Treg cell expansion negatively impacts immune control of the infection (42), with Treg cell activation resulting in increased suppression of T cell responses (43). In humanized mice expressing HLA-DR0401, Treg cell activation has been associated with suppression of protective anti-parasitic antibody responses (44) and failure to control parasite growth.

Resting CD25⁺ CD45RA⁺ Treg cells (rTregs) provide a reservoir for recently activated CD25^high CD45RA⁻ Treg cells (aTregs), which are highly suppressive, highly proliferative cells with a short life span (28). The increased frequency of rTreg-to-aTreg ratio in asymptomatic P. falciparum and P. vivax infection, which was particularly prominent in children, contrasts strikingly with the significant drop in rTreg-to-aTreg frequency found in acute malaria. In malaria, a relative increase in aTregs occurred alongside a relative decrease in rTregs, while in asymptomatic infection, the proportion of aTregs decreased and the rTreg proportion remained largely unchanged. These findings suggest that while acute malaria is accompanied by Treg cell overactivation, potentially leading to exhaustion of the Treg cell reservoir, asymptomatic Plasmodium infection is characterized by a dampened Treg cell response, which may indicate that a less suppressive environment for antiparasitic immune responses is contributing to control of parasitemia.

While it is unclear how Treg cell overactivation may be avoided, one study suggests that immature blood DC contribute to Treg cell differentiation via surface binding of the immunomodulatory cytokine transforming growth factor beta (TGF-β) to latency-associated peptide (LAP) (45). Curiously, downregulation of LAP on blood DC coincides with DC maturation, characterized by upregulation of HLA-DR and costimulatory molecules (45), suggesting that overcoming the maturation defect observed in acute malaria may be linked to reducing Treg cell overactivation. Whether LAP downregulation on DC is impaired in malaria remains to be established.

Enumeration of three major blood DC subsets in asymptomatic infection revealed notable differences according to Plasmodium species. Plasmacytoid DC expanded in asymptomatic P. vivax infection in children, suggesting that pDC may contribute to control of parasitemia and/or disease in asymptomatic infection. In contrast, there was no expansion of myeloid DC, with reduced CD1c⁺ mDC in children with asymptomatic infection and adults with malaria. CD141⁺ mDC were stable in children and adults with asymptomatic infection. Urban and colleagues have described increased CD141⁺ mDC in African children with severe P. falciparum malaria (46), but they did not investigate asymptomatic or uncomplicated malaria. The DC counts we report in control microscopy-negative Indonesian children were similar to the pDC, CD1c⁺ mDC, and CD141⁺ mDC counts reported in Kenyan control children (46).

The expansion or retention of blood DC in asymptomatic P. falciparum or P. vivax parasitemia contrasts with that seen in acute malaria, where P. falciparum and P. vivax parasitemias are associated with reductions in pDC and mDC numbers (8). In a limited number of samples where additional field flow cytometry was undertaken, HLA-DR expression appeared to be retained by CD1c and CD141 mDC during asymptomatic parasitemia but lost in malaria. These data support the notion that DC remain capable of antigen presentation in asymptomatic parasite carriers; however, further studies are required to comprehensively evaluate HLA-DR expression and the function of blood DC in asymptomatic infection.

Our study had a number of limitations. We evaluated parasitemia by microscopy only, and it is possible that a significant proportion of microscopy-negative aparasitemic controls had subpatent PCR-level parasitemia (47). Another limitation is that subjects were not followed longitudinally and those with asymptomatic parasitemia were treated, preventing assessment of whether people would remain asymptomatic, clear the infection, or develop malaria (48). Notwithstanding these limitations, at the time of blood collection, we can identify clear differences in DC and Treg cell activation between the groups. Future longitudinal studies of DC and Treg cell activation, including submicroscopic infection, will further enhance the understanding of immune activation in asymptomatic infection.

Asymptomatic parasitemia is largely ignored by immunization programs in countries where malaria is endemic, meaning a proportion of children and adults with asymptomatic parasitemia are routinely vaccinated with little understanding of the effects asymptomatic parasitemia has on the efficacy of current childhood vaccines (49). As all vaccines (standard childhood vaccines and new immunizations being developed) need functional DC to stimulate protective immune responses, our study begins to addresses this current knowledge gap. While more vaccine immunogenicity and efficacy studies are required, the data suggest inadequate immune responsiveness during acute malaria but that vaccinating asymptomatic carriers of Plasmodium may be potentially acceptable, at least in areas of low-to-medium transmission.