0014-2980/01/0101-265$17.50+.50/0© WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001

Neutrophil depletion exacerbates experimentalChagas’ disease in BALB/c, but protects C57BL/6mice through modulating the Th1/Th2 dichotomy indifferent directions

Lin Chen1, Tadashi Watanabe1, Hisami Watanabe2 and Fujiro Sendo1

1 Department of Immunology and Parasitology, Yamagata University School of Medicine,Yamagata, Japan

2 Department of Immunology, Niigata University School of Medicine, Niigata, Japan

To elucidate the roles of neutrophils in experimental Chagas’ disease, we depleted theperipheral neutrophils in BALB/c and C57BL/6 mice with a monoclonal antibody 1 daybefore Trypanosoma cruzi infection. Neutrophil depletion in BALB/c mice resulted in exacer-bation of the disease and decreased expression of mRNA for Th1 cytokines, including IL-2and IFN- + , IL-12p40 and TNF- § in their spleens after the infection, while a Th2 cytokine,IL-10, increased especially 1 day after infection. Neutrophils from infected BALB/c miceexpressed mRNA for IL-12p40, IFN- + , TNF- § and Th1 chemoattractive chemokines, mono-kine induced by IFN- + (MIG) and macrophage inflammatory protein-1 § (MIP-1 § ). In contrast,in C57BL/6 mice neutrophil depletion induced resistance to the disease and enhanced theexpression of the above Th1 cytokines, although IL-10 mRNA in neutrophil-depleted C57BL/6 mice was also higher than in control mice. Neutrophils from C57BL/6 mice did not expressIL-12p40, IFN- + and MIG but expressed TNF- § , MIP-1 § and IL-10. Therefore, neutrophilsmay play opposite roles in these two strains of mice with respect to protection versus exac-erbation of T. cruzi infection, possibly through modulating the Th1/Th2 dichotomy in differentdirections.

Key words: Trypanosoma cruzi / Neutrophil / Cytokine / Chemokine / T helper

Received 17/1/00Revised 28/9/00Accepted 5/10/00

[I 20489]

Abbreviations: MIG: Monokine induced by IFN- q MIP-1 § :Macrophage inflammatory protein-1 § MDC: Monocyte-derived chemokine PEN: Peritoneal exudate neutrophilsPEM: Peritoneal exudate macrophages

1 Introduction

Trypanosama cruzi, an obligate intracellular protozoanparasite, infects a variety of nucleated cells, replicateswithin the cytoplasm, and induces Chagas’ disease.Host resistance to experimental Chagas’ diseasedepends on both innate and acquired immunity. Miceinfected with T. cruzi have been used as a model for thehuman disease because like humans they developdetectable parasitemias during acute infection, which isthen followed by chronic tissue parasitism. However, it iswell known that different strains of inbred mice showmarkedly different patterns of acute infection. BALB/cand C57BL/6 mice, which have been regarded as mod-els of Th1- and Th2-dominant conditions, respectively, inLeishmania major infection [1], are widely used in experi-

mental Chagas’ disease. BALB/c mice are relatively sus-ceptible to T. cruzi infection, whereas C57BL/6 mice arerelatively resistant [2]. The basis of resistance is not yetunderstood. These two strains of mice also exhibit differ-ent patterns of cytokine response to T. cruzi infection [3,4]. Major histocompatibility complex (H-2) haplotype,which is important in regulating immune response, maybe involved in the resistance, but the genetic back-ground outside the H-2 complex may also influence thehost resistance [5].

For a long time the role of neutrophils in immunoregula-tion received little attention. However, recent evidencefrom our previous studies [6, 7] and those of others [8, 9]suggests that neutrophils can contribute significantly tothe immune response by modulating both cellular andhumoral immunity, especially via the synthesis andrelease of immunoregulatory cytokines.

Based on our previous work, we hypothesized that neu-trophils may be involved in immunomodulation in experi-mental Chagas’ disease. To address this, we infectedBALB/c and C57BL/6 mice with T. cruzi after their

Eur. J. Immunol. 2001. 31: 265–275 Different role of neutrophils in experimental Chagas’ disease 265

Fig. 1. Neutrophil depletion changes the courses of T. cruzi infection in BALB/c and C57BL/6 mice. BALB/c and C57BL/6 micewere infected with trypomastigotes 24 h after treatment with RB6-8C5 or normal rat IgG. (A, B) and (C, D) show the parasitemiaand survival after T. cruzi infection in neutrophil-depleted BALB/c and C5BL/6 mice, respectively. Results are expressed asmeans ± SD; pooled data from two separate comparable experiments are shown.

peripheral blood neutrophils had been depleted by anmAb. Interestingly, we found that neutrophils play oppo-site roles in these two strains of mice with respect to pro-tection or exacerbation of T. cruzi infection, possiblythrough modulating the Th1/Th2 dichotomy in differentdirections. Our results may add to the understanding ofthe role of Th1/Th2 dichotomy for pathogenesis ofT. cruzi infection, which has been regarded as more com-plicated than that in L. major infection [3, 10].

2 Results

2.1 Neutrophil depletion exacerbates T. cruziinfection in BALB/c, but inhibits it inC57BL/6 mice

To examine whether neutrophils are involved in protec-tion and/or pathogenesis in T. cruzi infection, BALB/cand C57BL/6 mice were infected with 1 × 104 trypomas-tigotes, and the course of infection was comparedbetween mice treated with RB6-8C5 mAb that selec-tively depletes neutrophils and those treated with controlIgG. During the acute phase (12–16 days after infection),neutrophil-depleted BALB/c mice exhibited an in-creased level of parasitemia compared to the non-depleted controls (p X 0.01) (Fig. 1 A). Moreover, all mAb-

treated BALB/c mice died between 14 and 18 days afterinfection, whereas one third of the control mice survivedthe acute phase (Fig. 1 B) (p X 0.001). This result demon-strates that neutrophils may play a protective role in theBALB/c mice infected with T. cruzi. However, it wasunexpected that treatment with the RB6-8C5 mAbdecreased the level of parasitemia in C57BL/6 mice inthe acute phase at 22 days after infection (Fig. 1 C), andincreased their survival compared to control mice,although the difference was not significant (p = 0.224),(Fig. 1 D). Therefore, neutrophils may play opposite rolesin BALB/c and C57BL/6 mice, protecting the former andexacerbating the infection in the latter. However, itshould be noted that parasitemias in IgG-treated BALB/c mice were significantly higher than those in similarlytreated C57BL/6 mice at 18, 20 and 22 days after theinfection (p X 0.05) (Fig. 1 A, C), which might influencethe effects of RB6-8C5 in these two mouse strains.

To determine the effects of neutrophil depletion by mAbRB6-8C5 on various types of heamtopoietic cells in nor-mal BALB/c and C57BL/6 mice, we examined each cellpopulation in the spleen, lymph node and the peripheralblood after treatment with this mAb. Using flow cytome-try, a slight decrease in CD8+ T cells was observed aftertreatment in these tissues in both group of mice, themechanisms for which are unknown, although these T

266 L. Chen et al. Eur. J. Immunol. 2001. 31: 265–275

Fig. 2. Neutrophil depletion alters the histopathology ininfected mice. BALB/c and C57BL/6 mice were infectedwith trypomastigotes 24 h after treatment with RB6-8C5 ornormal rat IgG. Spleens obtained 1 day after infection wereexamined for neutrophil infiltration by immunohistochemis-try (A, B). At 14 days after infection, both spleens and heartswere examined for their parasite burden using hematoxylin-eosin staining (C–F). (A) A spleen section from a BALB/cmouse treated with control IgG. Note many neutrophils inthe red pulp; (B) a spleen section from a BALB/c mousetreated with RB6-8C5. Note the absence of neutrophils. (C)Heart and (E) spleen sections from a BALB/c mouse treatedwith control igG. (D) Heart and (F) spleen sections from aBALB/c mouse treated with RB6-8C5. The arrows indicateparasite nests. The results with C57BL/6 mice are notshown. Bars = 20 ? m.

cells did not react in vitro with RB6-8C5 (data notshown).

2.2 Neutrophil depletion alters thehistopathology of the infected mice

To investigate whether neutrophils are involved in earlyinflammation or innate immunity in T. cruzi infection, weexamined infiltrating neutrophils in the spleens usingimmunohistochemistry. There were infiltrating neutro-phils in the red pulp of the spleens of control mice(Fig. 2A), and their number was higher in BALB/c mice(41 ± 4 in 10 microscopic fields) compared with that inC57BL/6 mice (2 ± 0). However, infiltration neutrophilswere hardly seen in the spleen of the neutrophil-depletedmice (Fig. 2B). We confirmed these immunohistochemi-cal findings by hematoxylin-eosin staining (data notshown). To determine whether mortality is associatedwith parasite burden, we counted the numbers of amas-tigote nests, which show parasite replication in heartsand spleens, at 14 days after infection. There were morefoci in the hearts and the spleens of RB6-8C5-treatedBALB/c mice than in control BALB/c mice, but not differ-ences were observed in C57BL/6 mice at this time(Fig. 2C–F and Table 1).

2.3 Neutrophil depletion changes cytokinemRNA expression in the spleen of infectedmice

Since cytokines such as IL-2 [11], IL-4 [4], IL-10 [12, 13],IL-12 [14], IFN- + [12, 15] and TNF- § [16] are involved inthe immune response to T. cruzi infection, and neutrophildepletion altered the course of the infection in BALB/cand C57BL/6 mice, we therefore investigated whetherthe selective neutrophil depletion results in changes inthe expression of various cytokine mRNA in the spleen,which may contribute to altering the course of the infec-tion. Figs. 3 and 4 and Table 2 show the expression lev-els of cytokine mRNA in BALB/c and C57BL/6 mice, asanalyzed by competitive reverse transcriptase (RT)-PCR.In BALB/c mice, IL-12p40 (day 1), IFN- + (days 1, 8), TNF-§ (days 8, 14), and IL-2 (days 1, 8, 14) were lower in the

neutrophil-depleted group than that in controls. In con-trast, in C57BL/6 mice expression of these cytokineswas higher in neutrophil-depleted mice than in controlsat some of days observed. As for IL-10, the expression ofmRNA in BALB/c mice was higher in the neutrophil-depleted group only on day 1, and in C57BL/6 mice thiswas the case observed at all the days observed. Theexpression of IL-4 was only detected on day 14 in BALB/c mice and there was no difference between neutrophil-depeleted and control groups.

2.4 Expression of cytokine and chemokinemRNA in neutrophils and macrophages fromT. cruzi-infected mice

A possible mechanism by which neutrophil depletioncaused modulation of cytokine production by spleencells in infected mice may be that cytokine production byneutrophils was eliminated. Therefore, we examinedcytokine expression in neutrophils in T. cruzi-infectedmice. In addition, recent studies demonstrated that neu-trophils produce chemokines [17–19], some of which

Eur. J. Immunol. 2001. 31: 265–275 Different role of neutrophils in experimental Chagas’ disease 267

Fig. 3. Competitive RT-PCR assay of IL-2, IL-4, IL-10, IL-12p40, IFN- + , and TNF- § mRNA in the spleens of BALB/c miceinfected with T. cruzi whose neutrophils were or were not depleted. BALB/c mice treated with RB6-8C5 or normal rat IgG (con-trols) were infected with trypomastigotes. Spleens were obtained at 1, 8 and 14 days after infection. mRNA were reverse tran-scribed and the concentrations of cDNA were adjusted to equal amounts of g -actin cDNA. The transcripts of these cytokineswere amplified in the presence of competitors. For quantification of IL-4 and IL-12p40, the competitors were serially diluted ten-fold starting at 40,000 and 60,000 attomol, respectively; for IFN- + , the competitor was serially diluted fourfold starting at 16 atto-mol (days 1 and 8) or 250 attomol (day 14); for TNF- § , the competitor was serially diluted fourfold starting at 16 attomol (days 1and 8, RB) or 250 attomol (days 8 IgG and 14); for IL-2, the competitor was serially diluted fourfold starting at 250 attomol (days1 and 8) or 1,000 attomol (day 14); for IL-10, the competitor was serially diluted forfold starting at 16 attomol (days 1, RB, and day8) or 4 attomol (day 1, IgG) or 1,000 attomol (day 14, RB) and 4,000 attomol (day 14, IgG). If there was no target cytokine cDNAamplified when the lowest dilution of competitor was used in competitive PCR, the negative result was confirmed using the samePCR conditions without adding competitor (data not shown). The amount of competitor which resulted in equal or approximatelyequal amounts of target cytokine cDNA in PCR is indicated above the amplified competitor. N: normal, RB: RB6-8C5, IgG: nor-mal rat IgG. Similar results were obtained in a duplicate experiment.

Table 1. Neutrophil depletion increases parasite burden in BALB/c, but not C57BL/6, mice 14 days after infectiona)

Mice Heartb) Spleenc) n

RB6-8C5 IgG RB6-8C5 IgG

BALB/c 29.5 ± 1.32* 11.67 ± 2.58 22.67 ± 9.60** 5.04 ± 2.08 5

C67BL/6 7.83 ± 1.61 5.67 ± 2.26 6.33 ± 2.52 8.00 ± 3.00 5

a) The results are expressed as the means ± SD. Similar results were obtained in another experiment.b) The numbers of parasite nests per section.c) Total numbers of parasite nests in 10 fields of a spleen section at a magnification of × 40.* p X 0.001, ** p X 0.05 between RB6-8C5 and normal rat IgG treated BALB/c mice.

seems to be key factors for determining the Th1/Th2dichotomy [20, 21]. Therefore, we also examined theexpression of some chemokine mRNA in neutrophilsfrom infected mice. Furthermore, since cytokines orchemokines produced by neutrophils should affectmacrophage functions, including cytokine or chemokine

production, we examined the expression of cytokine/chemokine mRNA in macrophages from neutrophil-depleted and control mice. As shown in Fig. 5 andTable 3, peritoneal exudate neutrophil (PEN) expressedhigh levels of IL-12p40 mRNA in BALB/c mice, but not inC57BL/6 mice, but IL-12p40 was also not detectable in

268 L. Chen et al. Eur. J. Immunol. 2001. 31: 265–275

Table 2. Competitive RT-PCR analysis of cytokine mRNA in the spleens of neutrophil-depleted or undepleted BALB/c andC57BL/6 mice infected with T. cruzia)

Cyto-kines

BALB/c mice C57BL/6 mice

Day 1 Day 8 Day 14 Day 1 Day 8 Day 14

RB IgG RB IgG RB IgG RB IgD RB IgG RB IgG

IL-12p40 ND 60 60–600 600 60–600 600 600 6 600 60 60–600 ND

IFN- + ND 0.06 0.06 0.25 4 4 0.25 0.06 4 0.25 1 0.25

TNF- § 0.25 0.25 0.25 16 4 16 4 1 16 1 4 0.25

IL-2 16 52.5 16 250 250 1000 62.5 16 62.5 0.06 16 0.25

IL-4 ND ND ND ND 40 40 ND ND ND ND ND ND

IL-10 0.06 0.015 0.25 0.25 62.5 62.5 16 4 16 4 62.5 16

a) The original data are shown in Fig. 3 and Fig. 4. Values indicate the attomoles of competitor which resulted in equal orapproximately equal amounts of target cytokine cDNA in competitive PCR. ND: not detectable, RB: RB6-8C5 mAb, IgG: ratIgG.

Fig. 4. Competitive RT-PCR assay of IL-2, IL-4, IL-10, IL-12p40, TNF- § and IFN- + mRNA from the spleens of C57BL/6 miceinfected with T. cruzi. mRNA were reverse transcribed and the concentrations of cDNA were adjusted to be equal to the g -actinin samples of the BALB/c mice. For quantification of IL-4 and IL-12p40, the competitors were serially diluted as for BALB/c mice(Fig. 3); for IFN- + , the competitor was serially diluted fourfold starting at 16 attomol for all samples except that the day 8 RBsample competitor was diluted starting at 62.5 attomol; for TNF- § , the competitor was serially diluted fourfold starting at 62.5attomol for all samples except the day 14 IgG sample competitor was diluted starting at 16 attomoles; for IL-2, the competitorwas serially diluted fourfold starting at 250 attomol (day 1) or 1,000 attomol (days 8 and 14, RB) or 16 (days 8 and day 14, IgG);for IL-10, the competitor was serially diluted fourfold starting at 16 attomol (days 1 and 8, RB) or 250 attomol (day 8, IgG, and day14). Similar results were obtained in a duplicate experiment.

peritoneal exudate macrophages (PEM) from neutrophil-depleted and control BALB/c and C57BL/6 mice. BothPEN and PEM from control BALB/c mice expressed IFN-+ and TNF- § . T and NK cells are generally assumed to be

IFN- + -producing cells. However, recent studies haveshown that both neutrophils and macrophages expressand secrete IFN- + [22, 23], and our results confirmedthese observations in T. cruzi-infected BALB/c mice.

Eur. J. Immunol. 2001. 31: 265–275 Different role of neutrophils in experimental Chagas’ disease 269

Table 3. RT-PCR analysis of cytokine and chemokine mRNA in neutrophils and macrophagesa)

Cytokines/chemokines

BALB/c mice C57BL/mice

N RB-M ¤ IgG-M ¤ N RB-M ¤ IgG-M ¤

IL-12p40 0.65 ND ND ND ND ND

IFN- + 0.12 ND 0.30 ND ND ND

TNF- § 0.52 ND 0.39 1.12 1.32 ND

IL-10 1.30 0.16 0.44 0.70 0.50 0.37

MIG 0.28 ND 0.26 ND ND ND

MIP-1 § 0.76 0.18 0.42 1.06 1.02 0.45

MDC ND 0.28 ND ND 0.09 ND

a) The original data are shown in Fig. 5. Values indicate target/ g -actin densities. N: Neutrophils, M ¤ : macrophages, RB: RB6-8C5, IgG: rat IgG, ND: not detectable.

Fig. 5. Expression of cytokine and chemokine mRNA byPEN and PEM after T. cruzi infection. PEN from control micetreated with IgG, and PEM from both neutrophil-depletedand control mice 24 h after infection were obtained asdescribed in Sect. 4.4. mRNA of cytokines IL-12p40, IFN- + ,TNF- § and IL-10 and chemokines MIG, MIP-1 § and MDCwere analyzed by RT-PCR.

PEN and PEM from C57BL/6 mice did not express IFN-+ , but TNF- § was highly expressed in these cells fromneutropenic C57BL/6 mice. PEN from both strains ofmice had high levels of IL-10 expression, but this wasdecreased in PEM from neutropenic BALB/c mice. PENalso expressed the Th1 chemoattractive chemokinesmonokine induced by IFN- + (MIG) and macrophageinflammatory protein-1 § (MIP-1 § ) in BALB/c mice andMIP-1 § in C57BL/6 mice. Neutrophil depletion causedan absence or lower expression of these chemokines in

macrophages from BALB/c mice, but not from C57BL/6mice. The Th2 chemoattractive monocyte-derivedchemokine (MDC) was detectable in PEM from bothBALB/c and C57BL/6 neutropenic mice, but wasexpressed at a higher level in neutropenic BALB/c thanin neutropenic C57BL/6 mice. Transcripts of thischemokine were not detectable in PEN from eithermouse strain. Neither PEN nor PEM from normal miceexpressed cytokines or chemokines (data not shown).These results provide the direct evidence thatneutrophils express Th1 cytokines and Th1chemoattractive chemokines at the very early stage ofT. cruzi infection in BALB/c mice, but except for MIP-1 §these molecules are not detectable in C57BL/6 mice.Neutrophil depletion changes the patterns of expressionof some cytokines and chemokines in macrophages,which may play a part in the results obtained from in vivostudies. To determine whether neutrophils actuallyproduce Th1 cytokine, we examined IFN- + production byneutrophils obtained under various conditions. PEN fromT. cruzi-infected BALB/c mice produced large amount ofIFN- + when stimulated with the trypomastigotes,whereas under the same conditions PEN from theinfected C57BL/6 mice did not (Table 4), which mimicsthe results of IFN- + mRNA expression under the similarconditions, although some differences were observedbetween IFN- + production and its mRNA expressionwhich might be resulted from differences in experimentalconditions.

3 Discussion

In the present study, we found that neutrophil depletioncauses exacerbation of T. cruzi infection in BALB/c mice,but protection in C57BL/6 mice, and reduces the

270 L. Chen et al. Eur. J. Immunol. 2001. 31: 265–275

Table 4. Production of IFN- + by neutrophils from uninfected and T. cruzi-infected BALB/c and C57BL/6 micea)

Concentration of IFN- + (pg/ml)

Strains of mice Sources of neutrophil Stimulants

Medium LPS Trypomastigotes

BALB/c Uninfected X 37 44.6 ± 7.9 686.6 ± 113.0

Infected X 37 117.7 ± 13.1 1307.0 ± 145.0*

C57BL/6 Uninfected X 37 X 37 94.3 ± 16.5

Infected X 37 X 37 139.2 ± 19.7*

a) PEN from uninfected BALB/c or C57BL/6 mice or at 24 h after trypomastigote infection were cultured in the presence ofeither LPS or live trypomastigotes. After 24 h of culture, supernatants were assessed for IFN- + content by ELISA. Results areexpressed as means ± SD of three experiments.

* p X 0.01.

expression of protective cytokine mRNA in BALB/c micebut enhances those in C57BL/6 mice.

Evidence obtained in the last 10 years shows that cyto-kines play key roles in regulating both parasite replica-tion and immune responses in infected animals. IL-12produced by macrophages is a potent inducer of IFN- +and mediate resistance to T. cruzi infection [14]. Adminis-tration of IL-2 to infected mice reduces parasitemia andincreases longevity [24], and TNF- § and IFN- + activatemouse macrophages synergistically to kill intracellularT. cruzi through a nitric oxide-dependent mechanism[16]. IL-10, one of the Th2 cytokines, mediates suscepti-bility to T. cruzi infection [13]. These studies clearly showthat Th1 cytokines and TNF- § play critical roles in resis-tance to T. cruzi infection, whereas Th2 cytokine may beinvolved in mediating susceptibility. These observationsare consistent with our present results with BALB/cmice, which show that at the very early phase of infectionthe lower expression of Th1 cytokines and TNF- § , andthe higher expression of IL-10 due to neutrophil deple-tion are related to susceptibility to T. cruzi infection. Ourstudy also confirms that the positive effect of IFN- + onhost resistance, which is perhaps triggered by IL-12,appears to be restricted to the very early stage of acuteinfection [14], since we found no difference in the expres-sion of this cytokine 14 days after infection. Furthermore,we found that neutrophils from BALB/c mice at the veryearly stage of infection expressed IL-12p40, IFN- + ,TNF- § and Th1 chemoattractive chemokines (MIG andMIP-1 § ), and neutrophils depletion resulted in down-regulation of IFN- + , TNF- § and Th1 chemoattractive che-mokines in macrophages, and induction of macrophageexpression of the Th2 chemoattractive chemokine,MDC.

Our study suggests that in infected BALB/c mice cyto-kines and chemokines produced by neutrophils may playa role in promoting the production of protective cyto-kines by other cells such as macrophages, NK cells andlymphocytes. Undoubtedly neutrophil depletion causesa change in cytokine and/or chemokine expression,which results in sequential modulation of the functions ofmonocytes and lymphocytes. Our previous study indi-cated that not only neutrophils, but also macrophagesand lymphocytes hardly infiltrated into the peritonealcavity of neutrophil-depleted rats after stimulation withOK-432, a streptococcal agent which acts as a biologicalresponse modifier and is used in cancer therapy. In thatstudy, neutrophils were shown to transmit informationfrom resident macrophages to exudated inflamma-tory macrophages in tumor inhibition [25]. A similarneutrophil-mediated activation of macrophages by Cory-nebacterium parvum has been reported [26]. Combinedwith these reports, our findings suggests that in T. cruzi-infected BALB/c mice neutrophils play a major part inthe afferent or inductive limb of the immune response byregulating the functions of macrophages and lympho-cytes.

It is of particular interest that infected neutrophil-depleted C57BL/6 mice had lower parasitemia andslightly longer survival than control mice. Neutrophildepletion resulted in higher expression of IL-2, IL-12p40,IFN- + , and TNF- § mRNA, which is the opposite to thatfound with BALB/c mice. Although the mechanism(s)responsible for this difference between BALB/c andC57BL/6 mice is not clear, possible explanations mayinclude the following: (1) the profile of cytokines pro-duced by neutrophils in T. cruzi-infected BALB/c andC57BL/6 mice differs because neutrophils in BALB/cmice express predominantly Th1 cytokine (IFN- + ), IL-

Eur. J. Immunol. 2001. 31: 265–275 Different role of neutrophils in experimental Chagas’ disease 271

12p40 and TNF- § as well as Th1 chemoattractive che-mokines (MIG and MIP-1 § ), but neutrophils in C57BL/6express only TNF- § and MIP-1 § . (2) Neutrophils inC57BL/6 mice infected with T. cruzi may produceunknown inhibitory factor(s) that suppress the produc-tion of protective cytokines and/or chemokines. Wehypothesize that the manner in which macrophage orlymphocyte cytokine/chemokine production is regulatedby neutrophils may differ in these two strains of mice. (3)Recently, Mac-1+/Gr-1+ myeloid suppressor cells havebeen reported to induce apoptosis of CD8+ lymphocytes[27]. These suppressor cells might be eliminated inT. cruzi infected mice by treatment with mAb RB6-8C5.However, this possibility does not seem likely sincethese suppressors appear 4 days after immunization[27], and but treatment of C57BL/6 mice with mAb RB6-8C5 at that time had no effects on parasitemia and mor-tality (Chen, unpublished results), suggesting that killingof myeloid suppressors in C57BL/6 mice with the mAbmay not be involved in the phenomena observed in thisstudy. Taken together, it seems that neutrophils maymodulate the immune response in different waysdepending on the conditions employed, although themechanisms are still obscure.

The difference in the effects of RB6-8C5 in BALB/c andC57BL/6 mice may also be explained by basic differenceof these two strains, e.g. as seen in parasitemia (Fig. 1),parasite numbers in the heart (Table 1), and neutrophilinfiltration into the spleen of T. cruzi-infected mice nottreated with RB6-8C5 (see Sect. 2.2).

The susceptibility or resistance to certain infections inexperimental models have been associated with certainprofiles of cytokine production. A predominant Th1-typeresponse, characterized by the production of IFN- + andIL-2, is associated with healing in experimental leish-mania infections, while a Th2 response, characterized bythe cytokines IL-4, IL-5, and IL-10, is associated withexacerbation of the disease [28]. The situation seems tobe different in T. cruzi infection. As we and others [12, 29,30] have found, both Th1 and Th2 cytokines are pro-duced in susceptible mice. On the other hand, Launois etal. [31, 32] demonstrated that IL-4 mRNA is detectable inNK1.1– CD4 T cells from draining lymph nodes of sus-ceptible BALB/c mice as early as 16 h after L. majorinfection, and this early produced IL-4 induces IL-12unresponsiveness, suggesting that a very early immuneresponse in terms of cytokine production is a key factorin the subsequent development of the Th1/Th2 dichot-omy. In this sense, neutrophils are one of the first effec-tors in acute inflammation and the immune response.Therefore, the profile of cytokines or chemokines pro-duced by neutrophils may determine the direction of thesubsequent immune response.

A puzzling result of our data on parasitemia, parasiteburden or mortality that is difficult to explain from theviewpoint of cytokine/chemokine modulation concernsIL-10. As mentioned above, neutrophil depletion modu-lates the Th1/Th2 dichotomy in a direction that canexplain the changes in parasitemia or mortality in bothBALB/c and C57BL/6 mice. However, although onlyslight changes of IL-10 mRNA expression were observedin the spleens of BALB/c mice on neutrophil depletion, itwas enhanced by neutrophil depletion in the spleens ofC57BL/6 mice (Table 2, Figs. 3 and 4). Furthermore, neu-trophil depletion reduced macrophage IL-10 expressionin BALB/c mice (Fig. 5). These results are difficult toexplained according to the present consensus about theTh1/Th2 dichotomy. However, recent studies show thatthe Th1/Th2 dichotomy is not as simple as commonlythought. Some human and mouse T cells produce bothIFN- + and IL-10 [33], and IL-10 is produced not only byTh2 but also Th1 [34]. Regarding the functions of IL-10, ithas been demonstrated that it inhibits not only IFN- +production by Th1 cells but also IL-4 and IL-5 productionby Th2 cells [34]. Thus, our present results may reflectunknown complex functions of IL-10.

Taken together, our findings indicate that neutrophilsplay a major part in the development of resistance toT. cruzi in BALB/c mice and susceptibility in C57BL/6mice. Lack of neutrophils at the early phase of T. cruziinfection in both BALB/c and C57BL/6 mice alters theexpression of Th1 cytokines (IL-2 and IFN- + ), IL-12p40and TNF- § , as well as the Th2 cytokine, IL-10, but notthe expression of IL-4 in the spleen. The changed patternin the expression of cytokines which may be caused bythe absence of neutrophil-derived cytokines or chemoki-nes may contribute to susceptibility or resistance toexperimental Chagas’ disease. The work presented heresheds new light on the roles of neutrophils in the resis-tance or pathogenesis of T. cruzi infection which will beuseful for future research. To further elucidate the mech-anisms of the immunoregulatory roles of neutrophils inthis infection, an investigation of how cytokines and che-mokines expressed by neutrophils regulate the functionsof macrophages and lymphocytes is underway.

4 Materials and methods

4.1 Mice and parasite

Eight- to ten-week-old female BALB/c and C57BL/6 micewere purchased from Japan SCL Inc. (Hamamatsu, Japan)and housed in a pathogen-free environment. The percent-age of neutrophils in peripheral blood nucleated cells were11.8 ± 1.0 and 11.5 ± 0.9 for these strains, respectively,showing no significant difference between them. The Tula-

272 L. Chen et al. Eur. J. Immunol. 2001. 31: 265–275

huen strain of T. cruzi was used for infection. Infection wasroutinely maintained by transfer of blood from infected touninfected mice. For in vivo experiments, mice were infectedi.p. with 104 blood-derived trypomastigotes. Parasite levelswere evaluated every other day starting from 6 days afterinfection. Survival was recorded daily until 8 weeks afterinfection.

4.2 Antibodies and reagents

mAb RB6-8C5, was used to deplete peripheral neutrophils.RB6-8C5 is an IgG2b which reacts with Ly-6G, a 21–25-kDaGPI-anchored protein previously known as a myeloid differ-entiation antigen, Gr-1 [35]. The hybridoma which secretesRB6-8C5 was a gift from Dr. R. Coffman (DNAX ResearchInstitute, Palo Alto, CA). RB6-8C5 from hybridoma culturesupernatants was purified by ammonium sulfate precipita-tion. To deplete neutrophils, 0.25 mg of the antibody wasadministered i.p. into mice 1 day before initiating infection.Treatment with this dose of the antibody induced severeneutropenia for up to 5 days, as assessed by counting morethan 200 leukocytes on blood smears, similar to thatreported [36]. Control mice received an equivalent amount ofnormal rat IgG (Caltag Laboratories, Burlingame, CA).

4.3 Histopathology and immunohistochemistry

To determine the parasite burdens 14 days after infectionfollowing treatment with the mAb or control IgG, we chosethe heart and the spleen as the target tissues. Paraffin-embedded 3- ? m-thick sections were stained withhematoxylin-eosin. The numbers of parasite nests in twosections per heart and those in 10 fields per spleen werecounted at a magnification × 40. Previously describedimmunohistochemistry [37] was employed to examine infil-trating neutrophils in the spleens 1 day after infection.

4.4 Preparation of PEN and PEM

PEN from control IgG-treated BALB/c and C57BL/6 mice (5mice/group) 24 h after i.p. injection with 104 blood-derivedtrypomastigotes were obtained as described previously [38].For preparation of PEM, BALB/c and C57BL/6 mice (5mice/group) were injected i.p. with 4–5 ml 3 % proteosepeptone (Difco Laboratories, Detroit, MI), and 24 h later theywere treated with 0.25 mg RB6-8C5 mAb or control IgG i.p.After a further 24 h, they were injected i.p. with 104 blood-derived trypomastigotes, and 24 h later PEM were har-vested by peritoneal lavage with cold heparinized RPMI1640. Adherent cells were obtained after 3-h incubation at37 °C in a 35 × 15-mm tissue culture dish. The nonadherentcells were removed by exhaustive washing with warm PBS.The purities of both the PEN and PEM preparations wereG 95 % as assessed by examining the smears of purified

cells. Total RNA was extracted immediately from purifiedPEN and PEM without live trypomastigotes or T. cruzi anti-gen stimulation in vitro.

4.5 RNA isolation and semiquantitative RT-PCRanalysis

For analysis of the expression of cytokine mRNA in spleens,total RNA was extracted from spleens of each three mice at1, 8 and 14 days after infection, we used competitive RT-PCR as described previously [37]. For simple analysis ofcytokine and chemokine mRNA expression in PEN andPEM, 4 ? l cDNA samples adjusted to the same levels of g -actin were used to perform PCR without the addition ofcompetitors. The primer pairs used for amplification of MIG,MIP-1 § and MDC were previously published by Kanegane etal. [39], Widmer et al. [40], and Gonzalo et al. [19], respec-tively, and they amplify 479-, 284- and 464-bp fragments ofthe cDNA, respectively. The number of amplification cycleused in PCR for cytokines was 35, and 30 for MIG, MIP-1 §and MDC. With these numbers of cycles, the amounts ofPCR products had not reached saturation. Photographs ofgels were scanned and analyzed by NIH Image software.Density was evaluated and expressed as cytokine or che-mokine band intensity divided by g -actin band intensity. Thetreatment with the mAb did not modify the basal levels ofcytokine transcripts in the spleen and PEM of either strain ofuninfected mice (data not shown).

4.6 ELISA for IFN- q measurement

PEN were obtained as described above from uninfectedBALB/c and C57BL/6 mice or at 24 h after infection with 1 ×104 trypomastigotes (13–15 mice/group). PEN were culturedat 5 × 105 cells/well in 0.5 ml RPMI 1640 medium with 10 %FCS, 5 × 10−5 M 2-mercaptoethanol and 2 mM L-glutaminein 48-well plates in the presence of either 100 ng/ml LPSfrom E. coli (Sigma) or 1 × 106 live trypomastigotes. After24 h of culture, supernatants were harvested and stored at− 80 °C until used. IFN- + production was assayed by ELISAusing mouse IFN- + ELISA system (Amersham Pharmacia).

4.7 Statistical analysis

Significance of survival curves was determined by Kaplan-Meier survival analysis. Analysis of differences of other datawas performed by using Student’s t-test. A value of p X 0.05was considered significant.

Acknowledgements: This work was supported in part byGrant-in-Aid no. 11470064 from the Ministry of Education,Science, Sports and Culture, Japan. We thank Drs. R. Coff-man (DNAX Research Institute, Palo Alto, CA) and R. L. Tar-leton (University of Georgia, Athens, GA) for generously pro-viding RB6-8C5 hybridoma and multicompetitor pmCK3.1,respectively.

Eur. J. Immunol. 2001. 31: 265–275 Different role of neutrophils in experimental Chagas’ disease 273

References

1 Mosmann, T. R. and Coffman, R. L., Th1 and Th2 cells: differentpatterns of lymphokine secretion lead to different functionalproperties. Annu. Rev. Immunol. 1989. 7: 145–173.

2 Nogueira, N., Ellis, J., Chaplan, S. and Cohan, Z., Trypano-soma cruzi: in vivo and in vitro correlation betwen T-cell activationand susceptibility in inbred strains of mice. Exp. Parasitol 1981.51: 325–334.

3 DosReis, G. A., Cell-mediated immunity in experimental Trypa-nosoma cruzi infection. Parasitol. Today 1997. 13: 335–342.

4 Hoft, D. F., Lynch, R. G. and Kirchhoff, L. V., Kinetic analysis ofantigen-specific immune responses in resistance and suscepti-ble mice during infection with Trypanosoma cruzi. J. Immunol.1993. 151: 7038–7047.

5 Wrightsman, R. A., Krassner, S. M., Watson, J. D. and Man-ning, J. E., Role of the H-2s haplotype in survival of mice afterinfection with Trypanosoma cruzi. Infect. Immun. 1984. 44:351–354.

6 Kudo, C., Yamashita, T., Araki, A., Terashita, M., Watanabe, T.,Atsumi, M., Tamura, M. and Sendo, F., Modulation of in vivoimmune response by selective depletion of neutrophils using amonoclonal antibody, RP-3. I. Inhibition by RP-3 treatment of thepriming and effector phases of delayed-type hypersensitivity tosheep red blood cells in rats. J. Immunol. 1993. 150: 3728–3738.

7 Kudo, C., Yamashita, T., Terashita, M. and Sendo, F., Modula-tion of in vivo immune response by selective depletion of neutro-phils using a monoclonal antibody, RP-3. II. Inhibition by RP-3treatment of mononuclear leukocyte recruitment in delayed-typehypersensitivity to sheep red blood cells in rats. J. Immunol.1993. 150: 3739–3746.

8 Lloyd, A. R. and Oppenheim, J. J., Poly’s lament: the neglectedrole of the polymorphonuclear neutrophil in the afferent limb ofthe immune response. Immunol. Today 1992. 13: 169–172.

9 Cassatella, M. A., The production of cytokines by polymorpho-nuclear neutrophils. Immunol. Today 1995. 16: 21–26.

10 Hoft, D. F., Schnapp, A. R., Eickhoff, C. S. and Roodman, S. T.,Involvement of CD4+ Th1 cells in systematic immunity protectiveagainst primary and secondary challenges with Trypanosomacruzi. Infect. Immun. 2000. 68: 197–204.

11 Tarleton, R. L., Trypanosoma cruzi-induced suppression of IL-2production. I. Evidence for the presence of IL-2-producing cells.J. Immunol. 1988. 140: 2763–2768.

12 Silva, J. S., Morrissey, P. J., Grabstein, K. H., Mohler, K. M.,Anderson, D. and Reed, S. G., Interleukin 10 and interferongamma regulation of experimental Trypanosoma cruzi infection.J. Exp. Med. 1992. 175: 169–174.

13 Reed, S. G., Brownell, C. E., Russo, D. M., Silva, J. S., Grab-stein, K. H. and Morrissey, P., IL-10 mediates susceptibility toTrypanosoma cruzi infection. J. Immunol. 1994. 153: 3135–3140.

14 Aliberti, J. C., Cardoso, M. A., Martins, G. A., Gazzinelli, R. T.,Vieira, L. Q. and Silva, J. S., Interleukin-12 mediates resistanceto Trypanosoma cruzi in mice and is produced by murine macro-phages in response to live trypomastigotes. Infect. Immun. 1996.64: 1961–1967.

15 Torrico, F., Heremans, H., Rivera, M. T., Van, M. E., Billiau, A.and Carlier, Y., Endogenous IFN-gamma is required for resis-tance to acute Trypanosoma cruzi infection in mice. J. Immunol.1991. 146: 3626–3632.

16 Silva, J. S., Vespa, G. N., Cardoso, M. A., Aliberti, J. C. andCunha, F. Q., Tumor necrosis factor alpha mediates resistance toTrypanosoma cruzi infection in mice by inducing nitric oxide pro-

duction in infected gamma interferon-activated macrophages.Infect. Immun. 1995. 63: 4862–4867.

17 Gasperini, S., Marchi, M., Calzetti, F., Laudanna, C., Vicentini,L., Olsen, H., Murphy, M., Liao, F., Farber, J. and Cassatella,M. A., Gene expression and production of the monokine inducedby IFN-gamma (MIG), IFN-inducible T cell alpha chemoattractant(I-ITAC), and IFN-gamma-inducible protein-10 (IP-10) chemo-kines by human neutrophils. J. Immunol. 1999. 162: 4928–4937.

18 Kasama, T., Strieter, R. M., Standiford, T. J., Burdick, M. D.and Kunkel, S. L., Expression and regulation of humanneutrophil-derived macrophage inflammatory protein 1 alpha.J. Exp. Med. 1993. 178: 63–72.

19 Gonzalo, J. A., Pan, Y., Lloyd, C. M., Jia, G. Q., Yu, G., Dus-sault, B., Powers, C. A., Proudfoot, A. E., Coyle, A. J., Gearing,D. and Gutierrez, R. J., Mouse monocyte-derived chemokineis involved in airway hyperreactivity and lung inflammation.J. Immunol. 1999. 163: 403–411.

20 Bonecchi, R., Bianchi, G., Bordignon, P. P., D’Ambrosio, D.,Lang, R., Borsatti, A., Sozzani, S., Allavena, P., Gray, P. A.,Mantovani, A. and Sinigaglia, F., Differential expression of che-mokine receptors and chemotactic responsiveness of type 1 Thelper cells (Th1s) and Th2s. J. Exp. Med. 1998. 187: 129–134.

21 Siveke, J. T. and Hamann, A., T helper 1 and T helper 2 cellsrespond differentially to chemokines. J. Immunol. 1998. 160:550–554.

22 Puddu, P., Fantuzzi, L., Borghi, P., Varano, B., Rainaldi, G.,Guillemard, E., Malorni, W., Nicaise, P., Wolf, S. F., Belardelli,F. and Gessani, S., IL-12 induces IFN-gamma expression andsecretion in mouse peritoneal macrophages. J. Immunol. 1997.159: 3490–3497.

23 Yeaman, G. R., Collins, J. E., Currie, J. K., Guyre, P. M., Wira,C. R. and Fanger, M. W., IFN-gamma is produced by polymor-phonuclear neutrophils in human uterine endometrium andby cultured peripheral blood polymorphonuclear neutrophils.J. Immunol. 1998. 160: 5145–5153.

24 Choromanski, L. and Kuhn, R. E., Interleukin 2 enhances spe-cific and nonspecific immune responses in experimental Chagas’disease. Infect. Immun. 1985. 50: 354–357.

25 Sato, T., Midorikawa, Y., Yamashita, T., Araki, A. and Sendo,F., The neutrophil as an information transmitter in tumor inhibitionby a streptococcal biological response modifier, OK-432. CancerImmunol. Immunother. 1996. 43: 77–86.

26 Lichtenstein, A. K., Berek, J., Kahle, J. and Zighelboim, J.,Role of inflammatory neutrophils in antitumor effects induced byintraperitoneal administration of Corynebacterium parvum inmice. Cancer Res. 1984. 44: 5118–5123.

27 Bronte, V., Wang, M., Overwijk, W. W., Surman, D. R., Pericle,F., Rosenberg, S. A. and Restifo, N. P., Apototic death of CD8+

T lymphocytes after immunization: induction of a suppressivepopulation of Mac-1+/Gr-1+ cells. J. Immunol. 1998. 161:5313–5320.

28 Heinzel, F. P., Sadick, M. D., Holaday, B. J., Coffman, R. L. andLocksley, R. M., Reciprocal expression of interferon gamma orinterleukin 4 during the resolution or progression of murine leish-maniasis. Evidence for expansion of distinct helper T cell sub-sets. J. Exp. Med. 1989. 169: 59–72.

29 Hunter, C. A., Ellis, N. L., Slifer, T., Kanaly, S., Grunig, G., Fort,M., Rennick, D. and Araujo, F. G., IL-10 is required to preventimmune hyperactivity during infection with Trypanosoma cruzi.J. Immunol. 1997. 158: 3311–3316.

30 Zhang, L. and Tarleton, R. L., Characterization of cytokine pro-duction in murine Trypanosoma cruzi infection by in situ immuno-

274 L. Chen et al. Eur. J. Immunol. 2001. 31: 265–275

cytochemistry: lack of association between susceptibility andtype 2 production. Eur. J. Immunol. 1996. 26: 102–109.

31 Launois, P., Swihart, K. G., Milon, G. and Louis, J. A., Early pro-duction of IL-4 in susceptible mice infected with Leishmaniamajor rapidly induces IL-12 unresponsiveness. J. Immunol. 1997.158: 3317–3324.

32 Launois, P., Ohteki, T., Swihart, K., MacDonald, H. R. andLouis, J. A., In susceptible mice Leishmania major induce veryrapid interleukin-4 production by CD4+ T cells which are NK1.1–.Eur. J. Immunol. 1995. 25: 3298–3307.

33 Assenmacher, M., Schmitz, J. and Radbruch, A., Flow cyto-metric determination of cytokines in activated murine T helperlymphocytes: expression of interleukin-10 in interferon-gammaand in interleukin-4-expressing cells. Eur. J. Immunol. 1994. 24:1097–1101.

34 Del, P. G., De, C. M., Almerigogna, F., Giudizi, M. G., Biagiotti,R. and Romagnani, S., Human IL-10 is produced by both type 1helper (Th1) and type 2 helper (Th2) T cells clones and inhibitstheir antigen-specific proliferation and cytokine production.J. Immunol. 1993. 150: 353–360.

35 Fleming, T. J., Fleming, M. L. and Malek, T. R., Selectiveexpression of Ly-6G on myeloid lineage cells in mouse bone mar-row. RB6-8C5 mAb to granulocyte-differentiation antigen (Gr-1)detects members of the Ly-6 family. J. Immunol. 1993. 151:2399–2408.

36 Jensen, J., Warner, T. and Balish, E., Resistance of SCID miceto Candida albicans administered intravenously or colonizing thegut: role of polymorphonuclear leukocytes and macrophages.J. Infect. Dis. 1993. 167: 912–919.

37 Chen, L., Zhang, Z.-H. and Sendo, F., Neutrophils play a criticalrole in the pathogenesis of experimental cerebral malaria. Clin.Exp. Immunol. 2000. 120: 125–133.

38 Hamasaki, A., Sendo, F., Nakayama, K., Ishida, N., Negishi, I.,Nakayama, K. and Hatakeyama, S., Accelerated neutrophilapoptosis in mice lacking A1-a, a subtype of the bcl-2-related A1gene. J. Exp. Med. 1998. 188: 1985–1992.

39 Kanegane, C., Sgadari, C., Kanegane, H., Teruya, F. J., Yao, L.,Gupta, G., Farber, J. M., Liao, F., Liu, L. and Tosato, G., Contri-bution of the CXC chemokines IP-10 and Mig to the antitumoreffects of IL-12. J. Leukoc. Biol. 1998. 64: 384–392.

40 Widmer, U., Yang, Z., van, D. S., Manogue, K. R., Sherry, B.and Cerami, A., Genomic structure of murine macrophageinflammatory protein-1 alpha and conservation of potential regu-latory sequences with a human homolog, LD78. J. Immunol.1991. 146: 4031–4040.

Correspondence: Fujiro Sendo, Department of Immunol-ogy and Parasitology, Yamagata University School of Medi-cine, 2-2-2, Iida-Nishi, Yamagata 990-9585, JapanFax: +81-23-628-5267e-mail: fsendo — med.id.yamagata-u.ac.jp

Eur. J. Immunol. 2001. 31: 265–275 Different role of neutrophils in experimental Chagas’ disease 275