B R I E F COM M U N I C AT I O N S

Escape of malaria parasites from host immunity requiresCD4+CD25+ regulatory T cellsHajime Hisaeda1, Yoichi Maekawa1, Daiji Iwakawa1, Hiroko Okada1,Kunisuke Himeno2, Kenji Kishihara1, Shin-ichi Tsukumo1 & Koji Yasutomo1

Infection with malaria parasites frequently induces totalimmune suppression, which makes it difficult for the host tomaintain long-lasting immunity. Here we show that depletionof CD4+CD25+ regulatory T cells (Treg) protects mice fromdeath when infected with a lethal strain of Plasmodium yoelii,and that this protection is associated with an increased T-cellresponsiveness against parasite-derived antigens. These resultssuggest that activation of Treg cells contributes to immunesuppression during malaria infection, and helps malariaparasites to escape from host immune responses.

Malaria parasites cause approximately 2 million deaths per year1.The generation of effector T cells is crucial for protective immu-nity against blood-stage malaria2. However, the parasites haveacquired mechanisms to escape T-cell immunity, including anti-genic diversity3, clonal antigenic variation4–6 and impairment ofdendritic cell maturation by parasitized red blood cells (RBCs)7.Malaria patients frequently show reduced immune responses not

only to the malaria parasite, but also to unrelated antigens8–10,suggesting that an active T-cell suppression mechanism may oper-ate during the course of malaria infection.

CD4+CD25+ Treg cells suppress CD4+ and CD8+ T-cell activa-tion11–13. Although these professional regulatory cells prevent theactivation and proliferation of potentially autoreactive T cells11,Treg cells have recently been shown to contribute to the mainte-nance of chronic infection by Leishmania major14. The functionalrole of Treg cells in immune responses against infectious organismsremains unclear, however. We therefore used a rodent malariamodel to address the question of whether Treg cells are crucial forthe immune suppression observed in malaria infection.

There are two substrains of the rodent malaria parasitePlasmodium yoelii strain 17X. One substrain, PyL, is highly viru-lent in mice and causes lethal infection; the other, PyNL, causes aself-limiting, nonlethal infection. PyL, a variant derived fromPyNL15, has a genetic and antigenic background identical to thatof PyNL. BALB/c mice infected with PyL showed rapid elevation ofparasitemia and die within 13 d, whereas mice infected with PyNLexperienced transient parasitemia (up to 30%) and eradicated theparasites within 25 d, with no mortality (Fig. 1a). In mice withsevere combined immunodeficiency, PyNL infection caused rapidelevation of parasitemia and death comparable to PyL infection(Fig. 1a). These observations suggest that immunocompetent micecan develop protective immunity against PyNL, but not againstPyL. We therefore compared the immune responses of miceinfected with PyL and PyNL. Splenocytes obtained from

1Department of Immunology & Parasitology, School of Medicine, The University of Tokushima, 3-18-15 Kuramoto, Tokushima, 770-8503, Japan. 2Department of Parasitology, Faculty of Medical Sciences, Kyushu University, 3 Maidashi, East Ward Fukuoka, 812-8582, Japan. Correspondence should be addressed to K.Y. (yasutomo@basic.med.tokushima-u.ac.jp).

Published online 21 December 2003; doi:10.1038/nm975

NATURE MEDICINE VOLUME 10 | NUMBER 1 | JANUARY 2004 29NATURE MEDICINE VOLUME 10 | NUMBER 1 | JANUARY 2004 29

a b

Figure 1 Immunosuppression in PyL-infected mice. (a) Immunocompetent BALB/c or immunodeficient (SCID) mice were infected with 10,000 RBCsparasitized with PyL or PyNL. Parasitemia was monitored by microscopic evaluation of thin blood films stained with Giemsa solution. Each symbol representsa value from an individual mouse. Numbers represent times of mouse death. (b) Proliferation of splenocytes (2 × 105) isolated from uninfected mice (�) ormice 7 d after infection with PyL (�) or PyNL (�). Splenocytes were cultured with the indicated number of PyL-parasitized RBCs (pRBC) for 72 h, andincubated with 1 µCi/well of [3H]thymidine for the last 6 h. Parasitized RBCs were purified from the blood of infected mice after eliminating host white bloodcells by passing through a CF-11 column, followed by discontinuous gradient centrifugation on 45% Percoll. Cultured cells were harvested onto glass-fiberfilters, and incorporated radioactivity was measured with a liquid scintillation counter. Results represent the mean ± s.d. of triplicate cultures.

©20

04 N

atur

e P

ublis

hing

Gro

up

http

://w

ww

.nat

ure.

com

/nat

urem

edic

ine

B R I E F COM M U N I C AT I O N S

PyNL-infected mice showed a dose-dependent proliferative response in vitro toRBCs parasitized with PyL, whereassplenocytes from PyL-infected miceshowed only a marginal proliferativeresponse (Fig. 1b). In addition, CD4+ Tcells obtained from PyL-infected miceshowed a very low proliferative response tostimulation with CD3-specific monoclonalantibody (data not shown). These resultssuggest active T-cell immune suppressionin PyL-infected mice.

To examine whether Treg cells contributeto the suppression of the T-cell response,we depleted BALB/c mice of CD25+ cellsand then infected them with PyL.Administration of CD25-specific mono-clonal antibody almost completely depletedCD4+CD25+ cells in BALB/c mice (Fig. 2a).Infection of untreated mice with PyLcaused severe parasitemia, whereas micedepleted of CD25+ cells suffered from twowaves of parasitemia and eventually com-pletely cleared the parasites (Fig. 2b). Incontrast, injection of CD25-specific mono-clonal antibody did not change the courseof infection with PyNL, in terms of para-sitemia and survival (Fig. 2b). This discre-pancy in the effect of CD25+ cell depletionstrongly suggests a specific contribution ofTreg cells to PyL infection. In addition,these, mice were not protected from PyL when injected with iso-type control IgM (data not shown), confirming that the effects ofthe CD25-specific antibody were due to depletion of Treg cells.

Several observations in this study suggest that antibody treat-ment of PyL-infected mice depleted immunosuppressiveCD4+CD25+ T cells. First, there were few (CD4–)CD25+ cells pres-ent in the spleen and lymph nodes before infection (data notshown), suggesting that most of the depleted CD25+ cells wereCD4+. Second, splenocytes from PyL-infected mice that weredepleted of CD25+ cells mounted a vigorous proliferative responsein vitro against parasitized RBCs (Fig. 2c). Finally, CD4+ T cellsfrom mice treated with the antibody to CD25 showed increased T-cell receptor–mediated responses compared with untreated mice(data not shown). These results indicate that depletion of Treg cellsreversed the immunosuppression observed in PyL-infected mice.

Our data indicate that Treg cells have a crucial role in immunesuppression during malaria infection, as well as the escape of para-sites from host immune surveillance. Although we have not yetidentified the mechanism by which PyL specifically activates Tregcells, these findings provide new insight regarding the escapemechanism of microorganisms from the host immune response andmay help to establish therapeutic strategies for parasitic infections.

ACKNOWLEDGMENTSWe thank M. Tsubosaka, N. Kanbe, K. Ishii and S. Hamano for assistance with

experiments, and J.R. Dorfman for helpful comments. This work was supported by grants-in-aid from the Ministry of Education, Culture, Sports, Science andTechnology in Japan and the Industrial Technology Research Grant Program fromNEDO of Japan (to K.Y.). The experiments were approved by the animal carecommittee of the University of Tokushima.

COMPETING INTERESTS STATEMENTThe authors declare that they have no competing financial interests.

Received 24 September; accepted 3 December 2003Published online at http://www.nature.com/naturemedicine/

1. Richie, T.L. & Saul, A. Nature 415, 694–701 (2002).2. Good, M.F., Kaslow, D.C. & Miller, L.H. Annu. Rev. Immunol. 16, 57–87

(1998).3. Plebanski, M. et al. Immunity 10, 651–660 (1999).4. Chen, Q. et al. Nature 394, 392–395 (1998).5. Su, X.Z. et al. Cell 82, 89–100 (1995).6. Baruch, D.I. et al. Cell 82, 77–87 (1995).7. Urban, B.C. et al. Nature 400, 73–77 (1999).8. Williamson, W.A. & Greenwood, B.M. Lancet 1, 1328–1329 (1978).9. Viens, P., Tarzaali, A. & Quevillon, M. Am. J. Trop. Med. Hyg. 23, 846–849

(1974).10. Ho, M. et al. J. Immunol. 141, 2755–2759 (1988).11. Shevach, E.M., McHugh, R.S., Piccirillo, C.A. & Thornton, A.M. Immunol. Rev.

182, 58–67 (2001).12. Shevach, E.M. J. Exp. Med. 193, F41–F46 (2001).13. Sakaguchi, S. Cell 101, 455–458 (2000).14. Belkaid, Y., Piccirillo, C.A., Mendez, S., Shevach, E.M. & Sacks, D.L. Nature

420, 502–507 (2002).15. Yoeli, M., Hargreaves, B., Carter, R. & Walliker, D. Ann. Trop. Med. Parasitol.

69, 173–178 (1975).

30 VOLUME 10 | NUMBER 1 | JANUARY 2004 NATURE MEDICINE

a b

c

Figure 2 Adverse contribution of Treg cells in protection to infection with PyL. (a) Flow cytometricanalysis of peripheral blood leukocytes obtained from mice before (Pre), and 1 d after (Post),intraperitoneal injection of CD25-specific monoclonal antibody (7D4; 500 µg; American Type CultureCollection). Cells were stained with FITC-conjugated monoclonal antibody to CD4 (GK1.5; BDPharMingen) and phycoerythrin-conjugated monoclonal antibody to CD25 (PC61.5; BD PharMingen).(b) Parasitemia of BALB/c mice infected with 10,000 PyL-parasitized (top) or PyNL-parasitized(bottom) RBCs, and intraperitoneally injected with 500 µg of CD25-specific monoclonal antibody 7D4(right) or PBS (left) on days –3, –1 and 5. Each symbol represents a value from an individual mouse. All PyL-infected control mice died at days 8, 9, 9, 10, 10 and 10. One PyL-infected antibody-treatedmouse died on day 10. (c) Proliferation of splenocytes isolated from mice treated with antibody toCD25 (�) or left untreated (�), 7 d after PyL infection, or from uninfected mice (�). Splenocytes werecultured with the indicated number of PyL-parasitized RBCs (pRBC). See Figure 1b legend for details.

©20

04 N

atur

e P

ublis

hing

Gro

up

http

://w

ww

.nat

ure.

com

/nat

urem

edic

ine