Regulatory T Cells Correlates with More RapidParasite Growth in Human Malaria Infection
Michael Walther,1,7 Jon Eric Tongren,2,8
Laura Andrews,3 Daniel Korbel,2 Elizabeth King,2
Helen Fletcher,1 Rikke F. Andersen,3 Philip Bejon,1
Fiona Thompson,1 Susanna J. Dunachie,1
Fanny Edele,2,9 J. Brian de Souza,2,4
Robert E. Sinden,5 Sarah C. Gilbert,3
Eleanor M. Riley,2,6,* and Adrian V.S. Hill1,3,6
1Center for Clinical Vaccinology and Tropical MedicineNuffield Department of Clinical MedicineOxford UniversityChurchill HospitalOxford OX3 7LJUnited Kingdom2Department of Infectious and Tropical DiseasesLondon School of Hygiene and Tropical MedicineKeppel StreetLondon WC1E 7HTUnited Kingdom3The Wellcome Trust Centre for Human GeneticsNuffield Department of Clinical MedicineOxford UniversityRoosevelt DriveOxford OX3 7BNUnited Kingdom4Department of Immunology and Molecular PathologyRoyal Free and UCL Medical SchoolWindeyer Building46 Cleveland StreetLondon W1T 4JFUnited Kingdom5Department of Biological SciencesSir Alexander Fleming BuildingImperial College LondonImperial College RoadLondon SW7 2AZUnited Kingdom
Summary
Understanding the regulation of immune responsesis central for control of autoimmune and infectiousdisease. In murine models of autoimmunity andchronic inflammatory disease, potent regulatory Tlymphocytes have recently been characterized. De-spite an explosion of interest in these cells, their rele-vance to human disease has been uncertain. In a lon-gitudinal study of malaria sporozoite infection via thenatural route, we provide evidence that regulatory Tcells have modifying effects on blood-stage infectionin vivo in humans. Cells with the characteristics of
6 These authors contributed equally to this work. 7 Present address: MRC Laboratories, PO Box 273, Fajara, TheGambia, West Africa. 8 Present address: Division of Parasitic Diseases, Centers for Dis-ease Control, 1600 Clifton Road, N.E. Atlanta, Georgia 30333. 9 Present address: Department of Dermatology, University of Frei-burg, Hauptstrasse 7, DE-79104, Freiburg, Germany.
regulatory T cells are rapidly induced followingblood-stage infection and are associated with a burstof TGF-� production, decreased proinflammatory cy-tokine production, and decreased antigen-specificimmune responses. Both the production of TGF-�and the presence of CD4+CD25+FOXP3+ regulatory Tcells are associated with higher rates of parasitegrowth in vivo. P. falciparum-mediated induction ofregulatory T cells may represent a parasite-specificvirulence factor.
Introduction
Initially characterized by their ability to control autoim-mune disease in mice (Sakaguchi et al., 1995), regula-tory T cells have now been shown, in model systems,to influence immune responses to a range of microbialinfections (Rouse and Suvas, 2004). Among severalforms of antigen-specific T cells with regulatory proper-ties (Barrat et al., 2002; Chen et al., 1994; Groux et al.,1997), a population of CD4+ T cells that express CD25(IL-2Rα) in the resting state has been described in mice(Sakaguchi et al., 1995) and humans (Dieckmann et al.,2001; Jonuleit et al., 2001; Levings et al., 2001; Ste-phens et al., 2001; Taams et al., 2001). These cells spe-cifically express the X-linked forkhead/winged helixtranscription factor FOXP3 (foxp3 in mice) (Brunkow etal., 2001; Schubert et al., 2001), which has been associ-ated with their development and function in both miceand humans (Fontenot et al., 2003; Hori et al., 2003;Khattri et al., 2003; Yagi et al., 2004). Some CD4+CD25+
regulatory T cells are thymus derived and are now des-ignated “endogenous T reg” or “natural T reg,” butthere is increasing evidence in mice and humans thatCD4+CD25+FOXP3-expressing T cells may be very rap-idly induced in the periphery (Chen et al., 2003; Fantiniet al., 2004; Park et al., 2004; Schramm et al., 2004;Zheng et al., 2004). Whatever their origin, T reg appearto mediate their effects either by direct cell contact(Thornton and Shevach, 2000) or by induction of theimmunoregulatory cytokines IL-10 or TGF-β (Powrie etal., 2003). In experimental situations, such as Leish-mania major infection of resistant mouse strains (Bel-kaid et al., 2002) or Herpes simplex virus infection ofthe murine cornea (Suvas et al., 2004), interactions be-tween pathogens and T reg have been shown to be ofmutual benefit to pathogen and host, allowing infectionto persist, thereby maintaining long-term memory andresistance to reinfection (Belkaid et al., 2002), whileminimizing T cell-mediated immunopathology (Suvas etal., 2004). However, disrupting the balance of T reg toT effector cells can lead to disease reactivation (Men-dez et al., 2004). Whether T reg play a similar role innatural host-parasite interactions is currently unknown,as is their potential role in modulating the human im-mune response to infection.
Malaria infection is characterized, following emer-gence of parasites from the liver, by an early phase oflogarithmic parasite replication in circulating red blood
Immunity288
cells, which is eventually brought under control by im-mune effectors; this is typically followed by low-grade,subclinical parasite persistence. In murine malaria, reg-ulation of Th-1 effectors by IL-10 (Li et al., 2003) andTGF-β (Omer and Riley, 1998) is essential for limitingimmunopathology, but induction (or administration) ofTGF-β during early, acute blood-stage infection sup-presses protective proinflammatory cytokine responses,leading to unrestrained parasite proliferation and in-creased disease severity (Omer et al., 2003b; Tsutsuiand Kamiyama, 1999). In a malaria model where dis-ease is exacerbated by early TGF-β production, neu-tralization of IL-10 and TGF-β signaling (Omer et al.,2003b), or depletion of CD25+ cells (Hisaeda et al.,2004), restores an early Th-1 response, retards earlyparasite replication, and allows the infection to be re-solved.
Unraveling the role of T reg in human infection ishampered by difficulties of following individual patientsfrom the uninfected state to acute infection and subse-quently to either chronic infection or cure. We havetherefore exploited the opportunity for studying the nat-ural history of human Plasmodium falciparum infectionfollowing experimental sporozoite infection of humanvolunteers via infectious mosquito bites within the con-text of phase IIa clinical trials to evaluate the efficacyof preerythrocytic stage malaria vaccines (McConkeyet al., 2003).
Results
We first assessed parasite density (detection of P. falci-parum DNA by quantitative PCR) and circulating cyto-kine concentrations in 14 vaccinated (but not pro-tected) individuals and 12 unvaccinated individualsover a period of 35 days, following live P. falciparumsporozoite infection. Serum and platelet-depletedplasma collected prior to sporozoite infection (day 0), 4days after infection, and every 12 hr from 6 days postin-fection until the first microscopic detection of parasi-taemia (at which point subjects were treated with a fullcurative dose of artemether with lumefantrine) weretested by multiplex cytokine bead array and (for latentand bioactive TGF-β) by ELISA. Blood samples col-lected every 12 hr over a period of 4 days from 2 unvac-cinated, uninfected subjects provided control data forphysiological variation in plasma cytokine concentra-tions. Data from the vaccinated and unvaccinatedgroups were pooled, as the vaccination regime em-ployed had no detectable protective effect and therewas no consistent or significant difference in values forany of the parameters measured between the twogroups; this can be further justified on the basis thatany vaccine-induced immune response would be spe-cific for preerythrocytic antigens and would be ex-pected to have no effect on subsequent blood-stage in-fection.
Baseline TGF-β concentrations for each individualwere calculated as the mean of the day 0 and day 4values (i.e., prior to emergence of parasites from theliver) and were no different from those observed in un-vaccinated, unchallenged controls (mean latent TGF-β:measured in nonvaccinated sporozoite infected = 8,129
pcms19
TfIvocid(19bfwca(tl0
frwbpspT9ps
lsePosa
RPWpmddpmw
βtib
g/ml [CI 95% 3,593–12,665], n = 12; and uninfectedontrols = 6,787 pg/ml [CI 95% 5,293–8,280], n = 16easurements; p = 0.5. Mean bioactive TGF-β: mea-
ured in all sporozoite infected = 23.1 pg/ml [CI 95%8.5–27.5], n = 26; uninfected controls = 19.5 pg/ml [CI5% 16.4–22.5], n = 16; p = 0.24).
GF-� Induction on Parasite Emergencerom the Livern 8 of the 12 unvaccinated subjects and in 5 of the 14accinated (but not protected) subjects, a marked peakf circulating, bioactive TGF-β was observed in plasmaollected within ±12 hr of the first detection of circulat-
ng P. falciparum DNA, which occurred on average 7.6ays (CI 95% 7.3–8) after sporozoite-initiated infection
data for 5 individuals are shown in Figure 1A). In these3 subjects, TGF-β peak concentrations were on average.2-fold (CI 95% 5.3–13.2) higher than their individualaseline TGF-β concentrations, in contrast to the 1.8-
old (CI 95% 1.2–2.4) increase observed in 13 subjectsithout a TGF-β peak (Figure 1B). Mean maximum TGF-βoncentrations for subjects with and without a TGF-β peakre shown in Table 1. Taking the whole cohort together
n = 26), the median maximum concentrations of bioac-ive TGF-β within 12 hr of parasite emergence from theiver were significantly higher than at baseline (p =.0001, Figure 1C).Latent TGF-β concentrations, assessed in samples
rom unvaccinated, sporozoite-infected subjects, mir-ored concentrations of bioactive TGF-β but were
100-fold higher (Table 1); the peak of spontaneouslyioactive TGF-β coincided with, or was preceded by, aeak in latent TGF-β. In 17/26 (65%) subjects, includingome subjects who did not show a distinct cytokineeak at the time of parasite emergence from the liver,GF-β levels began to fluctuate from approximately dayonward and remained slightly above baseline until
arasitaemia became patent by microscopy and theubjects were treated (Figures 1A and 1B).These observations are consistent with release of
atent, biologically inactive TGF-β from intracellulartores and its rapid conversion to the bioactive formither by endogenous activators or by components of. falciparum schizonts (Omer et al., 2003a) at the timef emergence of merozoites from hepatic schizonts; theubsequent TGF-β peak at 9–10 days may be associ-ted with the first wave of erythrocytic schizogony.
educed Proinflammatory Response and Enhancedarasite Growth in Presence of TGF-�e compared proinflammatory cytokine responses and
arasite growth rates between the 13 individuals whoade an early TGF-β peak and the 13 individuals whoid not. Although initial parasite growth rates were notifferent between the two groups, from day 8.5 onwardarasitaemia rose significantly faster in individuals whoade an early plasma TGF-β response than in thoseho did not (p = 0.008; Figure 2).Although similar proportions of subjects in the TGF-
-positive group (10/13; 77%) and in the TGF-β-nega-ive group 9/13 (69%) showed a 2-fold or more increasen serum IFN-γ concentration above their individualaseline, upregulation of IFN-γ occurred almost a day
Regulatory T Cells in Human Malaria289
Figure 1. Circulating Cytokine Concentrations at Time of Parasite Emergence from the Liver
(A and B) The kinetics of spontaneously bioactive TGF-β in plasma after sporozoite challenge are shown for (A) 5 subjects who showed atransient increase in TGF-β within ±12 hr of the first detection of circulating P. falciparum DNA, and (B) 5 individuals who did not show a TGF-β peak.(C) TGF-β level at baseline (mean TGF-β concentration on days 0 and 4) is compared to maximum TGF-β concentration within ±12 hr of firstdetection of parasite DNA.(D) Comparison of TGF-β (dotted line) and IFN-γ (continuous line) kinetics in subjects with (upper graph) and without (lower graph) initial TGF-β peak.
TGF-β-positive subjects was almost three times lower
Table 1. Circulating Concentrations of TGF-β and IFN-γ in Subjects with and without a Peak of TGF-β Production at the Time of Parasite Emergencefrom the Liver
With TGF-β Peak Without TGF-β Peak p
All TGF-β+
versus AllUnvaccinated Vaccinated All Unvaccinated Vaccinated All TGF-β−
Number of subjects 8 5 13 4 9 13Max. bioactive TGF-β at 191 (95–287) 203 (34–371) 196 (125–267) 74 (23–124) 26 (18–34) 41 (22–58)
1D). Furthermore, levels of IL-6 and IL-12, determined
later in the TGF-β-positive group than in the TGF-β-negative group (mean 10.5 dpi versus 9.6 dpi, respec-tively). Furthermore, the peak IFN-γ concentration in
than in those subjects without raised TGF-β levels (Ta-ble 1; p = 0.035), and in some TGF-β-positive subjects,no upregulation of IFN-γ was observed at all (Figure
Immunity290
Figure 2. Accelerated Parasite Growth in Presence of TGF-βGeometric mean parasite density following sporozoite challenge insubjects with a TGF-β peak at parasite emergence from the liver(continuous line, n = 13) and subjects without a TGF-β peak (dottedline, n = 13). From day 8.5, parasites grow significantly faster insubjects with TGF-β peak as compared to subjects without TGF-βpeak (slope coefficient = 1.2, p = 0.008)
iBsgnc
TwTCiioepelqz
s
in unvaccinated subjects only, were 2.5 and 2 timeslower, respectively, in TGF-β-positive subjects than inTGF-β-negative subjects, were detected in a lower pro-portion of subjects, and occurred a day (IL-6) or 2.5days (IL-12) later (data not shown). These observationssuggest that TGF-β may facilitate initial blood-stageparasite growth by suppressing proinflammatory re-sponses.
It is not yet clear why cytokine responses differamong individuals. We consider it very unlikely that thisis due to differences between individuals in the size ofthe sporozoite inoculum, as the estimated parasite bur-dens in the liver, calculated as described (Bejon et al.,2005), were very similar (p = 0.99) in the two groups.
Monocytes Are the Major Source of Early TGF-�To investigate the potential cellular source of TGF-β,PBMCs collected on days 4 and 7 after sporozoite in-fection from three subjects who had a peak of plasmaTGF-β on day 7 and from three subjects without aTGF-β peak were stained ex vivo, without restimulation,for intracellular latent TGF-β and expression of CD4,CD8, or CD14 (Figure 3). In all subjects, irrespective oftheir TGF-β status, the proportion of CD4+ cells de-clined and the percentage of CD14+ cells increasedfrom day 4 to day 7 (CD4+: median day 4 = 50.2%, day7 = 37.9%, p = 0.028; CD14+: median day 4 = 0.98%,day 7 = 5.9%, p = 0.028). Subjects with a plasma TGF-βpeak were found to have increased proportions ofTGF-β-positive cells on day 7 (median 7.64% of all via-ble PBMC), as compared to subjects without a TGF-βpeak (median 4.74%). As depicted in Figure 3D, the in-creased proportion of TGF-β-positive cells was due toan increase in the percentage of CD14+ TGF-β+ cells(the median percentage of CD14+ cells that were TGF-βpositive increased from 73% to 90% in subjects with aTGF-β plasma peak but declined from 83% to 58% insubjects without a TGF-β peak). Very few CD4+ or CD8+
cells were TGF-β positive, but a population of TGF-β+
cells that were neither monocytes nor T cells was seen
apiaap0
wpt1si2CtTT(Fafpijdpna
PSTPpiCfat
n all donors (Figure 3D); these cells may be NK cells orcells. In conclusion, the raised levels of TGF-β ob-
erved in some subjects at the time of parasite emer-ence from the liver appears to be due to increasedubers of CD14+ cells and a higher proportion of theseells making TGF-β.
GF-� Concentration Correlatesith Upregulation of T reghe observation that short-term culture of naive CD4+
D25− T cells with antigen in the presence of TGF-βnduces expression of FOXP3 and their developmentnto cells with the phenotypic and functional propertiesf T reg (Chen et al., 2003; Fantini et al., 2004; Parkt al., 2004; Schramm et al., 2004; Zheng et al., 2004)rompted us to compare FOXP3 expression in periph-ral blood mononuclear cells from unvaccinated, ma-
aria-infected subjects over time. FOXP3 mRNA wasuantified by real-time PCR in 10 unvaccinated, sporo-oite-infected individuals.As shown in Figure 4A, FOXP3 expression increased
ignificantly between day 7 and day 10 pi (p = 0.008)nd remained significantly elevated until at least day 35i (p = 0.021), and, even in this small data set, the fold
ncrease in FOXP3 from days 7 to 10 was significantlynd positively correlated with the fold increase in bio-ctive TGF-β (TGF-β level at first detection of P. falci-arum DNA compared to TGF-β level at baseline; r =.653, p = 0.041) (Figure 4B).To determine whether FOXP3 expression coincidesith the expansion of a regulatory T cell population, theroportion of T cells expressing the phenotypic charac-eristics of T reg was determined by flow cytometry in0 malaria-infected subjects. As suppressive functionegregates with the minor subset of CD4 cells exhibit-ng the CD25hi phenotype (Baecher-Allan et al., 2001,005), gates were set to include only the 2%–3% ofD4 cells expressing the highest levels of CD25. Fur-
hermore, as CD69 is upregulated on activated effectorcells but not on T reg (Park et al., 2004), we definedregs as being CD3+CD4+CD25hiCD69− lymphocytes
Figures 4C–4E). Consistent with the upregulation inOXP3 expression, the proportion of CD25hiCD69− cellsmong CD4+CD3+ lymphocytes increased significantlyrom day 7 to day 10 (Figure 4F). Importantly, the pro-ortion of CD4+ cells that are T reg shows a significant
nverse correlation with parasite growth rates; in sub-ects with a higher percentage of CD4+ T reg, parasiteensities of 1000 parasites/ml (or the highest recordedarasite count if less than 1000/ml) were reached sig-ificantly earlier than in subjects with lower percent-ges of T reg (r = −0.748, p = 0.013) (Figure 4G).
. falciparum-Induced T Reg Suppress Antigen-pecific Immune Responseso confirm the suppressive effect of putative T reg,. falciparum schizont extract (PfSE)-induced IFN-γroduction and lymphocyte proliferation was assessed
n PBMC with or without magnetic bead depletion ofD25hi cells (Figures 5A and 5B); PBMC were collected
rom 11 malaria-infected individuals on days 0, 7, 10,nd 35 pi and cultured with antigen for 72 hr. Prolifera-ion and IFN-γ production in response to culture me-
Regulatory T Cells in Human Malaria291
Figure 3. Increased Number of TGF-β+
CD14+ Cells among PBMC of Subjects withRaised Plasma TGF-β Concentrations
(A–C) PBMC collected on days 4 and 7 post-infection from 3 subjects with and 3 subjectswithout a peak of bioactive TGF-β on day 7were stained for CD4/CD8/CD14 and intra-cellular TGF-β. TGF-β-positive cells (con-tinuous line) were gated (horizontal bar) bycomparison with cells stained with isotypecontrol (dotted line) (A) and examined for ex-pression of CD4, CD8 (B), or CD14 (C).(D) The proportion of cells producing TGF-βand the contribution of CD4+, CD8+, andCD14+ cells producing TGF-β is shown foreach individual.
dium or uninfected RBC (uRBC) were consistently low,with no significant difference between depleted and un-depleted samples at any time point. By contrast, stimu-lation with PHA resulted in a strong proliferative re-sponse and induced high levels of IFN-γ at all timepoints, and these responses did not differ significantlybetween depleted and undepleted cells.
Figure 5 shows the effects of depletion of CD25hi
cells on the proliferative (Figure 5C) and IFN-γ (Figure5D) responses of PBMC, expressed as the ratio of stim-ulation indices (SI; cpm value or IFN-γ concentration forPfSE-stimulated cells/uRBC-stimulated cells) of de-pleted versus undepleted cells.
On days 0 and 7, CD25hi depletion did not affecteither proliferation or IFN-γ production, which is consis-tent with there being no change in T reg numbers orFOXP3 expression (Figure 4) and no effect of TGF-β onparasite densities at these time points (Figure 2). How-ever, for PBMC collected on day 10, by which timeFOXP3 expression and the proportion of CD4+CD25hi
CD69− T cells were significantly increased, depletion ofCD25hi cells resulted in a 2.34-fold increase in prolifera-tion, which was significantly higher than at day 7 (p =0.002) (Figure 5C). A similar, albeit not significant, trendwas observed for IFN-γ production (Figure 5D). Therewas a trend for the increase in proliferation in CD25+-depleted cultures to be greater for the TGF-β-positivedonors than for the TGF-β-negative donors and for en-hancement of both proliferation and IFN-γ productionto be greater in donors with a higher percentage of cellswith a T reg phenotype (data not shown). Importantly,the increase in proliferation after CD25hi depletion ofcells collected on day 10 correlates significantly withthe level of FOXP3 measured on day 10 (correlation co-efficient: 0.762, p = 0.01), supporting the view that
FOXP3 is a reliable marker for functional T reg. Takentogether, these data indicate that P. falciparum-inducedCD4+CD25hi cells are able to regulate the response ofP. falciparum-responsive effector cells.
Discussion
This study of immune responses during the prepatentphase of human P. falciparum malaria infection has re-vealed a burst of systemic TGF-β activity in 50% of in-fected subjects within 12 hr of the first detection ofblood-stage infection. This early burst of TGF-β activityis strongly associated with suppression of proinflam-matory cytokine responses, faster parasite growth, andinduction of a population of CD4+ T cells with the hall-marks of T reg. Therefore, this study suggests that notonly may T reg facilitate establishment of infection, butan individual’s propensity for T reg induction or activa-tion may be a key component of susceptibility to infec-tious disease. These data thus provide a firm founda-tion for future studies of the role of TGF-β and T reg inmalaria-endemic populations.
The marked increase in plasma levels of bioactiveTGF-β (compared with the much less significant in-crease in latent TGF-β) indicates that P. falciparum maydirectly activate latent TGF-β to its bioactive form in vivo,in a manner analogous to our previous in vitro observa-tions in which products of P. falciparum-schizont-infected erythrocytes activate latent TGF-β in a two-stepprocess involving parasite-derived metalloproteinasesand a thrombospondin-like molecule (Omer et al., 2003a).The reason for upregulation of TGF-β in only some sub-jects is unclear but, given that all subjects were in-fected with equivalent numbers of genetically identicalparasites, may reflect host genetic variation.
Immunity292
Figure 4. Upregulation of FOXP3 and T Reg Cells during Blood-Stage Infection
(A) Changes in FOXP3/HPRT ratio over time were determined in samples from 10 nonvaccinated, sporozite-challenged volunteers.(B) Fold increase of bioactive TGF-β (time of first detection of P. falciparum DNA versus baseline) correlates with fold increase in FOXP3 (day7 versus day 10).(C–E) T reg were identified as live cells that coexpressed CD4 and CD3 (R6) (C) and were simultaneously CD69neg (R3) (D) and CD25hi (R4)(E). (Only cells gated as R6 are shown in [D] and [E].)(F) The proportions of CD25hi/CD69neg/CD3+/CD4+ lymphocytes (R4) were determined for 10 malaria-infected subjects in comparison to threemalaria naive unchallenged controls.(G) The proportion of T reg at day 10 correlated negatively with parasite growth rate (days from infection to parasite density of 1000/ml).Although a negative linear correlation is significant (r = −0.748, p = 0.013), a better fit is obtained with a first-order inverse polynomialrelationship (r = 0.83 p = 0.0028).
The correlation between early TGF-β activation, de-layed and diminished proinflammatory cytokine re-sponses, and enhanced parasite growth is reminiscentof infection of C57BL/6 mice with the lethal 17XL strainof P. yoelii, which also induces high levels of TGF-βwithin 24 hr of blood-stage infection; in this model, thecausal relationship between TGF-β induction, T reg ac-tivity, and clinical outcome is proven since neutraliza-tion of TGF-β and IL-10 signaling (Omer et al., 2003b)or in vivo depletion of CD25hi cells during the earlyphase of infection (Hisaeda et al., 2004) restored theIFN-γ response and allowed parasitaemia to be con-trolled.
pitptresepqt
In humans, the kinetics of TGF-β activation (peaklasma levels at day 7.5 pi with CD14+ monocytes be-
ng the major source), induction of FOXP3 and detec-ion of functional T reg (after day 7 and before day 10i), and IFN-γ upregulation or suppression (day 10.5 pi),ogether with recent reports that TGF-β may induce Teg cells (Chen et al., 2003; Fantini et al., 2004; Parkt al., 2004; Schramm et al., 2004; Zheng et al., 2004),uggests the following sequence of events. In the pres-nce of malaria antigens, TGF-β may induce FOXP3 ex-ression in naive peripheral CD4+ T cells that subse-uently differentiate into T reg; these cells then suppresshe IFN-γ response, leading to less constrained parasite
Regulatory T Cells in Human Malaria293
Figure 5. Depletion of CD25hi Cells EnhancesPBMC Proliferative and IFN-γ Responses toMalaria Antigens
(A and B) CD25hi cells were depleted fromPBMC from 11 donors collected on days 0,7, 10, and 35 after sporozoite infection (plot[A] shows cells prior to depletion, [B] showscells after depletion) or PBMC were mockdepleted (not shown).(C and D) The increase in proliferation (C)and IFN-γ production (D) in response toPfSE, observed in CD25hi-depleted PBMC isshown, expressed as the ratio of stimulationindices (cpm value or IFN-γ concentration forPFSE-stimulated cells divided by cpm valueor IFN- γ concentration for uRBC-stimulatedcells) in depleted versus undepleted PBMC.
growth. It is possible that TGF-β directly exerts sus-tained inhibition of IFN-γ-producing T cells, but this isunlikely given the short half-life of this cytokine. How-ever, inhibition of the early burst of NK cell-derivedIFN-γ by TGF-β (as recently described in mice; Laouaret al., 2005) is a plausible explanation for our findings.A causal relationship between these events is sup-ported by the very tight correlations observed betweeneach step of the proposed pathway: upregulation ofTGF-β is strongly correlated with upregulation of FOXP3and with the increased proportion of fully functionalCD4+CD25hi T reg, while the proportion of T reg isstrongly correlated with parasite growth rates.
Our observation that TGF-β appears to emanate frommonocyte-macrophages is consistent with recent studiesin mice indicating important non-T cell sources forTGF-β (Fahlen et al., 2005; Marie et al., 2005). Further-more, it is consistent with recent studies in murine ma-laria showing that antigen-presenting cell populationscan modulate the T cell response during malaria infec-tions via release of modulatory cytokines, including IL-10 as well as TGF-β (Perry et al., 2005; K.N. Couper,D.G. Blount, J.B.d.S., and E.M.R., unpublished data),and with reports that malaria-infected red blood cellscan induce human dendritic cells to secrete anti-infla-matory cytokines (Urban et al., 1999).
Overall, our data demonstrate that numbers of regu-latory T cells increase upon infection with P. falciparumand, in parallel, parasite growth is facilitated by sup-pression of proinflammatory cytokine responses. Irre-spective of whether this represents de novo differentia-tion of T reg or expansion of a preexisting population,our data support the view that T reg can be induced inthe periphery via a mechanism that involves TGF-β andFOXP3. These cells are readily detected in peripheralblood, but we cannot rule out that cells are also se-
questered in the liver, spleen, or other organs whereparasitized erythrocytes accumulate. While initially ofbenefit to the parasite, facilitating the establishment ofblood-stage infection, P. falciparum-induced T reg maycontribute to the control of inflammatory responseslater in infection, thereby reducing immunopathologyand preventing the onset of severe malaria (de Souzaand Riley, 2002). In support of this possibility, immuno-epidemiological studies indicate that low levels of TGF-βare associated with acute (Wenisch et al., 1995) andsevere (Chaiyaroj et al., 2004; Perkins et al., 2000) ma-laria and that an imbalance between systemic levels ofproinflammatory cytokines and TGF-β increases therisk of clinical malaria (Dodoo et al., 2002). Thus, incases of mild clinical malaria, once parasitaemia is un-der control, T reg may produce—or induce productionof—TGF-β (Jonuleit et al., 2002) and IL-10 (Dieckmannet al., 2002; Jonuleit et al., 2002), which shut down theinflammatory response; individuals in whom this T regresponse is defective may then be at risk of pro-gressing to severe disease. Furthermore, by inhibitingongoing Th-1-mediated effector mechanisms, T reg ac-tivity may favor asymptomatic parasite persistence, fa-cilitating both long-term immunological memory (pre-munition) and parasite transmission to the next host.
These observations are important, as our study of therole of T reg in a natural infection of humans describesthe sequence of events and thus implies causal associ-ations. In human hepatitis C virus (HCV) infection, dem-onstration of T reg activity in chronically infected butnot recovered individuals (Cabrera et al., 2004; Sugi-moto et al., 2003) has been interpreted as evidence thatT reg activity can favor pathogen chronicity at the ex-pense of the host (who will eventually develop severedisease), but the cross-sectional nature of these studiesprecludes firm conclusions regarding cause and effect.
Immunity294
Studies of other chronic human infections (onchocerci-asis [Satoguina et al., 2002], human immunodeficiencyvirus [Aandahl et al., 2004; Kinter et al., 2004], and Heli-cobacter pylori [Lundgren et al., 2003]) suffer from thesame limitations.
In conclusion, the use of a well-established and safeparasite challenge model has allowed demonstration ofa direct correlation between activation of regulatory Tcells and enhanced blood-stage parasite growth, pro-viding evidence that FOXP3-expressing regulatory Tcells have infection-modifying activity in vivo in humans.
Experimental Procedures
Subject Recruitment, Vaccination, and Malaria Infection26 malaria-naive subjects that were exposed to the bites of fiveP. falciparum (strain 3D7) sporozoite-infected A. stephensi mosqui-toes in the context of two phase IIa vaccine trials (McConkey et al.,2003) were studied. 14 subjects had received experimental pre-erythrocytic malaria vaccines (FP9-MVA prime boost vaccines en-coding ME-TRAP [Bejon et al., 2005], CS, or both); 12 were unvac-cinated controls. All subjects were healthy, with a median age of22 years (range: 20–38 years). Venous blood was collected on day0, day 4, and twice daily from day 6.5 until the first microscopicdetection of parasitaemia, at which time subjects were cured withartemether/lumefantrine. A final blood sample was collected onday 35. The vaccination regimes employed did not induce any de-tectable protective immunity, and there was no significant differ-ence in the day of patency or subsequent parasite densities be-tween vaccinated and control subjects (M.W., unpublished data,data not shown). Serum and platelet-depleted plasma (van Waardeet al., 1997) was stored from each sample. PBMC were cryopre-served from control subjects on days 0, 4, 7, 10, and 35. Ethicalapproval was obtained from the Oxfordshire Research Ethics Com-mittee, UK, and the Human Subjects Protection Committee, Seat-tle, WA.
Multiplex Analysis of Serum Cytokine ConcentrationIL-6, IL-12p70, and IFN-γ were assayed in serum for each subjectand each time point using multiplex luminescent beads (Linco Re-search Inc., MO). Data were analyzed using StatLIA software (Bren-dan Scientific). The detection limit was 3.2 pg/ml for each of thecytokines tested; values below this threshold were set to 3.2 pg/ml.
TGF-� ELISAPlatelet-depleted plasma samples were assayed for TGF-β usingcommercial reagents (human TGF-β-1DuoSet, R&D Systems, Ab-ingdon, UK). Spontaneously bioactive TGF-β was measured di-rectly; total TGF-β (spontaneously bioactive plus latent) was mea-sured in samples after acid activation as described by themanufacturer. Latent TGF-β concentration was calculated by sub-traction of spontaneously bioactive TGF-β from total TGF-β. Thedetection limit was 15 pg/ml; extrapolated values below this cut-off were set to 15 pg/ml.
IFN-� ELISASerum samples from supernatants of the proliferation assay wereassayed for IFN-γ as described previously (Dodoo et al., 2002).
P. falciparum Detection by PCR and MicroscopyP. falciparum DNA was assayed by quantitative real-time PCR (An-drews et al., 2005), and at least 200 high-power fields of a thick filmwere examined by two expert slide readers for malaria parasites(McConkey et al., 2003).
P. falciparum Schizont AntigensFreeze-thawed lysates of P. falciparum-schizont-infected red bloodcells (PfSE) were prepared as described (Rhee et al., 2001) fromP. falciparum clone 3D7, maintained in continuous culture, and rou-tinely screened for mycoplasma by PCR (Bio Whittaker, Woking-
hw
FRF((taQUtaTG
CPgs(c1(awCnrmwtaard
FTmcFt[svPwcl
SQttmndcsgu
A
WtCaWLp
am, UK). Similar preparations of uninfected erythrocytes (uRBC)ere used as controls.
OXP3 Assay by RT-PCRT-PCR for FOXP3 was performed as described elsewhere (H.A.letcher et al., submitted). RNA was prepared from frozen PBMCs
Rneasy, Qiagen, Crawley, UK), stabilized with RNase inhibitorRNAsin, Promega/Invitrogen, Paisley, UK), and reverse transcribedo cDNA using oligo-dT-primers (MWG Biotech, Milton Keynes, UK)nd Omniscript reverse transcriptase (Omniscript kit, Qiagen).uantitative RT-PCR (ABI PRISM, Applied Biosystems, Warrington,K) was carried out using the SYBR Green (Qiagen) method with
he following primers: HPRT (F 5#-TATGGACAGGACTGAACGTC-3#nd R 5#-CTACAATGTGATGGCCTCCC-3#), and FOXP3 (F 5#-CACTACAGGCACTCCTCCAGG-3# and R 5#-CCACCGTTGAGAGCTGTGCAT-3#).
ell CultureBMCs were separated from heparinized blood by density centrifu-ation (Lymphoprep; Nycomed, Uppsala, Sweden) and cryopre-erved (5–10 × 106 PBMC/ml) in heat-inactivated fetal calf serumFCS) containing 10% DMSO (both Sigma, Poole, UK). For assays,ells were thawed, washed, and resuspended in RPMI 1640 with0% human serum, 100 IU/ml penicillin, 0.1 mg/ml streptomycin
all Sigma), and 2 mM L-glutamine (GIBCO/Invitrogen, Paisley, UK);ll samples from a single donor were analyzed concurrently. PBMCere split in two aliquots (each of 1.3 × 106 live cells in 1 ml); andD25hi (but not CD25low) cells were selectively removed using mag-etic beads (Dynal Biotech, Bromborough, UK) at a bead to PBMCatio of 7:1. Control cells were mock-depleted using sheep-anti-ouse-Ig-coated beads (Dynal Biotech). Depleted and control cellsere aliquoted (100 �l/well) into triplicate wells of 96× U-well plates
ogether with 5 × 104 pRBC or uRBC, 5 �g/ml PHA, or mediumlone and incubated for 5 days at 37°C in 5% CO2, and cell prolifer-tion was determined by [3H]-thymidine (Amersham, UK) incorpo-ation. Culture supernatant was saved for IFN-γ ELISA after 3.5ays.
low Cytometryhawed PBMC were directly stained using fluorochrome-labeledonoclonal mouse anti-human antibodies and appropriate isotype
ontrols (anti-CD25 FITC [IgG1], anti-CD4 PerCP [IgG1 k], anti-CD8ITC [IgG1], anti-CD69 PE [IgG1] [all BD Immunocytometry Sys-ems, Oxford, UK], anti-CD-14 APC [IgG2a], and anti-CD3 APCIgG2a] [both BD Pharmingen, Oxford, UK]), fixed (Cellfix, BD Bio-ciences), permeabilized, and stained for intracellular TGF-β (Arta-anis-Tsakonas and Riley, 2002) using anti-TGF-β1 PE (IgG1, IQroducts, Groningen, NL) or the appropriate isotype control. Cellsere analyzed by flow cytometry (FACSCalibur, BD Biosciences)ollecting a total of 100,000 events in a live gate; data were ana-
yzed using CellQuest (BD Biosciences).
tatistical Analysisuantitative data were assessed for normal distribution and log-
ransformed where appropriate. Correlation coefficients, Student’stest (paired or unpaired) comparing arithmetic or geometriceans, and Willcoxon rank sum or signed rank test performed on
onnormally distributed data were calculated using SPSS for Win-ows Version 11.0 (Lead Tools). 95% confidence intervals were cal-ulated using CIA 2.1 (Trevor Bryant, Southampton). Linear regres-ion analysis was performed to assess the difference of parasiterowth between subgroups with and without early TGF-β responsesing Stata 8.0 (StataCorp, Austin, TX).
cknowledgments
e are grateful to all the subjects who volunteered to take part inhe clinical studies, to Ian Poulton, Angela Hunt-Cooke, and Simonorrea for help with acquiring blood samples and slide reading,nd to Stephanie Sanos for assistance with the ABI PRISM cycler.e thank Jacqui Mendoza and Geoff Butcher (Imperial College,
ondon) and Dr. Jack Williams, WRAIR (Silver Spring, MD), for sup-lying P. falciparum-infected mosquitoes. This work was supported
Regulatory T Cells in Human Malaria295
by grants from the Wellcome Trust and the Malaria Vaccine Initia-tive at PATH. A.V.S.H. is a Wellcome Trust Principal ResearchFellow.
Received: May 15, 2005Revised: July 14, 2005Accepted: August 3, 2005Published: September 20, 2005
References
Aandahl, E.M., Michaelsson, J., Moretto, W.J., Hecht, F.M., andNixon, D.F. (2004). Human CD4+ CD25+ regulatory T cells controlT-cell responses to human immunodeficiency virus and cytomega-lovirus antigens. J. Virol. 78, 2454–2459.
Andrews, L., Andersen, R.F., Webster, D., Dunachie, S., Walther,R.M., Bejon, P., Hunt-Cooke, A., Bergson, G., Sanderson, F., Hill,A.B., and Gilbert, S.C. (2005). Quantitative real-time polymerasechain reaction for malaria diagnosis and its use in malaria vaccineclinical trials. Am. J. Trop. Med. Hyg. 73, 191–198.
Artavanis-Tsakonas, K., and Riley, E.M. (2002). Innate immune re-sponse to malaria: rapid induction of IFN-gamma from human NKcells by live Plasmodium falciparum-infected erythrocytes. J. Im-munol. 169, 2956–2963.
Baecher-Allan, C., Brown, J.A., Freeman, G.J., and Hafler, D.A.(2001). CD4+CD25high regulatory cells in human peripheral blood. J.Immunol. 167, 1245–1253.
Baecher-Allan, C., Wolf, E., and Hafler, D.A. (2005). Functionalanalysis of highly defined, FACS-isolated populations of humanregulatory CD4+ CD25+ T cells. Clin. Immunol. 115, 10–18.
Barrat, F.J., Cua, D.J., Boonstra, A., Richards, D.F., Crain, C., Savel-koul, H.F., de Waal-Malefyt, R., Coffman, R.L., Hawrylowicz, C.M.,and O’Garra, A. (2002). In vitro generation of interleukin 10-produc-ing regulatory CD4(+) T cells is induced by immunosuppressivedrugs and inhibited by T helper type 1 (Th1)- and Th2-inducingcytokines. J. Exp. Med. 195, 603–616.
Bejon, P., Andrews, L., Andersen, R.F., Dunachie, S., Webster, D.,Walther, M., Gilbert, S.C., Peto, T., and Hill, A.V.S. (2005). Calcula-tion of liver-to-blood inocula, parasite growth rates and pre-eryth-rocytic vaccine efficacy, from serial quantitative PCR studies of vol-unteers challenged with malaria sporozoite. J. Infect. Dis. 191,619–626.
Belkaid, Y., Piccirillo, C.A., Mendez, S., Shevach, E.M., and Sacks,D.L. (2002). CD4+CD25+ regulatory T cells control Leishmania majorpersistence and immunity. Nature 420, 502–507.
Brunkow, M.E., Jeffery, E.W., Hjerrild, K.A., Paeper, B., Clark, L.B.,Yasayko, S.A., Wilkinson, J.E., Galas, D., Ziegler, S.F., and Rams-dell, F. (2001). Disruption of a new forkhead/winged-helix protein,scurfin, results in the fatal lymphoproliferative disorder of the scurfymouse. Nat. Genet. 27, 68–73.
Cabrera, R., Tu, Z., Xu, Y., Firpi, R.J., Rosen, H.R., Liu, C., and Nel-son, D.R. (2004). An immunomodulatory role for CD4(+)CD25(+)regulatory T lymphocytes in hepatitis C virus infection. Hepatology40, 1062–1071.
Chaiyaroj, S.C., Rutta, A.S., Muenthaisong, K., Watkins, P., Na Ubol,M., and Looareesuwan, S. (2004). Reduced levels of transforminggrowth factor-beta1, interleukin-12 and increased migration inhibi-tory factor are associated with severe malaria. Acta Trop. 89, 319–327.
Chen, Y., Kuchroo, V.K., Inobe, J., Hafler, D.A., and Weiner, H.L.(1994). Regulatory T cell clones induced by oral tolerance: suppres-sion of autoimmune encephalomyelitis. Science 265, 1237–1240.
Chen, W., Jin, W., Hardegen, N., Lei, K.J., Li, L., Marinos, N.,McGrady, G., and Wahl, S.M. (2003). Conversion of peripheralCD4+CD25− naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. J. Exp. Med. 198,1875–1886.
de Souza, J.B., and Riley, E.M. (2002). Cerebral malaria: the contri-bution of studies in animal models to our understanding of immu-nopathogenesis. Microbes Infect. 4, 291–300.
Dieckmann, D., Plottner, H., Berchtold, S., Berger, T., and Schuler,G. (2001). Ex vivo isolation and characterization of CD4(+)CD25(+)T cells with regulatory properties from human blood. J. Exp. Med.193, 1303–1310.
Dieckmann, D., Bruett, C.H., Ploettner, H., Lutz, M.B., and Schuler,G. (2002). Human CD4(+)CD25(+) regulatory, contact-dependent Tcells induce interleukin 10-producing, contact-independent type1-like regulatory T cells. J. Exp. Med. 196, 247–253.
Dodoo, D., Omer, F.M., Todd, J., Akanmori, B.D., Koram, K.A., andRiley, E.M. (2002). Absolute levels and ratios of proinflammatoryand anti-inflammatory cytokine production in vitro predict clinicalimmunity to Plasmodium falciparum malaria. J. Infect. Dis. 185,971–979.
Fahlen, L., Read, S., Gorelik, L., Hurst, S.D., Coffman, R.L., Flavell,R.A., and Powrie, F. (2005). T cells that cannot respond to TGF-beta escape control by CD4(+)CD25(+) regulatory T cells. J. Exp.Med. 201, 737–746.
Fantini, M.C., Becker, C., Monteleone, G., Pallone, F., Galle, P.R.,and Neurath, M.F. (2004). Cutting edge: TGF-beta induces a regula-tory phenotype in CD4+CD25− T cells through Foxp3 induction anddown-regulation of Smad7. J. Immunol. 172, 5149–5153.
Fontenot, J.D., Gavin, M.A., and Rudensky, A.Y. (2003). Foxp3 pro-grams the development and function of CD4+CD25+ regulatory Tcells. Nat. Immunol. 4, 330–336.
Groux, H., O’Garra, A., Bigler, M., Rouleau, M., Antonenko, S., deVries, J.E., and Roncarolo, M.G. (1997). A CD4+ T-cell subset inhib-its antigen-specific T-cell responses and prevents colitis. Nature389, 737–742.
Hisaeda, H., Maekawa, Y., Iwakawa, D., Okada, H., Himeno, K., Ki-shihara, K., Tsukumo, S., and Yasutomo, K. (2004). Escape of ma-laria parasites from host immunity requires CD4+ CD25+ regulatoryT cells. Nat. Med. 10, 29–30.
Hori, S., Nomura, T., and Sakaguchi, S. (2003). Control of regulatoryT cell development by the transcription factor Foxp3. Science 299,1057–1061.
Jonuleit, H., Schmitt, E., Stassen, M., Tuettenberg, A., Knop, J., andEnk, A.H. (2001). Identification and functional characterization ofhuman CD4(+)CD25(+) T cells with regulatory properties isolatedfrom peripheral blood. J. Exp. Med. 193, 1285–1294.
Jonuleit, H., Schmitt, E., Kakirman, H., Stassen, M., Knop, J., andEnk, A.H. (2002). Infectious tolerance: human CD25(+) regulatory Tcells convey suppressor activity to conventional CD4(+) T helpercells. J. Exp. Med. 196, 255–260.
Khattri, R., Cox, T., Yasayko, S.A., and Ramsdell, F. (2003). Anessential role for Scurfin in CD4+CD25+ T regulatory cells. Nat. Im-munol. 4, 337–342.
Kinter, A.L., Hennessey, M., Bell, A., Kern, S., Lin, Y., Daucher, M.,Planta, M., McGlaughlin, M., Jackson, R., Ziegler, S.F., and Fauci,A.S. (2004). CD25(+)CD4(+) regulatory T cells from the peripheralblood of asymptomatic HIV-infected individuals regulate CD4(+)and CD8(+) HIV-specific T cell immune responses in vitro and areassociated with favorable clinical markers of disease status. J. Exp.Med. 200, 331–343.
Laouar, Y., Sutterwala, F.S., Gorelik, L., and Flavell, R.A. (2005).Transforming growth factor-beta controls T helper type 1 cell devel-opment through regulation of natural killer cell interferon-gamma.Nat. Immunol. 6, 600–607.
Levings, M.K., Sangregorio, R., and Roncarolo, M.G. (2001). Humancd25(+)cd4(+) t regulatory cells suppress naive and memory T cellproliferation and can be expanded in vitro without loss of function.J. Exp. Med. 193, 1295–1302.
Li, C., Sanni, L.A., Omer, F., Riley, E., and Langhorne, J. (2003).Pathology of Plasmodium chabaudi chabaudi infection and mortal-ity in interleukin-10-deficient mice are ameliorated by anti-tumornecrosis factor alpha and exacerbated by anti-transforming growthfactor beta antibodies. Infect. Immun. 71, 4850–4856.
Lundgren, A., Suri-Payer, E., Enarsson, K., Svennerholm, A.M., andLundin, B.S. (2003). Helicobacter pylori-specific CD4+ CD25high
regulatory T cells suppress memory T-cell responses to H. pylori ininfected individuals. Infect. Immun. 71, 1755–1762.
Immunity296
Marie, J.C., Letterio, J.J., Gavin, M., and Rudensky, A.Y. (2005).TGF-beta1 maintains suppressor function and Foxp3 expression inCD4+CD25+ regulatory T cells. J. Exp. Med. 201, 1061–1067.
McConkey, S.J., Reece, W.H., Moorthy, V.S., Webster, D., Dunachie,S., Butcher, G., Vuola, J.M., Blanchard, T.J., Gothard, P., Watkins,K., et al. (2003). Enhanced T-cell immunogenicity of plasmid DNAvaccines boosted by recombinant modified vaccinia virus Ankarain humans. Nat. Med. 9, 729–735.
Mendez, S., Reckling, S.K., Piccirillo, C.A., Sacks, D., and Belkaid,Y. (2004). Role for CD4(+) CD25(+) regulatory T cells in reactivationof persistent leishmaniasis and control of concomitant immunity. J.Exp. Med. 200, 201–210.
Omer, F.M., and Riley, E.M. (1998). Transforming growth factor betaproduction is inversely correlated with severity of murine malariainfection. J. Exp. Med. 188, 39–48.
Omer, F.M., de Souza, J.B., Corran, P.H., Sultan, A.A., and Riley,E.M. (2003a). Activation of transforming growth factor beta by ma-laria parasite-derived metalloproteinases and a thrombospondin-like molecule. J. Exp. Med. 198, 1817–1827.
Omer, F.M., de Souza, J.B., and Riley, E.M. (2003b). Differential in-duction of TGF-beta regulates proinflammatory cytokine produc-tion and determines the outcome of lethal and nonlethal Plasmo-dium yoelii infections. J. Immunol. 171, 5430–5436.
Park, H.B., Paik, D.J., Jang, E., Hong, S., and Youn, J. (2004).Acquisition of anergic and suppressive activities in transforminggrowth factor-beta-costimulated CD4+CD25− T cells. Int. Immunol.16, 1203–1213.
Perkins, D.J., Weinberg, J.B., and Kremsner, P.G. (2000). Reducedinterleukin-12 and transforming growth factor-beta1 in severechildhood malaria: relationship of cytokine balance with diseaseseverity. J. Infect. Dis. 182, 988–992.
Perry, J.A., Olver, C.S., Burnett, R.C., and Avery, A.C. (2005). Cuttingedge: the acquisition of TLR tolerance during malaria infection im-pacts T cell activation. J. Immunol. 174, 5921–5925.
Powrie, F., Read, S., Mottet, C., Uhlig, H., and Maloy, K. (2003).Control of immune pathology by regulatory T cells. Novartis Found.Symp. 252, 92–98.
Rhee, M.S., Akanmori, B.D., Waterfall, M., and Riley, E.M. (2001).Changes in cytokine production associated with acquired immunityto Plasmodium falciparum malaria. Clin. Exp. Immunol. 126, 503–510.
Rouse, B.T., and Suvas, S. (2004). Regulatory cells and infectiousagents: detentes cordiale and contraire. J. Immunol. 173, 2211–2215.
Sakaguchi, S., Sakaguchi, N., Asano, M., Itoh, M., and Toda, M.(1995). Immunologic self-tolerance maintained by activated T cellsexpressing IL-2 receptor alpha-chains (CD25). Breakdown of a sin-gle mechanism of self-tolerance causes various autoimmune dis-eases. J. Immunol. 155, 1151–1164.
Satoguina, J., Mempel, M., Larbi, J., Badusche, M., Loliger, C.,Adjei, O., Gachelin, G., Fleischer, B., and Hoerauf, A. (2002). Anti-gen-specific T regulatory-1 cells are associated with immunosup-pression in a chronic helminth infection (onchocerciasis). MicrobesInfect. 4, 1291–1300.
Schramm, C., Huber, S., Protschka, M., Czochra, P., Burg, J.,Schmitt, E., Lohse, A.W., Galle, P.R., and Blessing, M. (2004). TGF-beta regulates the CD4+CD25+ T-cell pool and the expression ofFoxp3 in vivo. Int. Immunol. 16, 1241–1249.
Schubert, L.A., Jeffery, E., Zhang, Y., Ramsdell, F., and Ziegler, S.F.(2001). Scurfin (FOXP3) acts as a repressor of transcription andregulates T cell activation. J. Biol. Chem. 276, 37672–37679.
Stephens, L.A., Mottet, C., Mason, D., and Powrie, F. (2001). HumanCD4(+)CD25(+) thymocytes and peripheral T cells have immunesuppressive activity in vitro. Eur. J. Immunol. 31, 1247–1254.
Sugimoto, K., Ikeda, F., Stadanlick, J., Nunes, F.A., Alter, H.J., andChang, K.M. (2003). Suppression of HCV-specific T cells withoutdifferential hierarchy demonstrated ex vivo in persistent HCV infec-tion. Hepatology 38, 1437–1448.
Suvas, S., Azkur, A.K., Kim, B.S., Kumaraguru, U., and Rouse, B.T.
(i
TAcJ
Tfs
TbP
UAiN
vafpA
WGt1
YSSh
Z(ca
2004). CD4+CD25+ regulatory T cells control the severity of viralmmunoinflammatory lesions. J. Immunol. 172, 4123–4132.
aams, L.S., Smith, J., Rustin, M.H., Salmon, M., Poulter, L.W., andkbar, A.N. (2001). Human anergic/suppressive CD4(+)CD25(+) Tells: a highly differentiated and apoptosis-prone population. Eur.. Immunol. 31, 1122–1131.
hornton, A.M., and Shevach, E.M. (2000). Suppressor effectorunction of CD4+CD25+ immunoregulatory T cells is antigen non-pecific. J. Immunol. 164, 183–190.
sutsui, N., and Kamiyama, T. (1999). Transforming growth factoreta-induced failure of resistance to infection with blood-stagelasmodium chabaudi in mice. Infect. Immun. 67, 2306–2311.
rban, B.C., Ferguson, D.J., Pain, A., Willcox, N., Plebanski, M.,ustyn, J.M., and Roberts, D.J. (1999). Plasmodium falciparum-
nfected erythrocytes modulate the maturation of dendritic cells.ature 400, 73–77.
an Waarde, M.A., van Assen, A.J., Kampinga, H.H., Konings, A.W.,nd Vujaskovic, Z. (1997). Quantification of transforming growthactor-beta in biological material using cells transfected with alasminogen activator inhibitor-1 promoter-luciferase construct.nal. Biochem. 247, 45–51.
enisch, C., Parschalk, B., Burgmann, H., Looareesuwan, S., andraninger, W. (1995). Decreased serum levels of TGF-beta in pa-
ients with acute Plasmodium falciparum malaria. J. Clin. Immunol.5, 69–73.
agi, H., Nomura, T., Nakamura, K., Yamazaki, S., Kitawaki, T., Hori,., Maeda, M., Onodera, M., Uchiyama, T., Fujii, S., and Sakaguchi,. (2004). Crucial role of FOXP3 in the development and function ofuman CD25+CD4+ regulatory T cells. Int. Immunol. 16, 1643–1656.
heng, S.G., Wang, J.H., Gray, J.D., Soucier, H., and Horwitz, D.A.2004). Natural and induced CD4+CD25+ cells educate CD4+CD25−
ells to develop suppressive activity: the role of IL-2, TGF-beta,nd IL-10. J. Immunol. 172, 5213–5221.
