of March 26, 2018.This information is current as

MalariaCerebralCells in the Control of Experimental

Regulatory T+Foxp3+Limited Role of CD4

Fleischer and Thomas JacobsChristiane Steeg, Guido Adler, Tim Sparwasser, Bernhard

http://www.jimmunol.org/content/183/11/7014doi: 10.4049/jimmunol.0901422November 2009;

2009; 183:7014-7022; Prepublished online 4J Immunol 

Referenceshttp://www.jimmunol.org/content/183/11/7014.full#ref-list-1

, 19 of which you can access for free at: cites 38 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2009 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Limited Role of CD4�Foxp3� Regulatory T Cells in theControl of Experimental Cerebral Malaria1

Christiane Steeg,* Guido Adler,* Tim Sparwasser,† Bernhard Fleischer,* and Thomas Jacobs2*

Cerebral malaria (CM) associated with Plasmodium berghei ANKA (PbA) infection is an accepted model of human CM. CMduring PbA infection critically depends on sequestration of T cells into the brain. Several studies aimed to address the role ofregulatory T cells (Treg) in modulating this pathogenic T cell response. However, these studies are principally hampered due tothe fact that until recently no reagents were available to deplete Foxp3� Treg specifically. To study the function of Treg in thegenesis of CM, we used depletion of Treg mice that are transgenic for a bacterial artificial chromosome expressing a diphtheriatoxin receptor-enhanced GFP fusion protein under the control of the foxp3 gene locus. These mice allow for a selective depletionof Foxp3� Treg by diphtheria toxin injection, and also their specific detection and purification during an ongoing infection. Usingdepletion of Treg mice, we found only a small increase in the absolute numbers of Foxp3� Treg during PbA infection and,consequently, the ratio of Treg to T effector cells (Teff) decreased due to the rapid expansion of Teff. Although the latter sequesterin the brains of infected mice, almost no Treg were found in the brains of infected mice. Furthermore, we demonstrate thatdepletion of Treg has no influence on sequestration of Teff and on the clinical outcome, and only minor influence on T cell activation.Using ex vivo analysis of purified Treg from either naive mice or PbA-infected mice, we found that both exhibit similar inhibitorycapacity on Teff. The Journal of Immunology, 2009, 183: 7014–7022.

C erebral malaria (CM)3 is a life-threatening sequela of hu-mans infected with Plasmodium falciparum (1). Infectionof susceptible mouse strains with Plasmodium berghei

ANKA (PbA) is an experimental model of CM that shares severalcharacteristics with the human disease (2, 3). During PbA infec-tion, parasitized erythrocytes, but also leukocytes, are sequesteredin the brain (4–6). Various studies clearly demonstrated a role forT cells secreting proinflammatory cytokines for the genesis of CM(7–10). We recently found that T cell activation during malaria isaccompanied by an increased expression of the negative costimu-lators CTLA-4 (11, 12) and B and T lymphocyte attenuator(BTLA) (13). Both of these were shown to modulate T cell func-tion during PbA infection. In PbA-infected C57BL/6 mice, wefound that blockade of CTLA-4 leads to exacerbation of the im-mune pathology triggered by T cells producing proinflammatorycytokines, whereas an agonistic Ab against BTLA prevented thegenesis of CM (13). This supports the idea that the induction ofnegative costimulators during PbA infection is a means of damp-ening the strong and polarized activation of the Th1 arm of theimmune system to prevent immune pathology. To further analyze

the role of T cell regulation, several studies have addressed the roleof CD4�CD25�Foxp3� natural regulatory T cells (Treg) duringthe pathogenesis of CM. However, until recently, no tools wereavailable to deplete Treg specifically. In naive mice, CD4�Foxp3�

T cells express CD25. Thus, Abs against CD25 have been used todeplete CD4�CD25� Treg (14, 15). Although this approach hasprovided novel information on the function of Treg, it is problem-atic because depletion is not complete in several organs (16) andmay also interfere with activated T effector cells (Teff) expressingCD25, as recently shown during chronic Toxoplasma gondii in-fection (17). To unequivocally study the function of Treg during thegenesis of CM, we used depletion of Treg (DEREG) mice (18).These mice are transgenic for a bacterial artificial chromosomeexpressing a diphtheria toxin (DT) receptor-enhanced GFP (eGFP)fusion protein under the control of the foxp3 gene locus. Thismethod allows for a selective depletion of Foxp3� Treg by DTinjection, and also their specific detection, quantification, and pu-rification during an ongoing infection.

Using DEREG mice, we show that activated CD8� and CD4�

T cells sequester into brains of infected mice, whereas Foxp3�

Treg are virtually excluded. Furthermore, although Treg that werepurified from infected mice exhibit regulatory activity in vitro,depletion of Treg does not alter the severity and incidence of CM.Thus, our data suggest that CM in the PbA model is triggered bya rapid activation and proliferation of potentially pathogenic Tcells that cannot be controlled by Treg.

Materials and MethodsMice and parasites

Male DEREG mice (18) were crossed with female C57BL/6 mice in theanimal facility of the Bernhard Nocht Institute for Tropical Medicine. Theresulting offspring was typed by PCR and flow cytometry. In each exper-iment, hemizygotic DEREG mice were used with sex- and age-matchedwild-type littermates as controls.

PbA was maintained by alternating cyclic passage of the parasites inAnopheles stephensi mosquitoes and BALB/c mice at the mosquito colonyof the Bernhard Nocht Institute for Tropical Medicine. Blood was collectedfrom highly parasitemic mice, and aliquots were stored in liquid nitrogen

*Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany; and †Instituteof Infection Immunology, Centre for Experimental and Clinical Infection Research,Twincore, Hanover, Germany

Received for publication May 7, 2009. Accepted for publication September 29, 2009.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by a grant of the Deutsche Forschungsgemeinschaft to T.J.(JA 1451).2 Address correspondence and reprint requests to Dr. Thomas Jacobs, Department ofImmunology, Bernhard Nocht Institute for Tropical Medicine, Bernhard-Nocht-Str.74, 20359 Hamburg, Germany. E-mail address: tjacobs@bni-hamburg.de3 Abbreviations used in this paper: CM, cerebral malaria; BTLA, B and T lymphocyteattenuator; DEREG, depletion of regulatory T cell; DT, diphtheria toxin; eGFP, en-hanced GFP; PbA, Plasmodium berghei ANKA; p.i., postinfection; Teff, T effectorcell; Treg, regulatory T cell.

Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00

The Journal of Immunology

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901422

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

in a solution of 0.9% NaCl, 4.6% sorbitol, and 35% glycerol. DEREG andC57BL/6 (5–8 wk old) were infected i.p. with 2 � 106 PbA-infectedRBCs. Parasitemia was determined in Wright-stained blood smears fromtail blood. C57BL/6 mice infected with blood stages of PbA have beenshown to develop neurological behavioral changes, such as ataxia, convul-sions, and coma, and usually die between days 6 and 8 postinfection (p.i.).Similarly, in our study, 90–100% of PbA-infected mice exhibited signs ofcerebral involvement between days 6 and 8 p.i., characterized by weightloss, reduced locomotion, and marked ataxia. Mice developing cerebralmalaria were scored according to the severity of the symptoms and sub-sequently euthanized to avoid unnecessary suffering between days 7 and9 p.i. All experiments were in accordance with local Animal Ethics Com-mittee regulations.

Abs, reagents, and flow cytometry

Anti-CD3 mAb was purified from the hybridoma cell line 145-2C11. Forflow cytometry, the following FITC-, PE-, PerCP-Cy5.5-, PerCP-, or al-lophycocyanin-labeled Abs and appropriate isotype controls were used:anti-CD4 (RM4-5), anti-CD8a (53-6.7), anti-CD45 (30-F11), and anti-CD62L (MEL-14) from BD Pharmingen. Anti-CD25 (PC61.5.3) was pur-chased from Cedarlane Laboratories; polyclonal anti-eGFP from NovusBiologicals; and anti-Foxp3 (FJK-16s) from eBioscience. Data were ac-quired on a FACSCalibur flow cytometer (BD Biosciences) and analyzedwith the CellQuest program (BD Biosciences).

Staining of blood and spleen lymphocytes

A total of 30 �l of blood from the tail vein was collected in a heparinizedcapillary tube. RBC were lysed with 0.17 M ammonium chloride, and afterblocking with anti-Fc�R, cells were counted and stained with anti-CD4,anti-CD8, anti-CD62L, or anti-CD25. Spleen cells were homogenized andpassed through a cell strainer. After lysis of RBC, 2 � 106 cells werestained with the indicated Abs. Intracellular staining for Foxp3 was con-ducted according to the manufacturer’s instruction (eBioscience).

Staining of brain lymphocytes

C57BL/6 or DEREG mice were infected with 2 � 106 PbA-infected RBC.On day 6 p.i., mice were sacrificed and brains were removed. For quanti-fication of cell numbers, brains were weighed, homogenized, passedthrough a cell strainer, and stained with anti-CD45, anti-CD4 or anti-CD8,and anti-CD62L or anti-CD25. Cells prepared from one-tenth of the brainwere collected on a FACSCalibur, and the number of CD45� cells wasused to calculate the total number of sequestered cells. In native samples,detection of Treg was conducted by analysis of eGFP expression. In someexperiments, Treg were analyzed after fixation and permeabilization by in-tracellular staining of Foxp3 and eGFP using anti-Foxp3-PE and a poly-clonal anti-eGFP-FITC Ab, according to the manufacturer’s instruction.

Inhibition assay

Spleen cells from uninfected and PbA-infected mice at day 6 p.i. weresorted on a FACSAria (BD Biosciences) into eGFP� Treg and eGFP� Teff.Cells were mixed in different ratios and stimulated with 1 �g/ml anti-CD3Ab for 48 h. IL-2 in the supernatants was analyzed in an ELISA (R&DSystems), and proliferation was measured by analyzing the incorporationof [3H]thymidine after incubation of cells with 0.2 �Ci/well for 16 h.

Statistical analysis

Statistical analyses were performed with unpaired Student’s t test. One-way ANOVA was used to compare parasitemia between different groups ofmice. The log rank test was used to test the equality of survival functionsacross different groups of mice. All statistical analyses were performedwith Prism software (GraphPad).

ResultsQuantification of Foxp3� Treg during blood-stage infection withPbA

The blood stage of a PbA infection induces a strong activation ofCD8� T cells as well as CD4� cells in spleen and blood, leadingto an increased number of activated T cells, as indicated by the lossof CD62L expression on the surface (Fig. 1). To verify whether theexpansion of Teff is also accompanied by an increasing number ofFoxp3� Treg, we made use of the recently described DEREG micethat allow for the detection of Treg by expression of eGFP and alsotheir selective depletion by application of DT (18). Analysis of the

absolute number of both Teff and Treg revealed only marginalchanges in blood and spleen at day 7 p.i., which were of borderlinesignificance or not statistically significant (Fig. 2, A and B). How-ever, by comparing the percentage of GFP� Treg in the whole

FIGURE 1. The blood stage of a PbA infection leads to a strong acti-vation of T cells. C57BL/6 mice were infected with 2 � 106 PbA-infectedRBCs at day 0. Lymphocytes from spleen or blood were obtained fromuninfected control mice or from mice at day 7 p.i. Cells were stained withallophycocyanin-labeled anti-CD4 or allophycocyanin-labeled anti-CD8,and with PE-labeled anti-CD62L. Samples were subsequently analyzed byflow cytometry. A summary of data (n � 5) expressed as mean � SD isdepicted in the lower panel.

7015The Journal of Immunology

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

CD4� T cell population in blood and spleen, we found an overalldecrease in the proportion of Treg at day 7 p.i. (Fig. 2C).

PbA infection was also accompanied by an expansion ofCD4�CD25� T cells that are often considered to represent Treg innaive mice. To further characterize these cells, we stained CD25on T cells in the spleen from uninfected control mice or PbA-infected mice at day 7 p.i. and analyzed the number ofCD25�Foxp3� and CD25�Foxp3� T cells (Fig. 3A). Whereas innaive mice almost all CD4�CD25� T cells were Foxp3� Treg, ininfected mice roughly 50% of CD4�CD25� T cells were Foxp3�

(Fig. 3B). Similar results were obtained using analysis of eGFPexpression (data not shown). This demonstrates that CD25 expres-sion as a marker is not sufficient to define the population of Treg

during PbA infection, and consequently, anti-CD25 applicationcannot be regarded to specifically deplete only Treg during PbAinfection. CTLA-4 is expressed on Foxp3� Treg constitutively. Wefound recently that CTLA-4 is induced on a very high number ofCD4� T cells during experimental malaria. In light of our presentresults, in which no increase in the numbers of Treg was found, we

performed an intracellular staining for Foxp3 and CTLA-4 (Fig.3C). In this experiment, we again found no significant increase inthe number of Foxp3� Treg at day 7 p.i. Furthermore, whereas innaive mice almost all CTLA-4� T cells express Foxp3, a strongincrease of CTLA-4�CD4�Foxp3� T cells was found upon blood-stage infection.

Analysis of T cell sequestration in the brain during developmentof CM

Infection of C57BL/6 mice with PbA causes CM in the majority ofinfected mice. Mice suffering from CM usually die between days7 and 9 p.i., showing severe neurological symptoms. A hallmark ofCM in this model is the sequestration of large numbers of CD8�

T cells and, albeit at lower numbers, also of CD4� T cells into thebrain (Fig. 4, A, D, and E). These cells produce large amounts ofIFN-�, which is implicated in the genesis of CM (data not shown).We asked whether Treg also sequester in the brain, which would bethe prerequisite for a local modulation of CM on inflamed endo-thelium in the brain of infected mice. Interestingly, hardly anyCD4� eGFP� Treg or CD8� eGFP� Treg can be found in thefraction of T cells that were recovered from the brain (Fig. 4, B andC). The vast majority of sequestered cells are activated CD62Llow

CD8� or CD4� T cells that were eGFP� (Fig. 4, B and C) andFoxp3� (data not shown). Thus, the ratio between Treg and Teff inbrain during PbA infection decreased compared with control mice,probably due to a rapid infiltration of Teff. These Teff can be re-covered from brains of PbA-infected mice, whereas in this studyalmost no Treg can be detected.

Effects of Treg depletion on the activation of T cells and theincidence of CM during PbA infection

To further study the role of Treg during the course of malaria,C57BL/6 wt and DEREG mice were infected with PbA and wereinjected with DT on days �1, �2, and �3. This protocol leads toa depletion of eGFP� Treg in DEREG mice during the course ofinfection (Fig. 5A, upper panel). The absence of Treg in DEREGmice was also verified by intracellular staining of Foxp3 (data notshown). Interestingly, we found an increasing number ofCD4�CD25� T cells during infection in Treg-depleted DEREGmice (Fig. 5A, lower panel, and B). This further confirmed ourobservation that CD25 is not only expressed on Treg, but also onactivated Teff during PbA infection. Depletion of Treg does not alterthe parasitemia of mice, and mice also showed a similar weightloss during the course of infection (Fig. 6, A–C). In addition, weobserved no general differences in the incidence of CM. Also,when we used a more sensitive score to monitor early neurologicalsymptoms occurring before the onset of CM (Fig. 6D), we ob-served no statistically significant increases in the clinical score inTreg-depleted DEREG mice between days 7 and 8 just before onsetof CM. However, it is noteworthy that some Treg-depleted miceexhibit an increased clinical score. Depletion of Treg in DEREGmice resulted in an enhanced activation of CD4� as well as CD8�

T cells upon PbA infection in spleen (Fig. 6, E and F) and blood(data not shown) when compared with DT-treated C57BL/6 wtmice. We also used a modified protocol for Treg depletion, inwhich DEREG mice were treated on days �3, �2, and �1 withDT. Similarly to our original protocol, no effect of Treg depletionon weight loss and incidence of CM was found (Fig. 6, G and H).

Although the depletion of Treg was accompanied by an en-hanced activation of T cells in the spleen (Fig. 6, E and F), thepercentage and the absolute numbers of T cells recovered frombrains of Treg-depleted and nondepleted mice were similar, whichfurther supports the view that Treg are not capable of controlling

FIGURE 2. Quantification of Treg during PbA infection. DEREG micewere infected with 2 � 106 PbA-infected RBCs at day 0. Lymphocytesfrom spleen or blood were obtained from uninfected control mice or frommice at day 7 p.i., stained with allophycocyanin-labeled anti-CD4, andanalyzed for eGFP expression. Absolute numbers of cells correspond to 1�l of blood (A) or one spleen (B). C, The ratio of GFP� Treg of Teff in blood(left) and spleen during PbA infection is depicted. Data are expressed aspercentage of CD4�GFP� Treg per CD4� T cells. Data are expressed asmean � SD of n � 5 mice.

7016 ROLE OF Treg DURING CM

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

the pathological T cell response in vivo and, hence, cannot preventCM (Fig. 7).

Ex vivo analysis of Treg function during the blood stage ofmalaria

Numerous studies have been undertaken to characterize the inhib-itory activity of CD4�CD25� Treg from naive mice. Because thesemarkers are not exclusively defining Treg during T cell activation,this strategy cannot be applied to analyze Treg function during anongoing inflammation or infection. Based on our results, we hy-pothesize that during the course of the PbA infection, Treg functionis either temporarily or irreversibly impaired. Thus, we askedwhether PbA infection alters the function of Treg. To this end, wesorted eGFP� Treg from spleens of uninfected and infectedDEREG mice. This method allows for the identification and pu-rification of eGFP� Treg without staining for additional markers,which might interfere with their function. Treg that were sortedfrom naive and infected DEREG mice were highly pure (�95%were eGFP�Foxp3�; data not shown). To compare the regulatory

function of Treg from naive mice with Treg from PbA-infectedmice, these cells were sorted and added to eGFP� T cells fromuninfected control mice that were stimulated with anti-CD3 (Fig.8). Interestingly, eGFP� Treg from naive DEREG mice and eGFP�

Treg from infected DEREG mice were both capable of inhibitingIL-2 production and proliferation. These data suggest that eitherTreg function is not impaired during the blood stage of PbA infec-tion or these cells regain their regulatory capacity very fast uponcultivation in vitro. In addition, purified GFP� Teff from PbA-infected mice were also susceptible to inhibition by purified GFP�

Treg. It is noteworthy that Teff from PbA-infected mice produceless IL-2 (and other cytokines) upon stimulation.

DiscussionCM during PbA infection critically depends on T cells producingproinflammatory cytokines. CD8� T cells and, to a lesser extent,also CD4� T cells sequester in the brain and produce proinflam-matory cytokines (13, 19, 20). These cells can be found in direct

FIGURE 3. Expansion of CD4�

CD25�Foxp3� T cells during PbAinfection. C57BL/6 mice were in-fected with 2 � 106 PbA-infectedRBCs at day 0. At day 7 p.i., spleencells were stained with PerCP-labeledanti-CD4 and PE-labeled anti-CD25,and stained intracellularly for Foxp3using allophycocyanin-labeled anti-Foxp3. In naive control mice, mostFoxp3� T cells were CD25�,whereas during infection CD25� isinduced on Foxp3�CD4� T cells (A).B, A summary of two experimentswith n � 4 mice each is shown. Dataare expressed as mean � SD. C,DEREG mice were infected, as de-scribed. At day 7 p.i., cells werestained intracellularly using PE-la-beled anti-CTLA-4 and FITC-labeledanti-Foxp3 (middle). As control,spleen cells from uninfected micewere stained (left). PE-labeled anti-hamster served as istotype control(right). This experiment was repeatedtwice with similar results. An identi-cal experiment using C57BL/6 miceshowed similar results (data notshown).

7017The Journal of Immunology

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

proximity to the brain endothelium. Several lines of evidence sug-gest that this process directly contributes to CM. Thus, immuneregulatory pathways that control either the magnitude and/or thecytokine bias of this pathogenic T cell response are of particularinterest. In general, such regulation can either take place: 1) inlymphoid organs during priming by influencing T cell activation,or 2) by the modulation of T cells that are already retained in thebrain on activated endothelium. In the present study, we sought todetermine whether Treg contribute to the control of pathogenic T

cell responses during CM by either one of these mechanisms. Treg

can be separated into natural Treg that are already present in naivemice and inducible Treg that arise during inflammation or infection(21), needing TGF-ß for their induction, a cytokine that was al-ready shown to influence pathology during malaria (22, 23). Bothof these cell types express Foxp3; however, expression is onlytransient in the latter. Using the recently described DEREG mice,it is possible to specifically deplete both subtypes of Foxp3� Treg

by DT treatment during the whole course of PbA infection. In

FIGURE 4. Infiltration of T cells into the brain during PbA infection. DEREG mice were infected with 2 � 106 PbA-infected RBCs. On day 6 p.i., micewere sacrificed and cells from brains were isolated. Samples were stained with PerCP-labeled anti-CD45, allophycocyanin-labeled anti-CD4 or allophy-cocyanin-labeled anti-CD8, and PE-labeled anti-CD62L. Foxp3� Treg were quantified by eGFP expression. Representative dot plots are shown for cells inthe brain of uninfected control mice and PbA-infected mice. One-tenth of the brain was analyzed and gated on CD45� cells (A). Flow cytometric analysisshows an increased number of cells that were CD4�/CD62Llow (D) or CD8�/CD62Llow (E) in PbA-infected mice compared with uninfected controls.Although an increased number of T cells was found in PbA-infected mice compared with control mice, almost none of these cells were CD4� eGFP� (B)or CD8� eGFP� T cells (C). Experiments were performed at least three times with n � 5 mice per group. A summary of this data expressed as mean �SD is depicted in the lower panel.

7018 ROLE OF Treg DURING CM

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

addition, this technique does not only allow for the quantificationof Treg in various organs, but also their purification by cell sortingdue to their expression of eGFP. This enabled us to perform func-tional comparisons of Treg from naive and PbA-infected mice.

PbA infection is accompanied by an expansion of T cells thatexpress activation markers such as CD69 and are CD62Llow. Al-though the absolute numbers of eGFP�Foxp3�CD4� Treg inDEREG mice remained constant upon infection, due to a morerapid expansion of Teff, the Treg:Teff ratio declines. This decreasedratio of Treg might contribute to a strong antiparasitic T cell re-sponse with the eventual cost of pathology. By analyzing the cel-lular infiltrate of brains of PbA-infected DEREG mice, we foundthat Treg do not accumulate in the brain before or during the onsetof CM, whereas activated conventional T cells massively accumu-late in the brain during this time window. This supports recentfindings using intracellular staining of Foxp3 as a marker for Treg

in cells recovered from brains of PbA-infected mice (24), in whichno enrichment of Treg in the brain was found. Therefore, a localregulatory activity of Treg in inflamed tissue in the brain is veryunlikely. However, the principal ability of Treg to be retained in thebrain was demonstrated in the experimental autoimmune enceph-alomyelitis model (25). This indicates that the homing capacitiesof Treg vary in different disease settings, which might also influ-ence the outcome of the immune response (26).

To exclude the possibility that Treg are functionally impairedduring PbA infection, we used the eGFP expression of Treg inDEREG mice to compare the regulatory effect of purified Treg

from naive and PbA-infected mice. Functional impairment wasplausible because malaria parasites trigger TLR-2 by GPI mole-cules and TLR-2 agonists have been shown to induce a temporaryrestriction of Treg function in vitro (27). However, Treg from bothgroups display a similar capacity to suppress T cell responses invitro. This excludes a functional impairment of Treg during infec-tion with PbA. An alternative explanation might be a fast recoveryof Treg in the absence of TLR-2 agonistic molecules after purifi-cation in vitro.

Our results are in agreement with experiments in whichCD25high cells purified from PbA-infected mice that are enrichedfor Foxp3� Treg were even more active in vitro (24). Similarly toour results using the specific depletion of Treg in DEREG mice, the

depletion of Treg using anti-CD25 did not alter the incidence ofCM in this study (24). However, when we looked for subtlechanges in the early development of neurological symptoms, wefound that depletion of Treg leads to an increase in the clinicalscore. In addition, we found that Treg-depleted mice exhibit anincreased number of activated T cells in the periphery, but not inthe brain. This indicates that Treg can attenuate the activation of Tcells during PbA infection to some extent, which might dampenseveral pathological responses occurring during infection, but arenot capable of controlling the development of CM.

During the blood stage of malaria, the immune system is con-fronted with large amounts of TLR agonistic molecules. In partic-ular, hemozoin stimulates cells of the innate immune system viaTLR-9 (28), whereas GPI molecules of the parasite bind to TLR-2(29). TLR-derived signaling contributes to the polarization of theadaptive immune response toward Th1, although its contribution toCM is controversial (30, 31). Recently, we have shown that TLR-2/4/9-deficient mice still develop CM during PbA infection (32).Even the increased parasitemia of MyD88-deficient mice in thePlasmodium yoelii model is due to the disturbed IL-1/IL-18 sig-naling rather than a disturbed TLR signaling (33).

Several studies using different models revealed that TLR-2-de-rived signaling in particular may exert not only inhibitory effectson Treg, but may also activate Treg function. This might explainwhy TLR-2-deficent mice have a decreased Treg function and, as aconsequence, suffer from an enhanced pathology during schisto-soma infection (34). Interestingly, we found an increased numberof activated T cells in TLR-2/4/9-deficient mice during P. yoeliiinfection, which may also suggest a decreased Treg activity (33).Further studies are needed to dissect the consequences ofGPI-induced TLR-2 signaling on Treg function and on Th1differentiation.

The development of the pathology of CM during PbA infectionis a rapid and temporary highly coordinated process, and patho-genic CD8� T cells accumulate in the brain just before the onsetof symptoms. Thus, Treg that are often implicated in the control ofchronic inflammatory processes might not be dynamic enough tocontrol such a rapidly evolving pathology. Our data argue againsta general restriction of Treg function during malaria, but may sug-gest that Treg lack a signature of homing molecules that allow their

FIGURE 5. Depletion of Treg in DEREG mice during PbA infection. DEREG mice were infected with 2 � 106 PbA-infected RBCs at day 0. Micereceived DT injection i.p. on days �1, �2, and �3. At days 0, �3, and �6, spleen cells were isolated, stained with PE-labeled anti-CD25, and analyzedfor eGFP expression (A). A summary of data expressed as mean � SD from three independent experiments is shown in B.

7019The Journal of Immunology

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

FIGURE 6. Effects of Treg depletion in DEREG mice during PbA infection. DEREG mice were infected with 2 � 106 PbA-infected RBCs at day 0. Micereceived DT injection on days �1, �2, and � 3. Parasitemia (A), incidence of CM (B), and body weight (C) were monitored at the indicated time points.To detect more subtle changes in the neurological outcome, mice were monitored twice per day and were scored blinded (D; see Materials and Methodsfor definition). To study the effects of Treg depletion on T cell activation, similar experiments were performed, but mice were sacrificed on day 7 p.i. andspleen cells were isolated subsequently. Samples were stained with allophycocyanin-labeled anti-CD4 or allophycocyanin-labeled anti-CD8, and PE-labeledanti-CD62L or PE-labeled anti-CD25. CD4� or CD8� T cells were analyzed for CD62L expression (E) or CD25 expression (F). The experiment wasperformed three times with each n � 5 mice. Data are expressed as mean � SD. In a modified depletion protocol, DEREG mice received DT injectionon days �3, �2, and �1. Mice were then infected with 2 � 106 PbA-infected RBCs at day 0. Body weight (G) and incidence of CM (H) were monitoredat the indicated time points. This experiment was performed twice with each n � 5 mice.

7020 ROLE OF Treg DURING CM

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

homing to or retention in the inflamed brain, as observed for con-ventional T cells. Several studies have shown that during experi-mental malaria infection different regulatory mechanisms are trig-gered to avoid overwhelming pathology, such as production ofIL-10 and TGF-ß (35, 36). We have found that the negative co-stimulators CTLA-4 and BTLA expressed on T cells play a criticalrole during PbA infection by dampening the immune response(11–13). Recently, it was found that CTLA-4 is not only expressedby Foxp3� Treg constitutively, but is also critical for their function(37). In light of the present study and our previous work, in whichwe demonstrate that during experimental malaria many more cells

are CTLA-4�CD4�Foxp3� T cells than CD4�Foxp3� Treg, it istempting to speculate that counterregulatory mechanisms on acti-vated Teff such as CTLA-4 induction represent a more suitablemechanism to control a rapidly evolving pathology (11, 38). How-ever, we cannot exclude the possibility that the exacerbation ofpathology by anti-CTLA-4 treatment during PbA malaria (11, 12)is a consequence of a functional restriction of CTLA-4 on bothactivated Teff and Treg.

Interestingly, we found an increase of CD4�CD25�Foxp3� Tcells upon infection that was even higher when Treg were depletedin DEREG mice during PbA infection. In contrast, in naive mice,

FIGURE 7. Depletion of Treg is not accompanied by an increased infiltration of T cells into brain during PbA infection. DEREG mice were infected with2 � 106 PbA-infected RBCs on day 0. Mice received DT injection i.p. on days �1, �2, and �3. On day 7 p.i., mice were sacrificed and cells from brainswere isolated. Samples were stained with PerCP-labeled anti-CD45, allophycocyanin-labeled anti-CD4 or allophycocyanin-labeled anti-CD8, and PE-labeledanti-CD62L or PE-labeled anti-CD25. CD45�CD4� or CD45�CD8� T cells were analyzed for CD62L expression (A, upper panel) or CD25 expression (A, lowerpanel). A similar experiment was performed, and the total numbers of cells/100 mg brain tissue were analyzed for CD62L expression (B, upper panel) or CD25expression (B, lower panel). The experiment was performed three times with each n � 5 mice. Data are expressed as mean � SD.

FIGURE 8. Purified Treg from PbA-infected mice are functional. DEREG mice were infected with 2 � 106 PbA-infected RBCs on day 0. On day 6 p.i.,mice were sacrificed and eGFP� Treg were purified by cell sorting (d6 p.i.). As control, eGFP� Treg were sorted from uninfected control mice (not inf.).These cells were incubated with the indicated ratio of eGFP� spleen cells from infected mice or uninfected control mice and stimulated with anti-CD3.After 48 h, IL-2 production was quantified by ELISA (A) or the proliferation was measured by [3H]thymidine uptake (B). Data are mean � SD from threeexperiments.

7021The Journal of Immunology

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

virtually all CD4�CD25� T cells are Foxp3�. Currently, we donot know whether these CD4�CD25�Foxp3� T cells that are in-duced during infection are activated Teff or whether these cellsrepresent an uncanonical population of Foxp3� Treg. However,their presence indicates that previous studies addressing Treg func-tion using anti-CD25 depletion must be interpreted with caution,because this strategy will not only have an effect on Foxp3� Treg

during infection. This is in agreement with previous studies show-ing that anti-CD25 deplete Teff during chronic T. gondii infection(17). This might explain the contrasting results obtained in previ-ous studies, in which anti-CD25 treatment confers resistance toCM when applied during infection, but not when given 30 daysbefore infection (15, 24). The effects of anti-CD25 treatment dur-ing PbA infection on different CD25� T cell subsets, either beingCD25� Treg or CD25� Teff, may critically depend on the availableAb concentration, and thus, on the time point of application.

Collectively, our data demonstrate that Treg, defined by the ex-pression of Foxp3, have only a moderate effect on general pathol-ogy and attenuate T cell activation in the periphery during PbAinfection, but are not effective in the prevention of CM. In addi-tion, our study suggests that data obtained using anti-CD25 appli-cation during malaria should be interpreted with caution becausethis may not only deplete Treg, but instead will also impair other Tcell populations.

AcknowledgmentsWe thank Iris Gaworski for expert technical assistance.

DisclosuresThe authors have no financial conflict of interest.

References1. Miller, L. H., D. I. Baruch, K. Marsh, and O. K. Doumbo. 2002. The pathogenic

basis of malaria. Nature 415: 673–679.2. Hearn, J., N. Rayment, D. N. Landon, D. R. Katz, and J. B. de Souza. 2000.

Immunopathology of cerebral malaria: morphological evidence of parasite se-questration in murine brain microvasculature. Infect. Immun. 68: 5364–5376.

3. Jennings, V. M., A. A. Lal, and R. L. Hunter. 1998. Evidence for multiple patho-logic and protective mechanisms of murine cerebral malaria. Infect. Immun. 66:5972–5979.

4. Belnoue, E., M. Kayibanda, A. M. Vigario, J. C. Deschemin, N. van Rooijen,M. Viguier, G. Snounou, and L. Renia. 2002. On the pathogenic role of brain-sequestered �� CD8� T cells in experimental cerebral malaria. J. Immunol. 169:6369–6375.

5. Belnoue, E., M. Kayibanda, J. C. Deschemin, M. Viguier, M. Mack,W. A. Kuziel, and L. Renia. 2003. CCR5 deficiency decreases susceptibility toexperimental cerebral malaria. Blood 101: 4253–4259.

6. Neill, A. L., and N. H. Hunt. 1992. Pathology of fatal and resolving Plasmodiumberghei cerebral malaria in mice. Parasitology 105: 165–175.

7. Clark, I. A., S. Ilschner, J. D. MacMicking, and W. B. Cowden. 1990. TNF andPlasmodium berghei ANKA-induced cerebral malaria. Immunol. Lett. 25:195–198.

8. Jennings, V. M., J. K. Actor, A. A. Lal, and R. L. Hunter. 1997. Cytokine profilesuggesting that murine cerebral malaria is an encephalitis. Infect. Immun. 65:4883–4887.

9. De Kossodo, S., and G. E. Grau. 1993. Profiles of cytokine production in relationwith susceptibility to cerebral malaria. J. Immunol. 151: 4811–4820.

10. Bagot, S., F. Nogueira, A. Collette, V. do Rosario, F. Lemonier, P. A. Cazenave,and S. Pied. 2004. Comparative study of brain CD8� T cells induced by sporo-zoites and those induced by blood-stage Plasmodium berghei ANKA involved inthe development of cerebral malaria. Infect. Immun. 72: 2817–2826.

11. Jacobs, T., S. E. Graefe, S. Niknafs, I. Gaworski, and B. Fleischer. 2002. Murinemalaria is exacerbated by CTLA-4 blockade. J. Immunol. 169: 2323–2329.

12. Jacobs, T., T. Plate, I. Gaworski, and B. Fleischer. 2004. CTLA-4-dependentmechanisms prevent T cell induced-liver pathology during the erythrocyte stageof Plasmodium berghei malaria. Eur. J. Immunol. 34: 972–980.

13. Lepenies, B., K. Pfeffer, M. A. Hurchla, T. L. Murphy, K. M. Murphy, J. Oetzel,B. Fleischer, and T. Jacobs. 2007. Ligation of B and T lymphocyte attenuatorprevents the genesis of experimental cerebral malaria. J. Immunol. 179:4093–4100.

14. Nie, C. Q., N. J. Bernard, L. Schofield, and D. S. Hansen. 2007. CD4� CD25�

regulatory T cells suppress CD4� T-cell function and inhibit the development ofPlasmodium berghei-specific TH1 responses involved in cerebral malaria patho-genesis. Infect. Immun. 75: 2275–2282.

15. Amante, F. H., A. C. Stanley, L. M. Randall, Y. Zhou, A. Haque, K. McSweeney,A. P. Waters, C. J. Janse, M. F. Good, G. R. Hill, and C. R. Engwerda. 2007. Arole for natural regulatory T cells in the pathogenesis of experimental cerebralmalaria. Am. J. Pathol. 171: 548–559.

16. Couper, K. N., D. G. Blount, J. B. de Souza, I. Suffia, Y. Belkaid, andE. M. Riley. 2007. Incomplete depletion and rapid regeneration of Foxp3� reg-ulatory T cells following anti-CD25 treatment in malaria-infected mice. J. Im-munol. 178: 4136–4146.

17. Couper, K. N., P. A. Lanthier, G. Perona-Wright, L. W. Kummer, W. Chen,S. T. Smiley, M. Mohrs, and L. L. Johnson. 2009. Anti-CD25 antibody-mediateddepletion of effector T cell populations enhances susceptibility of mice to acutebut not chronic Toxoplasma gondii infection. J. Immunol. 182: 3985–3994.

18. Lahl, K., C. Loddenkemper, C. Drouin, J. Freyer, J. Arnason, G. Eberl,A. Hamann, H. Wagner, J. Huehn, and T. Sparwasser. 2007. Selective depletionof Foxp3� regulatory T cells induces a scurfy-like disease. J. Exp. Med. 204:57–63.

19. Boubou, M. I., A. Collette, D. Voegtle, D. Mazier, P. A. Cazenave, and S. Pied.1999. T cell response in malaria pathogenesis: selective increase in T cells car-rying the TCR V�8 during experimental cerebral malaria. Int. Immunol. 11:1553–1562.

20. Renia, L., S. M. Potter, M. Mauduit, D. S. Rosa, M. Kayibanda, J. C. Deschemin,G. Snounou, and A. C. Gruner. 2006. Pathogenic T cells in cerebral malaria. Int.J. Parasitol. 36: 547–554.

21. Mills, K. H. 2004. Regulatory T cells: friend or foe in immunity to infection? Nat.Rev. Immunol. 4: 841–855.

22. Omer, F. M., J. A. Kurtzhals, and E. M. Riley. 2000. Maintaining the immuno-logical balance in parasitic infections: a role for TGF-�? Parasitol. Today 16:18–23.

23. Omer, F. M., and E. M. Riley. 1998. Transforming growth factor � production isinversely correlated with severity of murine malaria infection. J. Exp. Med. 188:39–48.

24. Vigario, A. M., O. Gorgette, H. C. Dujardin, T. Cruz, P. A. Cazenave, A. Six,A. Bandeira, and S. Pied. 2007. Regulatory CD4� CD25� Foxp3� T cells ex-pand during experimental Plasmodium infection but do not prevent cerebral ma-laria. Int. J. Parasitol. 37: 963–973.

25. Selvaraj, R. K., and T. L. Geiger. 2008. Mitigation of experimental allergic en-cephalomyelitis by TGF-� induced Foxp3� regulatory T lymphocytes throughthe induction of anergy and infectious tolerance. J. Immunol. 180: 2830–2838.

26. Wei, S., I. Kryczek, and W. Zou. 2006. Regulatory T-cell compartmentalizationand trafficking. Blood 108: 426–431.

27. Sutmuller, R. P., M. H. den Brok, M. Kramer, E. J. Bennink, L. W. Toonen,B. J. Kullberg, L. A. Joosten, S. Akira, M. G. Netea, and G. J. Adema. 2006.Toll-like receptor 2 controls expansion and function of regulatory T cells. J. Clin.Invest. 116: 485–494.

28. Parroche, P., F. N. Lauw, N. Goutagny, E. Latz, B. G. Monks, A. Visintin,K. A. Halmen, M. Lamphier, M. Olivier, D. C. Bartholomeu, et al. 2007. Malariahemozoin is immunologically inert but radically enhances innate responses bypresenting malaria DNA to Toll-like receptor 9. Proc. Natl. Acad. Sci. USA 104:1919–1924.

29. Nebl, T., M. J. De Veer, and L. Schofield. 2005. Stimulation of innate immuneresponses by malarial glycosylphosphatidylinositol via pattern recognition recep-tors. Parasitology 130: S45–S62.

30. Coban, C., K. J. Ishii, S. Uematsu, N. Arisue, S. Sato, M. Yamamoto, T. Kawai,O. Takeuchi, H. Hisaeda, T. Horii, and S. Akira. 2007. Pathological role ofToll-like receptor signaling in cerebral malaria. Int. Immunol. 19: 67–79.

31. Togbe, D., L. Schofield, G. E. Grau, B. Schnyder, V. Boissay, S. Charron,S. Rose, B. Beutler, V. F. Quesniaux, and B. Ryffel. 2007. Murine cerebralmalaria development is independent of Toll-like receptor signaling.Am. J. Pathol. 170: 1640–1648.

32. Lepenies, B., J. P. Cramer, G. D. Burchard, H. Wagner, C. J. Kirschning, andT. Jacobs. 2008. Induction of experimental cerebral malaria is independent ofTLR2/4/9. Med. Microbiol. Immunol. 197: 39–44.

33. Cramer, J. P., B. Lepenies, F. Kamena, C. Holscher, M. A. Freudenberg,G. D. Burchard, H. Wagner, C. J. Kirschning, X. Liu, P. H. Seeberger, andT. Jacobs. 2008. MyD88/IL-18-dependent pathways rather than TLRs controlearly parasitaemia in non-lethal Plasmodium yoelii infection. Microbes Infect. 10:1259–1265.

34. Layland, L. E., R. Rad, H. Wagner, and C. U. da Costa. 2007. Immunopathologyin schistosomiasis is controlled by antigen-specific regulatory T cells primed inthe presence of TLR2. Eur. J. Immunol. 37: 2174–2184.

35. Kossodo, S., C. Monso, P. Juillard, T. Velu, M. Goldman, and G. E. Grau. 1997.Interleukin-10 modulates susceptibility in experimental cerebral malaria. Immu-nology 91: 536–540.

36. Omer, F. M., J. B. de Souza, and E. M. Riley. 2003. Differential induction ofTGF-� regulates proinflammatory cytokine production and determines the out-come of lethal and nonlethal Plasmodium yoelii infections. J. Immunol. 171:5430–5436.

37. Wing, K., Y. Onishi, P. Prieto-Martin, T. Yamaguchi, M. Miyara, Z. Fehervari,T. Nomura, and S. Sakaguchi. 2008. CTLA-4 control over Foxp3� regulatory Tcell function. Science 322: 271–275.

38. Lepenies, B., I. Gaworski, S. Tartz, J. Langhorne, B. Fleischer, and T. Jacobs.2007. CTLA-4 blockade differentially influences the outcome of non-lethal andlethal Plasmodium yoelii infections. Microbes Infect. 9: 687–694.

7022 ROLE OF Treg DURING CM

by guest on March 26, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from