Frimpong et al. Malar J  (2018) 17:263  
https://doi.org/10.1186/s12936-018-2410-6 Malaria Journal
RESEARCH Open Access
Characterization of T cell activation 
and regulation in children with asymptomatic 
Plasmodium falciparum infection
Augustina Frimpong1,2,3* , Kwadwo Asamoah Kusi1,2, Bernard Tornyigah2, Michael Fokuo Ofori1,2 
and Wilfred Ndifon3,4*
Abstract 
Background: Asymptomatic Plasmodium infections are characterized by the absence of clinical disease and the 
ability to restrict parasite replication. Increasing levels of regulatory T cells (Tregs) in Plasmodium falciparum infections 
have been associated with the risk of developing clinical disease, suggesting that individuals with asymptomatic 
infections may have reduced Treg frequency. However, the relationship between Tregs, cellular activation and parasite 
control in asymptomatic malaria remains unclear.
Methods: In a cross-sectional study, the levels of Tregs and other T cell activation phenotypes were compared using 
flow cytometry in symptomatic, asymptomatic and uninfected children before and after stimulation with infected 
red blood cell lysates (iRBCs). In addition, the association between these T cell phenotypes and parasitaemia were 
investigated.
Results: In children with asymptomatic infections, levels of Tregs and activated T cells were comparable to those 
in healthy controls but significantly lower than those in symptomatic children. After iRBC stimulation, levels of Tregs 
remained lower for asymptomatic versus symptomatic children. In contrast, levels of activated T cells were higher 
for asymptomatic children. Strikingly, the pre-stimulation levels of two T cell activation phenotypes (CD8+CD69+ 
and CD8+CD25+CD69+) and the post-stimulation levels of two regulatory phenotypes (CD4+CD25+Foxp3+ and 
CD8+CD25+Foxp3+) were significantly positively correlated with and explained 68% of the individual variation in 
parasitaemia. A machine-learning model based on levels of these four phenotypes accurately distinguished between 
asymptomatic and symptomatic children (sensitivity = 86%, specificity = 94%), suggesting that these phenotypes 
govern the observed variation in disease status.
Conclusion: Compared to symptomatic P. falciparum infections, in children asymptomatic infections are character-
ized by lower levels of Tregs and activated T cells, which are associated with lower parasitaemia. The results indicate 
that T cell regulatory and activation phenotypes govern both parasitaemia and disease status in paediatric malaria in 
the studied sub-Saharan African population.
Keywords: Malaria, Regulatory T-cells, T-cell activation, Asymptomatic, Symptomatic, Children, falciparum, Immunity
*Correspondence:  tinafrimp@gmail.com; wndifon@aims.ac.za 
1 West African Centre for Cell Biology of Infectious Pathogens (WACCBIP), 
Department of Biochemistry, Cell and Molecular Biology, University 
of Ghana, Legon, P. O. Box LG 54, Accra, Ghana
4 African Institute for Mathematical Sciences, University of Stellenbosch, 7 
Melrose Rd, Muizenberg, Cape Town 7945, South Africa
Full list of author information is available at the end of the article
© The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License 
(http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, 
provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, 
and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat ivecom mons .org/
publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Frimpong et al. Malar J  (2018) 17:263 Page 2 of 13
Background an increased parasite density [9–12]. Furthermore, the 
Falciparum malaria occurs when sporozoites inoculated expansion of Tregs in malaria has been associated with 
into the human host develop in the liver into merozoites decreased antigen-specific immune responses [11].
that infect red blood cells and cause clinical disease. The Also, a recent study by Kurup et  al. [13] has shown 
acquisition of natural immunity to falciparum malaria is that CTLA-4 Tregs expand during symptomatic malaria 
slow and requires frequent exposure to the parasite over in both human and murine models, which is associ-
a period of time [1]. Despite previous exposure to the ated with decreased parasite clearance and impedes the 
parasite, people in endemic areas may remain suscepti- acquisition of immunity in murine models. Other stud-
ble to clinical disease or they may be asymptomatic car- ies have also reported the upregulation of TNFRII on 
riers of parasites as clinical immunity is only partial and Tregs with asymptomatic parasitaemia [14]. There have 
never sterile. Also, repeated parasite exposure has been also been reports on the upregulation of FOXP3 mRNA 
associated with limited protectiveness to vaccine can- transcripts during acute malaria infections in children 
didates [2, 3]. The lack of a proper understanding of the and naïve adults, which negatively correlated with Th1 
immune responses occurring during natural infections memory responses [9, 15]. Nonetheless, other studies 
has, for example, resulted in an inability to develop effec- have also shown conflicting data whereby no association 
tive interventions such as vaccines. was found between the levels of Tregs and Plasmodium 
Understanding the regulatory and protective immune infection [16–19]. Collectively, these imply that the activ-
responses during asymptomatic and clinical infections ity of Tregs associated with the development of protec-
remain necessary to comprehend mechanisms that ena- tive immunity needs to be comprehended. The likely 
ble the control of infections as well as the persistence and suggestions are that infections may cause the expansion 
survival of the parasite. A number of studies in malaria of Tregs, which in turn may cause immune suppression 
have associated protection from clinical disease with hav- and enhance parasite growth as observed in other studies 
ing a broad antibody repertoire [4–6]. Nonetheless, the [11, 20, 21].
presence of asymptomatic infections in children who may This study aims to compare the expression levels of T 
not have a broad antibody repertoire suggests that there cell activation and regulatory markers across sympto-
is some level of immunity to the parasite by the host and matic, asymptomatic and healthy control children living 
this is characterized by the absence of clinical manifesta- in hyperendemic areas with stable malaria transmission 
tions of the disease. Moreover, during Plasmodium falci- in Ghana. The Treg markers CD25+Foxp3+, the early 
parum infections, it is believed that the effector function activation marker CD69, and the late activation marker 
of immune cells will be compromised due to immune CD25 were measured. Tregs have an established role in 
regulation [7]. This may be induced by the specific expan- suppressing effector immune responses to a variety of 
sion of certain T or B cell sub-sets and modulation of cer- pathogens, including malarial parasites [22, 23]. CD69 
tain antigen presenting cells, such as the dendritic cells expression in CD4+ T cells has been shown to correlate 
[8]. T cells express receptors that enable co-stimulation, with the development of antigen-specific antibodies in 
activation, memory formation, and immune regulation to experimental human falciparum malaria [24]. CD69 is a 
ensure effective and timely immune response induction transmembrane glycoprotein expressed during early acti-
upon antigen recognition. The expansion of specific cell vation and increases with inflammation with the poten-
sub-sets, especially those that express regulatory mark- tial to induce cytotoxic activity once crosslinked [25–29], 
ers, may either enhance or inhibit the development of whereas CD25 (IL-2α receptor) has been associated with 
immunity against an infection. However, the association T cell proliferation and differentiation through the IL-2 
between such cellular activation and regulatory markers cytokine [28]. This suggests that their combined expres-
and parasite control during asymptomatic infections is sion may lead to an enhanced cellular immune activity. 
inadequately understood. Therefore, it was hypothesized that asymptomatic infec-
Regulatory T cells are unique cell phenotypes that func- tions have reduced Treg levels, such that exposure to P. 
tion to maintain homeostasis when the immune response falciparum is associated with increased cellular response 
is activated. The establishment of immune homeosta- and lower parasitaemia, which in turn feeds back to 
sis may result in blocking the activity of other immune reduce cellular activation.
cells. For instance, CTLA-4 (also known as CD152), once 
activated, functions to inhibit activation of both antigen Methods
presenting cells and other T cells. Even though the role Study sites
of Tregs during P. falciparum infections remains con- Participants for the study were recruited from Asutsu-
troversial, it has been observed that in both human and are and Paakro sub-districts, which are hyperendemic 
rodent malaria an early induction of Tregs may result in for malaria transmission. Asutsuare has two malaria 
Frimpong et al. Malar J  (2018) 17:263 Page 3 of 13
transmission seasons; June to August (high transmis- Stimulation of PBMCs with infected and uninfected red 
sion season) and November to December (low trans- blood cells
mission season) with an entomological inoculation rate Plasmodium falciparum parasites of the NF54 strain 
of 14.6 infective bites/man/year whereas Paakro has were cultured in O + red blood cells at 3% haematocrit 
May to June as the high transmission season and Sep- in culture medium (RPMI 1640 medium, 25  µg/ml of 
tember to October as the low transmission season [30, gentamycin, 10% heat-inactivated  O+ human serum). 
31]. Samples from participants were obtained during The culturing was done in the presence of 7.5% sodium 
the high transmission seasons. bicarbonate at 37  °C in a 5%  O2, 5%  CO2 and 90%  N2 
atmosphere. PBMCs were later thawed and rested for 
6 h in 10% fetal bovine serum. About 400,000 cells were 
Participants and sample collection later stimulated with intact P. falciparum trophozoites/
The study was approved by the Institutional Review schizont (NF54 clone)-infected RBCs (iRBCs; 3 iRBCs: 1 
Board of the Noguchi Memorial Institute for Medical PBMC) or uninfected RBCs (uRBCs; 3 uRBCs: 1 PBMC) 
Research, University of Ghana (Permit No. 096/15-16). and cultured in complete RPMI 1640 in 5% C O2 at 37 °C. 
A written informed consent was obtained from parents After 4 h of stimulation, brefeldin A was added at a con-
or guardians and assent appropriately received from centration of 10  µg/ml. Cells were washed and stained 
the children before they were enrolled. Samples were after 18 h with the following monoclonal antibodies; anti-
obtained in a cross-sectional study from 57 children CD3 (APC-H7), CD4 (BUV 395), CD8 (PerCP-Cy5.5), 
under 13  years old who satisfied the inclusion criteria CD25 (PE-CF594), CD69 (PE-Cy7), CD152/CTLA-4 
with no known conditions that could interfere with (APC), FOXP3 (PE) all from BD. The cells were washed 
the experiments. The participants were grouped into by centrifugation before staining for both extracellular 
P. falciparum-infected asymptomatic children (n = 18), and intracellular markers.
symptomatic malaria patients (n = 22) and healthy con-
trols (n = 17). About 5 ml of venous blood was collected Statistical analysis
into heparin tubes before anti-malarial treatment. Both Data analyses were performed with the GraphPad Prism 
thick and thin smears were prepared and stained with version 6.01 (GraphPad Software, Inc.) and the R statis-
Giemsa for parasite identification after screening for tical software version 3.4.0 (R Foundation for Statistical 
infection with rapid diagnostic tests. Haematological Computing). The demographics and clinical characteris-
indices were determined by an automated haemato- tics of the study participants were compared among the 
logical analyzer. PBMCs were isolated by ficoll gradi- 3 study groups using Chi square test for categorical varia-
ent centrifugation and stored in liquid nitrogen until bles, Kruskal–Wallis or One-way ANOVA for continuous 
the time of the experiment. The PBMCs were cryopre- variables, Mann–Whitney U test and Wilcoxon-Signed 
served in fetal bovine serum with 10% dimethyl sulfox- Rank Test for paired comparisons for data that were not 
ide (DMSO). normally distributed. For comparing the markers of T 
cells among the 3 study populations, the Kruskal–Wallis 
test was used with a Dunn’s post hoc test or a Bonferroni 
Flow cytometry analysis correction for multiple comparisons where necessary. 
Stored PBMCs were thawed and washed. The viability Spearman’s rank correlation was used to determine 
was assessed by trypan blue dye exclusion method. Cells associations between markers. Support vector machine 
with viability greater than 95% were used in the assay. model, a supervised machine-learning algorithm was 
The cells were surface stained with the following antibod- used to predict disease status. Statistical significance was 
ies for T cell sub-sets (anti-CD3, anti-CD4, anti-CD8), set at P-values < 0.05.
co-stimulation markers (anti-CD28, anti-CD57) and 
activation markers (anti-CD25, anti-CD69) (Additional 
file  1). The cells were washed, fixed and permeabilized Results
using FOXP3 buffer set (BD) and intracellularly stained Characteristics of the study population
for regulatory markers Foxp3 (Biolegend) and CTLA-4 Venous blood samples were obtained from 57 children 
(BD). Fluorescence minus one controls and compen- including 18 with asymptomatic P. falciparum infec-
sation were performed to set gates using single colour tions, 22 with symptomatic malaria, and 17 with no P. 
stained or unstained PBMCs. Data were compensated falciparum parasites detected in blood by microscopy 
and analysed using Flowjo V10 software (Tree Star, San or rapid diagnostic test (Table 1). There was no statisti-
Carlos, CA, USA). The gating strategies are outlined in cally significant difference between the ages of children 
Figs. 1a and 2a. in the asymptomatic versus symptomatic groups. In 
Frimpong et al. Malar J  (2018) 17:263 Page 4 of 13
a
CD4 FOXP3 CTLA-4
b T r e g s T r e g s
60 P = 0 .0 0 3 9 60 P = 0 .0 1 2 7
P = 0 .0 0 6 5
P = 0 .0 3 9 4
40 40
E x v ivo iR B C s
20 20
0 0
c
ro
l t i t ic ls t ica t ict a o a a
C o
n m r
to to
m
n t om mp p o t o
m pm C p
t
s y y y
m ym
A S A s S
c C T L A -4 + T re g s C T L A -4 + T re g s
P = 0 .0 0 1 7 100 P = 0 .0 2 6 8
100 P = 0 .0 4 9 3
P = 0 .0 1 7 4
50
50
E x v ivo iR B C s
0 0
ls t i
c
t ic ls t i
c ic
ro a a ro
a a t
t m m n t om m
C o
n o o t o
p t t
m p
t C o p p
m ym
s y y s S y
m
A S A
Fig. 1 Percentage expression of CD4+CD25+FOXP3+ regulatory T cells among study groups. a Representative flow cytometry gating strategy 
for phenotyping CD4+CD25+FOXP3+ and CTLA-4+ Tregs in peripheral blood. The percentage expression and activation of regulatory T cells 
were analysed in healthy controls (n = 16), children with asymptomatic infections (n = 18) and symptomatic falciparum malaria (n = 22); for levels 
of b Tregs analysed as CD4+CD25+FOXP3+; and, c activated Tregs analysed as CD4+CD25+FOXP3+CTLA-4+ T cells both ex vivo and after iRBC 
stimulation. The data are presented as box plots with inter-quartile ranges. The 10th and 90th percentiles are denoted by whiskers. Medians are 
indicated by the horizontal lines across the boxes. Kruskal–Wallis test was used for comparisons, followed by Dunn’s test where necessary
contrast, the healthy controls were significantly older groups. Total leukocyte counts were significantly higher 
than the asymptomatic (P = 0.0404) and symptomatic in the asymptomatic children compared to symptomatic 
children (P = 0.0123). Also, the mean haemoglobin lev- children (P = 0.0478) but comparable to controls. Even 
els in asymptomatic children were significantly higher though the median lymphocyte counts did not differ sig-
compared to the symptomatic children (P = 0.0348). nificantly amongst the groups, lower levels were found 
However, levels were comparable between the children in the symptomatic group than in the asymptomatic and 
in the control group and asymptomatic or symptomatic control groups. Also, platelet levels decreased with the 
% C D 25 + FO X P 3+C T L A -4 + T % C D 4+C D 25 + F O X P 3+ T c e lls CD3
ce lls
CD25
% C D 25 + FO X P 3+C T L A -4 + T % C D 25 + FO X P 3 + o f C D 4+ T c e lls
ce lls
Frimpong et al. Malar J  (2018) 17:263 Page 5 of 13
a
CD4 CD69
CD8
C D 4 +
b C D 8 +
80 P = 0 .0 0 0 7 80 P = 0 .0 1 8
P = 0 .0 0 1 6 E x v ivo
60 60 P = 0 .0 3 7 6
40 40
20 20
0 0
ls t ic t ic ls t ic ca ia a t
n t
ro m o ao om
r
o t t o n
t
to
m
p to
m
C p C pm pm m m
A s
y
S y s yA S
y
Fig. 2 Expression of T-cell activation markers CD25/CD69 on T cells in PBMCs from the study cohort. a Representative flow cytometry gating 
strategy for phenotyping activation markers on CD4+ and CD8+ T cells ex vivo; the expression of the activation markers b CD25−CD69+ on T cells 
was analysed in healthy controls (n = 17), asymptomatic P. falciparum-infected children (n = 18), and symptomatic P. falciparum-infected children 
(n = 21). The data are presented as box plots with inter-quartile ranges. The 10th and 90th percentiles are presented as whiskers. Medians are 
indicated by the horizontal lines across the boxes. Kruskal–Wallis test was used for comparisons, followed by Dunn’s test where necessary
CD3
% C D 25 -C D 6 9 + T c e lls
% C D 25 -C D 6 9 + T c e lls
CD25
Frimpong et al. Malar J  (2018) 17:263 Page 6 of 13
Table 1 Demographics and clinical characteristics of the study participants
Characteristics Control Asymptomatic Symptomatic P values
Sample size n = 17 n = 18 n = 22
Age (IQR), years 9 (8–11) 7 (4.5–9) 6 (4.8–7) 0.0087a
Female (%) 52.94 44.44 50 0.8765b
Mean haemoglobin (IQR), g/dl 11.5 (10.8–12.1) 12.7 (11.7–13.58) 10.7 (8.8–13.1) 0.0402c
Parasitaemia (IQR), µl NA 845 (260.7–3812) 13,973 (7238–58,764) 0.0009d
Leukocytes ( 109/l) 7 (5.7–8.0) 7.7 (6.1–9.6) 5.1 (1.2–8.3) 0.0436a
Lymphocytes  (106/l) 2.9 (2.5–3.6) 2.1 (1.2–3.45) 1.9 (1.3–3.9) 0.0889a
Platelets ( 109/l) 305 (237–356) 223 (193–280) 101 (61–198) > 0.0001a
IQR interquartile range, NA not applicable
a Kruskal–Wallis test
b Chi square test
c One-way ANOVA
d Mann–Whitney U test
severity of P. falciparum infections. The median platelet The levels of CTLA4+ Tregs in children with asympto-
counts in the symptomatic children were significantly matic malaria were significantly lower than in those with 
lower than in the asymptomatic (P = 0.0029) and con- clinical malaria (P = 0.0174, Fig. 1c) but comparable with 
trol (P < 0.0001) groups. Children in the asymptomatic the control group. However, levels of CTLA4+ Tregs in 
group had statistically similar platelet counts as the con- the symptomatic group were significantly higher than in 
trol group. Parasitaemia levels were significantly lower in the control group (P = 0.0034). In addition, after iRBC 
asymptomatic children compared to children with symp- stimulation, CTLA-4+ Treg levels remained significantly 
tomatic infection (P = 0.0009). lower in the asymptomatic group compared to the symp-
tomatic group (P = 0.0493) but were comparable to levels 
Decreased levels of regulatory T cells in asymptomatic observed in healthy controls (P = 0.5457, Fig. 1c).
Plasmodium falciparum infections
To investigate if there are any differences between the Decreased expression of CD69 activation marker on T 
P. falciparum-infected group and healthy controls with cells in asymptomatic Plasmodium falciparum infections 
respect to T cell regulation, the levels of Treg popula- before infected red blood cell stimulation
tions in the 3 study groups were determined. Here, Treg With the observed levels of Tregs being lower in the 
populations were classified as CD3+, CD4+, CD25+ and asymptomatic children compared to symptomatic chil-
FoxP3+ (Fig. 1a). dren, the extent of cellular activation was measured to 
For the peripheral blood mononuclear cells (PBMCs) determine if they may differ across the study groups. The 
analysed directly without stimulation (ex vivo), it was expression of the CD69 activation marker on both CD4+ 
observed that Tregs had a lower frequency in the asymp- and CD8+ T cell sub-sets before in vitro stimulation was 
tomatic children compared to the symptomatic children investigated (Fig. 2a). Children with asymptomatic malaria 
(P = 0.0065), but levels were comparable between asymp- had significantly lower levels of CD69+ expression on 
tomatic and control groups (P > 0.05). Also, the Treg fre- CD4+ T cells compared to children with symptomatic 
quency in the symptomatic children was higher than in disease (P = 0.0016) but had comparable levels with the 
the controls (P = 0.0209, Fig. 1b). This trend remained the controls (P > 0.05, Fig. 2b). In addition, children with symp-
same after the cells were stimulated with iRBCs in vitro; tomatic malaria had a higher level of the CD4+CD69+ T 
lower levels of Tregs were found in the asymptomatic cells than the controls (P = 0.0068, Fig. 2b). This trend was 
children than in the symptomatic children (P = 0.0394), the same for the CD8+CD69+ T cells.
while levels were comparable to the controls. Similarly, 
levels in the symptomatic children were higher than in The pattern of expression of CD25 on T cells 
the controls (P = 0.0391, Fig. 1b). among symptomatic, asymptomatic Plasmodium 
Furthermore, the levels of activated Tregs based on the falciparum infections and healthy controls before infected 
expression of CTLA4, an immunosuppressive marker red blood cell stimulation
that inhibits activation of immune cells by direct contact, The expression levels of CD25, a late activation marker, on 
were determined. The levels of CTLA4+ Tregs differed CD4+ and CD8+ T cells were determined (Fig. 2a). Except 
significantly across the study populations (P = 0.0017). for CD8+CD25+CD69− T cells, levels of CD25 were not 
Frimpong et al. Malar J  (2018) 17:263 Page 7 of 13
a C D 4 + C D 8 +
100
P = 0 .0 3 0 1
P = 0 .6 8 0 8 P = 0 .0 2 5 7
100
50
50
0 0
l t ic ic ic c
n t
ro a a t lso a t a t
i
o to
m r
om n t omt omC p p o t t
ym
p
m C p
s S y s y
m ym
A A S
C D 4 + C D 8 +
b
60 P = 0 .0 0 3 1 60 P = 0 .1 1 9 2
40 P = 0 .0 1 1 0 40
20 20
0 0
ls t ic t ic o l t i
c
t ic
tro
a a tr am an o om o n omt t t to
m
C o p p C pm pm m m
A s
y y yS yA s S
C D 8 +
c C D 4 + 30 P = 0 .0 0 2 9
20
P = 0 .0 0 9 5 P = 0 .0 0 6 2
P = 0 .0 1 6 20
10
10
0 0
ls t ic
ic c
t ic l
s t t i
t ro
a a ro a a
n om om n
t om m
o p t
o
p t C
o p t
C p
t
m m ymy ms S y s S
y
A A
Fig. 3 Expression of T-cell activation markers CD25/CD69 on T cells in PBMCs from the study cohort. a CD25+CD69−; b CD25+FOXP3− cells; 
and, c CD25+CD69+ on T cells was analysed in healthy controls (n = 17), asymptomatic P. falciparum-infected children (n = 18), and symptomatic 
P. falciparum-infected children (n = 21). The data are presented as box plots with inter-quartile ranges. The 10th and 90th percentiles are presented 
as whiskers. Medians are indicated by the horizontal lines across the boxes. Kruskal–Wallis test was used for comparisons, followed by Dunn’s test 
where necessary
% C D 25+ FO X P 3 - T c e lls % C D 25 + C D 6 9 - T c e lls
% C D 2 5 + C D 6 9 + T c e lls
% C D 2 5 +C D 6 9 + T c e lls % C D 25 + FO X P 3 - T c e lls % C D 25 + C D 6 9 - T c e lls
Frimpong et al. Malar J  (2018) 17:263 Page 8 of 13
significantly different between asymptomatic and symp- children  increased significantly above those found in 
tomatic children. There was no significant difference in the symptomatic children (P = 0.0002) and controls 
the expression of CD25 on CD4+ T cells across the study (P = 0.0008, Fig. 4a). Levels of CD4+CD69+ T cells did 
populations (P = 0.4971, Fig.  3a). However, for CD8+ T not differ significantly between symptomatic children 
cells, the expression of CD25 was significantly lower in the and controls. Higher expression of CD8+CD69+ cells 
asymptomatic group compared to the symptomatic chil- was also observed in the asymptomatic cohort compared 
dren (P = 0.0257), whereas levels between the asympto- to both symptomatic children (P = 0.0057) and controls 
matic and control groups were comparable (Fig. 3a). (P = 0.0054) (Fig.  4a). As was observed in the CD4+ T 
CD25+FOXP3− T cells have been classified as acti- cell compartment, levels of CD8+CD69+ T cells did not 
vated effector Th 1 cells capable of secreting effec- differ significantly between symptomatic children and 
tor cytokines, such as IFNγ (interferon gamma), TNF controls (Fig. 4a).
(tumour necrosis factor) and IL-10 (interleukin-10). Also, no significant difference was found in the 
Therefore, the ex  vivo expression of these markers was expression of CD25+CD69− T cells on any of the T 
compared across the study groups. No significant differ- cell sub-sets across the study groups (Fig. 4b). Levels of 
ence was observed in the expression of CD25+FOXP3− CD25+FOXP3− activated effector T cells were increased 
on either CD4+ or CD8+ T cell sub-sets between in the asymptomatic children compared to symptomatic 
asymptomatic and symptomatic children. However, the and healthy controls (Fig.  4c). Increased levels of the 
levels of CD25+FOXP3− on CD4+ T cells were signifi- CD4+CD25+FOXP3− activation marker were found in 
cantly lower in the healthy controls when compared to the asymptomatic group compared to the symptomatic 
the asymptomatic (P = 0.0063) or symptomatic children (P = 0.035) and control (P = 0.0007) groups. Likewise, 
(P = 0.0024). In addition, no significant difference was CD8+CD25+FOXP3− activated T cells were signifi-
observed in the expression of CD8+CD25+FOXP3− T cantly increased in the asymptomatic group compared 
cells across the study groups (P = 0.1192, Fig. 3b). to the symptomatic (P = 0.0208) and control (P < 0.0001) 
groups (Fig.  4c). In addition, a significant increase in 
Decreased expression of CD25+CD69+ T cell activation levels of double-positive CD4+CD25+CD69+ T cells 
markers during asymptomatic Plasmodium falciparum were observed in the asymptomatic children compared 
infections before infected red blood cell stimulation to the symptomatic children (P = 0.035) and controls 
The expression of CD69 activation marker on both (P = 0.0025, Fig.  4d). Likewise, higher expression of 
CD4+ and CD8+ T cell subsets before in vitro stimula- CD8+CD25+CD69+ cells was observed in the asymp-
tion was investigated in the three study groups (Fig. 2a). tomatic cohort when compared to the healthy controls 
For the PBMCs analysed ex  vivo, CD4+ and CD8+ T (P = 0.023) but not the symptomatic children (P > 0.05, 
cells expressing both CD25 and CD69 were higher in Fig. 4d).
the symptomatic when compared to the asymptomatic 
group (P = 0.016 and P = 0.0062) and healthy controls Cellular activation and Treg frequency govern parasitaemia 
(P = 0.047 and P = 0.0232), respectively (Fig.  3c). How- and disease status
ever, no significant difference was observed between the Correlations  between the 24 considered  T cell phe-
asymptomatic group and controls (P > 0.05) for both T notypes (including the pre- and post-iRBC stimula-
cell sub-sets (Fig. 3c). tion levels of Tregs) and parasite control (as measured 
by parasitaemia levels) were investigated. After apply-
Increased expression of activation marker on T cells ing a Bonferroni correction for multiple comparisons, 
in asymptomatic infections after infected red blood cell significant positive correlations were found  between 
stimulation parasitaemia and the levels of both CD8+CD69+ 
When PBMCs were stimulated in vitro with iRBCs, levels (r = 0.4128658, P = 0.0016) and CD8+CD25+CD69+ 
of activated CD4+CD69+ T cells in the asymptomatic (r = 0.4070214, P = 0.0018) T cells measured before iRBC 
(See figure on next page.)
Fig. 4 Expression of activation markers CD25/CD69 on T-cells from the study cohorts after iRBC stimulation. PBMCs were stimulated with iRBC 
lysates (iRBCs) to determine the levels of activation markers (CD25/CD69) on both CD4+ and CD8+ T cell sub-sets. The percentage expression 
of a CD25−CD69+; b CD25+CD69−; c CD25+FOXP3−; and, d CD25+CD69+ T cells was analysed in healthy controls (n = 17), asymptomatic 
P. falciparum-infected children (n = 18) and symptomatic P. falciparum-infected children (n = 21). The data are presented as box plots with 
inter-quartile ranges. The 10th and 90th percentiles presented as whiskers. Medians are indicated by the horizontal lines across the boxes. The 
Kruskal–Wallis test was used for statistical comparisons between groups. P values < 0.05 were considered to be significant after Dunn’s test to 
correct for multiple comparisons
Frimpong et al. Malar J  (2018) 17:263 Page 9 of 13
a C D 4 + C D 8 +
100 P < 0 .0 0 0 1
100 P = 0 .0 0 1 7
P = 0 .0 0 0 8 P = 0 .0 0 0 2
P = 0 .0 0 5 4 P = 0 .0 0 5 7 iR B C s
50 50
0 0
l t ico t ic lr a o t i
c ic
n t
a tr a a t
o om m n
C t to o to
m
omp p C p p t
s y
m
S y
m ym ym
A A s S
b C D 4 + C D 8 +
100
100 P = 0 .1 6 3 0P = 0 .4 9 7 1
50 50
0 0
l l c c
ro a t
ic ic o t i t i
n t
a t tr a a
o m m
n m m
C p t
o
p t
o oC o op t p t
s y
m m
S y
m m
A s
y
A S
y
c C D 4 + C D 8 +
80 P < 0 .0 0 0 1
80 P = 0 .0 0 0 8
60 P < 0 .0 0 0 1 P = 0 .0 2 0 8
60
P = 0 .0 0 0 7 P = 0 .0 3 5
40
40
20 20
0 0
l t ic ic o
l t ic c
tro a a t
i
t r a a
t
n m m
C o
n om om o
p t p t C
o
p t t
o
p
m
ym ym msy
A s S A S
y
d C D 4 + C D 8 +
40 P = 0 .0 0 2 6
40 P = 0 .0 2 8 0
30 P = 0 .0 0 2 5 P = 0 .0 3 5 30 P = 0 .0 2 3 0
20
20
10 10
0 0
l t icro t
ic
a la i
c
t i
c
n m tr
o a t a t
o
C o t to
m n m m
p op C p t
o
p t
o
ym ym ym ym
A s S sA S
% C D 2 5 +C D 6 9 + T c e lls % C D 25 + FO X P 3 - T c e lls % C D 25 + C D 6 9 - T c e lls % C D 25 -C D 6 9 + T c e lls
% C D 2 5 +C D 6 9 + T c e lls % C D 25 + FO X P 3 - T c e lls % C D 25 + C D 6 9 - T c e lls % C D 25 -C D 6 9 + T c e lls
Frimpong et al. Malar J  (2018) 17:263 Page 10 of 13
stimulation, and the levels of both CD4+CD25+Foxp3+ of the infected children and then used to predict disease 
(r = 0.4772815, P = 0.0002) and CD8+CD25+Foxp3+ severity in the remaining children. Using a 5-fold cross-
(r = 0.4772714, P = 0.0003) T cells measured after stimu- validation analysis to prevent overfitting, it was found 
lation. Strikingly, levels of these four T cell phenotypes that the model accurately distinguishes between asymp-
together accounted for 68% of the variation in parasitae- tomatic and symptomatic children, with a sensitivity of 
mia observed in asymptomatic and symptomatic children 86%, a specificity of 94%, and an area under the receiver-
(Additional files 2, 3). operator-characteristic curve (AUC) of 90% (Fig.  5). 
Machine learning was used to  determine whether the Together, the results suggest that expression levels of the 
levels of these four T cell phenotypes alone could be used considered regulatory and activation markers determine 
to predict disease status in infected children. A model most of the individual variation in parasitaemia and pre-
based on a support vector machine was fitted to the lev- dict disease status in asymptomatic and symptomatic P. 
els of the four T cell phenotypes measured in a sub-set falciparum infections.
Fig. 5 T cell regulatory and activation markers distinguish between asymptomatic and symptomatic Plasmodium falciparum infections. It 
was assessed whether a machine-learning model based on pre-iRBC stimulation levels of CD8+CD69+ and CD8+CD25+CD69+ T cells and 
post-stimulation levels of CD4+CD25+Foxp3+ and CD8+CD25+Foxp3+ T cells could accurately predict disease status in asymptomatic and 
symptomatic children. Thirty-eight children had data for all the 4 T cell phenotypes considered. The children were randomly separated into 5 
groups. Fixing one group as a test group, we trained a machine-learning model (precisely a support vector machine) on the other 4 groups and 
then predicted the disease status of children found in the test group. The process was repeated until each of the groups was used exactly once as a 
test group. The plot shows a representative heatmap of the predicted probability that each child is either asymptomatic or symptomatic. Strikingly, 
as expected, most asymptomatic (respectively symptomatic) children have a higher predicted probability of being asymptomatic (respectively 
symptomatic)
Frimpong et al. Malar J  (2018) 17:263 Page 11 of 13
Discussion It has recently been shown that Tregs expressing 
The aim of this study was to investigate the frequency of CTLA-4 in murine models of malaria interfere with the 
activated Tregs and T cell early and late activation mark- acquisition of long-term immunity to malaria infec-
ers during P. falciparum infections, and the correlations tions [13]. The increase in CTLA-4 in Tregs observed 
between these and levels of parasitaemia. It was observed in individuals with P. falciparum infections compared 
that asymptomatic infections are associated with lower to uninfected controls could reflect their direct role in 
levels of Tregs with reduced Treg activation, and reduced controlling immune responses during human malaria 
expression of T cell activation markers compared to infections. Also, the increased expression of CTLA-4 
symptomatic infections. Also, T cells from asymptomatic on Tregs in the symptomatic children suggests that 
P. falciparum-infected children were more responsive to immune regulation associated with clinical malaria may 
iRBC stimulation compared to cells from symptomatic affect cellular activation, consequently, affecting the 
children. Importantly, the measured variations in regula- downstream development of anti-malaria immunity.
tory and activation marker levels explained most (68%) of Importantly, persistent immune activation has been 
the variation in parasitaemia observed in asymptomatic described as a major factor in predicting disease with 
and symptomatic infections. These results indicate that increased levels of activation being associated with 
in contrast to children with symptomatic malaria, there clinical disease progression [34–36]. Resting T cells are 
seems to be appropriate levels of immune regulation and identified phenotypically by the absence of CD25/CD69 
activation in children with asymptomatic malaria, which markers [37]. In this study, activated T cells were clas-
favor the control of parasitaemia. Another, non-mutually sified by the expression of CD25+/CD69+ markers. A 
exclusive possibility not ruled out by the analyses is that significant increase in immune activation in the CD4 
asymptomatic children might have higher levels of pro- and CD8 T cells was observed in clinical malaria. This 
tective antibodies compared to symptomatic children, is in line with a previous study that observed increased 
which might contribute to the observed differences in immune activation during clinical malaria infections 
parasitaemia. [38]. However, it should be noted that the increased 
Previous data have shown that malaria-exposed indi- activation in symptomatic children may not directly 
viduals can harbour infection without clinical symptoms, connote an effective T cell response since cytokine pro-
implying that there is some level of immune restriction on files were not measured.
parasite replication [5]. Increased levels of Tregs have been In contrast, in the asymptomatic children, it was 
associated with higher parasitaemia [10, 11] and delayed found that fewer cells expressed both activation mark-
parasite clearance [20] as well as the development of clini- ers on either CD4 or CD8 T cells indicating reduced 
cal disease [15, 32, 33]. In this study, it was found that the cellular activation when compared to children with 
level of Tregs is higher in children with clinical malaria symptomatic malaria. A plausible interpretation of 
compared to children with asymptomatic infections and these results is that lower immune suppression by 
healthy controls. This supports findings from other stud- Tregs in asymptomatic children leads to more effector 
ies which have also associated increased Treg frequen- T cell activation and greater parasite control, which in 
cies with symptomatic malaria infections [9, 23]. Also, turn feeds back to reduce T cell activation. Conversely, 
the significant increase in the Treg frequency which was higher immune suppression in symptomatic children 
observed in the symptomatic children after iRBC stimula- might limit parasite control leading to higher levels 
tion may indicate that during clinical malaria Tregs from of parasitaemia and T cell activation. This is in line 
the pre-clinical state are expanded or being induced. with the observation that P. falciparum activates the 
The low levels of Tregs observed in the asymptomatic immune system in a dose-dependent manner [39]. In 
children corroborates the findings of Boyle et  al. [32] addition, the lack of symptoms, lower levels of immune 
who identified lower levels of Tregs in children with activation, and lower parasitaemia in the asympto-
asymptomatic infections. This was interesting since in matic children might also result from higher levels of 
another previous study by Jangpatarapongsa et  al. [7] parasite-specific antibodies that reduce parasitaemia 
they also identified lower amounts of Treg cytokines levels below the threshold required to induce a T cell 
in individuals with asymptomatic Plasmodium vivax response. Additional research is needed to elucidate 
infections, suggesting there is less Treg activation in these hypotheses.
individuals with asymptomatic Plasmodium infections. It has been shown that asymptomatic children main-
This supports the view that lower levels of Tregs may be tain levels of CD4+CD25+FOXP3- effector T cells 
associated with a decreased risk of developing clinical that co-produce IFNγ, TNF and IL-10, describing these 
disease and possibly an increased likelihood of develop- cells as self-regulatory [9, 40]. Even though lower lev-
ing immunity to malaria. els of Tregs and activated Tregs were observed in the 
Frimpong et al. Malar J  (2018) 17:263 Page 12 of 13
asymptomatic children, CD4+CD25+FOXP3− T cells Authors’ contributions
were not significantly different between the asympto- AF and WN conceived the idea, designed the experiments. WN, MFO and KAK supervised the work. AF performed the experiments in the study. BT assisted 
matic and symptomatic children. This may suggest that in the experiment. AF, KAK, MFO and WN wrote the paper. All authors read and 
during asymptomatic malaria, restriction of parasite rep- approved the final manuscript.
lication and inflammation may be mediated by these self- Author details
regulating effector T cells. 1 West African Centre for Cell Biology of Infectious Pathogens (WACCBIP), 
Unfortunately, this study had a number of limita- Department of Biochemistry, Cell and Molecular Biology, University of Ghana, 
tions since a longitudinal study could not be conducted Legon, P. O. Box LG 54, Accra, Ghana. 
2 Immunology Department, Noguchi 
Memorial Institute for Medical Research, College of Health Sciences, University 
to determine if any of the asymptomatic cohorts may of Ghana, P.O. Box LG 581, Accra, Ghana. 3 African Institute for Mathematical 
develop clinical disease because they were treated when Sciences, P.O. Box DL 676, Cape-Coast, Ghana. 4 African Institute for Math-
diagnosed. Consequently, the possibility that the immune ematical Sciences, University of Stellenbosch, 7 Melrose Rd, Muizenberg, Cape Town 7945, South Africa. 
dynamics observed reflect changes that occur during the 
natural course of P. falciparum infections could not be Acknowledgements
ruled out. In addition, because parasitaemia was  deter- The authors thank the parents, guardians and children who participated in this study. We are also grateful to the Directors and Clinical staff at the Shai-Osu-
mined by microscopy,  it was not possible to determine doku Health Directorate in the Shai-Osudoku district of Greater Accra region, 
conclusively that  none of the healthy cohort had sub- John Tetteh, Sophia Ampah, Jones Amponsah, Emmanuel K. Dickson and Alex 
microscopic infection. It was also, not possible to evaluate Danso-Coffie of the Immunology Department, NMIMR for their technical sup-port, and the University of Ghana’s Malaria Centre of Excellence.
the humoral response in the study population to deter-
mine its contribution to the immune dynamics observed. Competing interests
The authors declare that they have no competing interests.
Conclusion Availability of data and materials
The study shows evidence that Tregs are lower and asso- The datasets generated and/or analyzed during the current study are available 
ciated with reduced Treg activation in children with from the corresponding author(s) on reasonable request.
asymptomatic P. falciparum infections, which corre- Consent for publication
sponds to reduce cellular activation and lower levels of The authors have read and agreed to the content of this manuscript and its 
parasitaemia. Also, the greater expansion of activation publication upon acceptance.
markers after iRBC stimulation in asymptomatic chil- Ethics approval and consent to participate
dren compared to symptomatic children suggests that Ethical approval was obtained from the ethics committee of the Noguchi 
the former children harbour a larger latent repertoire Memorial Institute for Medical Research, University of Ghana, Accra, Ghana. Participation was voluntary and written informed consent and assent were 
of parasite-responsive T cells. Alternatively, this obser- obtained from parents/guardians and the children respectively.
vation could reflect less Treg-mediated suppression of 
T cell activation in cells from asymptomatic children. FundingAugustina Frimpong is supported by a Ph.D. fellowship from a World Bank 
Together, these data support the view that the dynamics African Centres of Excellence Grant (ACE02-WACCBIP: Awandare) and the 
of T cell regulation and activation may contribute to the International Development Research Center research grant from the African 
acquisition of anti-parasite and/or anti-disease immunity Institute for Mathematical Sciences, Ghana, the L’Oreal-UNESCO for Women in Science Grant, the Carnegie Corporation of New York and the University of 
to malaria. Insights into these dynamics might inform the Ghana BanGA Ph.D. Research Grant.
development of malaria vaccines that induce appropriate 
levels of cellular activation and regulation as well as opti- Publisher’s Note
mal control of parasitaemia and disease. Springer Nature remains neutral with regard to jurisdictional claims in pub-
lished maps and institutional affiliations.
Additional files
Received: 29 March 2018   Accepted: 7 July 2018
Additional file 1. Antibody panel and clones used.
Additional file 2. Linear regression analysis to determine the level of vari-
ation in parasitaemia using 4 T cell phenotypes. References
Additional file 3. Diagnostic plots for the regression analysis described in  1. Gonçalves BP, Huang C-Y, Morrison R, Holte S, Kabyemela E, Prevots DR, 
the legend of Additional file 2. et al. Parasite burden and severity of malaria in Tanzanian children. N Engl 
J Med. 2014;370:1799–808.
 2. Sissoko MS, Healy SA, Katile A, Omaswa F, Zaidi I, Gabriel EE, et al. Safety 
and efficacy of PfSPZ vaccine against Plasmodium falciparum via direct 
Abbreviations
venous inoculation in healthy malaria-exposed adults in Mali: a ran-
Tregs: regulatory T cells; CTLA-4: cytotoxic T-cell lymphocyte antigen 4; PBMC: 
domised, double-blind phase 1 trial. Lancet Infect Dis. 2017;17:498–509.
peripheral blood mononuclear cells; iRBCs: infected red blood cells; uRBCs: 
 3. White MT, Verity R, Griffin JT, Asante KP, Owusu-Agyei S, Greenwood B, 
uninfected red blood cells; IFNγ: interferon gamma; TNF: tumour necrosis fac-
et al. Immunogenicity of the RTS, S/AS01 malaria vaccine and implica-
tor; IL-10: interleukin 10; DMSO: dimethyl sulfoxide.
tions for duration of vaccine efficacy: secondary analysis of data from a 
phase 3 randomised controlled trial. Lancet Infect Dis. 2015;15:1450–8.
Frimpong et al. Malar J  (2018) 17:263 Page 13 of 13
 4. Osier FH, Feng G, Boyle MJ, Langer C, Zhou J, Richards JS, et al. Opsonic  22. Torcia MG, Santarlasci V, Cosmi L, Clemente A, Maggi L, Mangano VD, 
phagocytosis of Plasmodium falciparum merozoites: mechanism in et al. Functional deficit of T regulatory cells in Fulani, an ethnic group 
human immunity and a correlate of protection against malaria. BMC with low susceptibility to Plasmodium falciparum malaria. Proc Natl Acad 
Med. 2014;12:108. Sci USA. 2008;105:646–51.
 5. Portugal S, Pierce SK, Crompton PD. Young lives lost as B cells falter:  23. Hansen DS, Schofield L. Natural regulatory T cells in malaria: host or 
what we are learning about antibody responses in malaria. J Immunol. parasite allies? PLoS Pathog. 2010;6:e1000771.
2013;190:3039–46.  24. Burel JG, Apte SH, Groves PL, Boyle MJ, Langer C, Beeson JG, et al. 
 6. McCallum FJ, Persson KE, Mugyenyi CK, Fowkes FJ, Simpson JA, Richards Dichotomous miR expression and immune responses following primary 
JS, et al. Acquisition of growth-inhibitory antibodies against blood-stage blood-stage malaria. JCI Insight. 2017;2:e93434.
Plasmodium falciparum. PLoS ONE. 2008;3:e3571.  25. Sancho D, Gómez M, Sánchez-Madrid F. CD69 is an immunoregulatory 
 7. Jangpatarapongsa K, Chootong P, Sattabongkot J, Chotivanich K, Siri- molecule induced following activation. Trends Immunol. 2005;26:136–40.
chaisinthop J, Tungpradabkul S, et al. Plasmodium vivax parasites alter the  26. Ziegler SF, Ramsdell F, Hjerrild KA, Armitage RJ, Grabstein KH, Hennen KB, 
balance of myeloid and plasmacytoid dendritic cells and the induction of et al. Molecular characterization of the early activation antigen CD69: a 
regulatory T cells. Eur J Immunol. 2008;38:2697–705. type II membrane glycoprotein related to a family of natural killer cell 
 8. Kho S, Marfurt J, Noviyanti R, Kusuma A, Piera KA, Burdam FH, et al. activation antigens. Eur J Immunol. 1993;23:1643–8.
Preserved dendritic cell HLA-DR expression and reduced regulatory T cell  27. Ziegler SF, Ramsdell F, Alderson MR. The activation antigen CD69. Stem 
activation in asymptomatic Plasmodium falciparum and P. vivax infection. Cells. 1994;12:456–65.
Infect Immun. 2015;83:3224–32.  28. Clausen J, Vergeiner B, Enk M, Petzer AL, Gastl G, Gunsilius E. Functional 
 9. Walther M, Jeffries D, Finney OC, Njie M, Ebonyi A, Deininger S, et al. significance of the activation-associated receptors CD25 and CD69 on 
Distinct roles for FOXP3+ and FOXP3−CD4+ T cells in regulating cellular human NK-cells and NK-like T-cells. Immunobiol. 2003;207:85–93.
immunity to uncomplicated and severe Plasmodium falciparum malaria.  29. Mardiney M III, Brown MR, Fleisher TA. Measurement of T-cell CD69 
PLoS Pathog. 2009;5:e1000364. expression: a rapid and efficient means to assess mitogen-or antigen-
 10. Minigo G, Woodberry T, Piera KA, Salwati E, Tjitra E, Kenangalem E, et al. induced proliferative capacity in normals. Cytometry A. 1996;26:305–10.
Parasite-dependent expansion of TNF receptor II—positive regulatory  30. Ononye NC. Characterization of malaria transmission and insecticide suscep-
T cells with enhanced suppressive activity in adults with severe malaria. tibility status of Anopheles Gambiae Sensu Lato Gilles (Diptera: Culicidae). In: 
PLoS Pathog. 2009;5:e1000402. Shai-Osudoku District of Southern Ghana. University of Ghana; 2015.
 11. Walther M, Tongren JE, Andrews L, Korbel D, King E, Fletcher H, et al.  31. Musah A. Effectiveness of the National Malaria Control Programme in 
Upregulation of TGF-β, FOXP3, and CD4+CD25+ regulatory T cells Akwapim South Municipality. University of Ghana; 2013.
correlates with more rapid parasite growth in human malaria infection.  32. Boyle MJ, Jagannathan P, Farrington LA, Eccles-James I, Wamala S, McIn-
Immunity. 2005;23:287–96. tyre TI, et al. Decline of FoxP3+ regulatory CD4 T cells in peripheral blood 
 12. Finney OC, Riley EM, Walther M. Regulatory T cells in malaria—friend or of children heavily exposed to malaria. PLoS Pathog. 2015;11:e1005041.
foe? Trends Immunol. 2010;31:63–70.  33. Semple PL, Binder AB, Davids M, Maredza A, van Zyl-Smit RN, Dheda K. 
 13. Kurup SP, Obeng-Adjei N, Anthony SM, Traore B, Doumbo OK, Butler NS, Regulatory T cells attenuate mycobacterial stasis in alveolar and blood-
et al. Regulatory T cells impede acute and long-term immunity to blood- derived macrophages from patients with tuberculosis. Am J Respir Crit 
stage malaria through CTLA-4. Nat Med. 2017;23:1220–5. Care Med. 2013;187:1249–58.
 14. Wammes LJ, Wiria AE, Toenhake CG, Hamid F, Liu KY, Suryani H, et al.  34. Karim R, Mack WJ, Stiller T, Operskalski E, Frederick T, Landay A, et al. 
Asymptomatic plasmodial infection is associated with increased tumor Association of HIV clinical disease progression with profiles of early 
necrosis factor receptor II—expressing regulatory t cells and suppressed immune activation: results from a cluster analysis approach. AIDS. 
type 2 immune responses. J Infect Dis. 2013;207:1590–9. 2013;27:1473–81.
 15. Todryk SM, Walther M, Bejon P, Hutchings C, Thompson FM, Urban BC,  35. Fletcher HA, Snowden MA, Landry B, Rida W, Satti I, Harris SA, et al. T-cell 
et al. Multiple functions of human T cells generated by experimental activation is an immune correlate of risk in BCG vaccinated infants. Nat 
malaria challenge. Eur J Immunol. 2009;39:3042–51. Commun. 2016;7:11290.
 16. Couper KN, Blount DG, Wilson MS, Hafalla JC, Belkaid Y, Kamanaka M,  36. Hazenberg MD, Otto SA, van Benthem BH, Roos MT, Coutinho RA, Lange 
et al. IL-10 from CD4+CD25−Foxp3−CD127− adaptive regulatory T cells JM, et al. Persistent immune activation in HIV-1 infection is associated 
modulates parasite clearance and pathology during malaria infection. with progression to AIDS. AIDS. 2003;17:1881–8.
PLoS Pathog. 2008;4:e1000004.  37. Baldauf H-M, Pan X, Erikson E, Schmidt S, Daddacha W, Burggraf M, 
 17. Torres KJ, Villasis E, Bendezú J, Chauca J, Vinetz JM, Gamboa D. Relation- et al. SAMHD1 restricts HIV-1 infection in resting CD4+ T cells. Nat Med. 
ship of regulatory T cells to Plasmodium falciparum malaria symptomatol- 2012;18:1682–9.
ogy in a hypoendemic region. Malar J. 2014;13:108.  38. Worku S, Troye-Blomberg M, Christensson B, Björkman A, Fehniger 
 18. Gonçalves RM, Salmazi KC, Santos BA, Bastos MS, Rocha SC, Boscardin SB, T. Activation of T cells in the blood of patients with acute malaria: 
et al. CD4+CD25+Foxp3+ regulatory T cells, dendritic cells, and circulat- proliferative activity as indicated by Ki-67 expression. Scand J Immunol. 
ing cytokines in uncomplicated malaria: do different parasite species 2001;53:296–301.
elicit similar host responses? Infect Immun. 2010;78:4763–72.  39. Finney OC, Lawrence E, Gray AP, Njie M, Riley EM, Walther M. Freeze-thaw 
 19. Cambos M, Belanger B, Jacques A, Roulet A, Scorza T. Natural regulatory lysates of Plasmodium falciparum-infected red blood cells induce dif-
(CD4+CD25+FOXP+) T cells control the production of pro-inflammatory ferentiation of functionally competent regulatory T cells from memory T 
cytokines during Plasmodium chabaudi adami infection and do not cells. Eur J Immunol. 2012;42:1767–77.
contribute to immune evasion. Int J Parasitol. 2008;38:229–38.  40. Portugal S, Moebius J, Skinner J, Doumbo S, Doumtabe D, Kone Y, et al. 
 20. Hisaeda H, Tetsutani K, Imai T, Moriya C, Tu L, Hamano S, et al. Malaria Exposure-dependent control of malaria-induced inflammation in chil-
parasites require TLR9 signaling for immune evasion by activating regula- dren. PLoS Pathog. 2014;10:e1004079.
tory T cells. J Immunol. 2008;180:2496–503.
 21. Vigário AM, Gorgette O, Dujardin HC, Cruz T, Cazenave P-A, Six A, et al. 
Regulatory CD4+CD25+Foxp3+ T cells expand during experimental 
Plasmodium infection but do not prevent cerebral malaria. Int J Parasitol. 
2007;37:963–73.