Highly up-regulated CXCR3 expression on eosinophils in mice infected with Schistosoma japonicum

Abstract

CXCR3, predominately expressed on memory/activated T cells, is a receptor for both interferon-γ inducible protein-10/CXC ligand 10 (CXCL10) and monokine induced by interferon-γ/CXCL9. We reported here that CXCR3 was highly up-regulated on infiltrating eosinophils in Schistosoma japonicum egg-induced granuloma in the mouse liver. It was also highly and functionally up-regulated on peritoneal exudate eosinophils in mice infected with S. japonicum. The phenomena were demonstrated at protein and mRNA levels using immunohisto- and immunocytochemistry evaluation of biopsy, flow cytometry and real-time quantitative reverse transcriptase–polymerase chain reaction technique, and verified by Northern blotting and chemotaxis assay in vitro. We also found that CCR3 expression on the infiltrating and peritoneal exudate cells was significantly decreased, CXCR4 expression was unchanged during the 42-day period of infection. We screened mRNA expression levels of the all known chemokine receptors in purified peritoneal exudate eosinophils and liver granuloma dominated by eosinophils. CXCR3 was highly and functionally up-regulated on peritoneal exudate eosinophils in mice infected with S. japonicum, meanwhile CCR3 was significantly and functionally down-regulated in these cells. The findings could lead to a better understanding of the chemokine receptor expression pattern of eosinophils at inflamed tissue sites caused by parasites. These could be also crucial for establishing a therapeutic strategy for eosinophilic inflammation via intervention in chemokine actions.

Introduction

Schistosoma japonicum is a major schistosome species in Asia, infecting not only humans but also wild or domestic animals. Despite the availability of very successful control programmes, schistosomiasis japonica remains a serious public health problem in endemic regions in South-east Asia.¹ Many parasites, especially tissue-invasive helminths, cause local eosinophilia as well as systemic eosinophilia in mammalian hosts. Hypersensitivity reactions in the host to the parasite's eggs lead to the formation of eosinophil-rich granulomatous lesions in the intestine and liver, which is the pathological hallmark of intestinal schistosomiasis. The mechanisms by which eosinophils are recruited to local inflammatory sites via the circulation have been elucidated at least in part using eosinophil chemotactic factors (ECFs).^2,3 The infiltrating eosinophils have been linked to T helper 2 (Th2)-associated interleukin (IL)-5 production, promoting eosinophil differentiation. More specific support for a Th2-dominated response comes from a report showing elevated mRNA-expression for IL-4 and IL-10, but not for interferon-γ (IFN-γ), in intestine and liver of S. japonicum-infected animals.⁴ Although many components and regulatory features of the eosinophil recruitment pathway have been elucidated, the molecular basis for the massive infiltration of eosinophils during helminth infection is not fully understood.^5–8 Further, the role of eosinophils in immunity to helminths remains the subject of considerable debate.⁹

Evidence from different groups has indicated that signals via chemokine receptors play an essential role in the pathogenesis of eosinophilic inflammation. CCR3 is strongly expressed on eosinophils¹⁰ and is responsible for both migration and degranulation.^11–13 Eotaxin/CCL11 represents the most potent chemoattractant for eosinophils.^14,15 CXCR4 expression is inducible on eosinophils under certain circumstances. Stromal cell-derived factor-1 (SDF-1)/CXC ligand 12 (CXCL12), a CXCR4-specific ligand, induces strong migration comparable to that induced by eotaxin/CCL11.^10,16 We have demonstrated the expression of functional CXCR3 on human eosinophils.¹⁷ CCR1 expression is observed to a lesser extent than CCR3 expression, and its ligand macrophage inflammatory protein 1α (MIP-1α)/CCL3 induces eosinophil chemotaxis in a small and selected subset of atopic patients.¹⁸ The observations of expression of CXCR1 and CXCR2 have also been reported.^19,20 The appearance of CCR2, CCR4, and CCR5 on murine eosinophils, which leads to the cells migrating into the peritoneal cavity has been reported by Lukacs et al.²¹ Interestingly, mouse eosinophils isolated from IL-5 transgenic mice express transcripts encoding the chemokine receptors CCR1, CCR2, CCR3, CCR5, CCR8, CXCR2, and CXCR4, but not CCR4.²² Though expression profiles of eosinophil chemokine receptors have become increasingly understood, most studies have used eosinophils isolated from peripheral blood. As eosinophils move from the vascular compartment to inflamed tissue sites, microenvironmental signals may affect the receptor expression. Eosinophils that have migrated into inflamed tissue sites may exhibit a pattern of receptor expression that is distinct from that of circulating eosinophils.²³

In the present study, we have found that CXCR3 is highly and functionally up-regulated on infiltrating eosinophils in S. japonicum egg-induced granuloma in the mouse liver as well as on peritoneal exudate eosinophils in mice infected S. japonicum, demonstrated at protein and mRNA levels in vitro. Meanwhile CCR3 is significantly and functionally down-regulated in these cells. The findings can lead to a better understanding of the chemokine receptor expression pattern of eosinophils at inflamed tissue sites caused by parasites.

Materials and methods

Reagents and antibodies

All recombinant mouse chemokines eotaxin/CCL11, MIP-1α/CCL3, MIP-1β/CCL4, RANTES (regulated on activation, normal, T-cell expressed, and secreted)/CCL5, monokine induced by interferon-γ (Mig)/CXCL9 and interferon-γ inducible protein-10 (γ IP-10)/CXCL10 were purchased from R & D Systems Europe Ltd (Abingdon, UK). Fluorescein isothiocyanate (FITC)-labelled anti-CD15 was purchased from DAKO (Glostrup, Denmark). Phycoerythrin (PE)-labelled rat anti-mouse monoclonal CCR3 (clone 83101.111) and goat anti-mouse polyclonal CXCR3 (clone SC-6226) were purchased from R & D Systems Europe Ltd and Santa Cruz Biotechnology Inc. (Santa Cruz, CA), respectively.

Animals and S. japonicum infection

C57BL/6 (H-2b) mice used in this study were obtained from Provincial Experimental Animal Centre, Wuhan, Hubei, P. R. China. Adult females of 8–12 week of age were used throughout this study. C57BL/6 wild-type mice were used as uninfected controls. The mice were inoculated by percutaneous injection with Iscove's medium-suspended cercariae (30 per exposure) of a Chinese mainland strain of S. japonicum originating from Hubei Province and maintained in Oncomelania hupensis hupensis snails at the Department of Parasitology, Medical College, Wuhan University, Wuhan. The mice were then sacrificed at different stages of the inflammatory process, usually at day 14 (early stage), or at day 42 (late stage), or at time intervals as indicated.

Histological evaluation of liver biopsy

The livers were harvested from killed mice at time intervals as indicated for histological analysis. For routine histological examination (HE staining), liver samples were fixed with 10% formalin and paraffin-embedded sections (4 µm thickness) were stained with haematoxylin and eosin for 3 min each. For immunohistology, a portion of the liver tissue was placed in OCT compound and frozen in liquid nitrogen for immunohistological studies. Cryostat sections (6–8 µm thickness) were prepared on slides coated with poly l-lysine, and fixed in cold acetone for 10 min, and air-dried overnight, immersed in phosphate-buffered saline (PBS) for 10 min and then in 0.03% H₂O₂ for 8 min to eliminate endogenous peroxidase activity. The antibodies were diluted to 5 µg/ml in 0.05% Tris-HCl buffer with 2% bovine serum albumin (BSA). The slides were then stained with the primary antibody for 4 hr at room temperature. Control slides were incubated with appropriate isotype antibodies as the primary antibody. After three washes in PBS for 5 min each, slides were incubated for 20 min at room temperature with an appropriate biotinylated secondary antibody diluted 1/200 in PBS, followed with three washes in PBS, then incubated with streptavidin–horseradish peroxidase (HRP; DAKO) for 30 min at room temperature. The freshly prepared substrate–chromagen solution added 12 µl 30% H₂O₂ was applied to each slide and incubated for about 5 min at room temperature. After a final wash in water, slides were counterstained with haematoxylin, rinsed, and immersed in 37 nm NH₄OH for 10 s. All slides were viewed and evaluated in a blinded fashion by a qualified pathologist.

Real-time quantitative reverse transcription (RT)–polymerase chain reaction (PCR) assay

All real-time quantitative RT–PCR reactions were performed as described elsewhere.^23–26 Briefly, total RNA was purified from liver tissues or purified peritoneal exudate eosinophils (5 × 10³, purity >93%). Total RNA was prepared by using Quick Prep® total RNA extraction kit (Pharmacia Biotech, Piscataway, NJ). The RNA was reverse transcribed by using oligo (dT)₁₂₋₁₈ and Superscript II reverse transcriptase (Life Technologies, Grand Island, NY). RT was performed for 60 min at 37°, and any potential contaminating protein was denatured by incubation for 10 min at 95°. The real time quantitative PCR was performed in special optical tubes in a 96 well microtitre plate (Applied Biosystems, Foster City, CA) with an ABI PRISM® 7700 Sequence Detector Systems (Applied Biosystems). By using SYBR® Green PCR Core Reagents Kit (Applied Biosystems, P/N 4304886), fluorescence signals were generated during each PCR cycle via the 5′ to 3′ endonuclease activity of AmpliTaq Gold²³ to provide real-time quantitative PCR information. The sequences of the specific primers generated for target genes are shown in Table 1.

Table 1.

The sequences of senses and antisense for mRNA detection of chemokine receptors and relevant ligands in real-time quantitative RT–PCR assay

Target	Sequence of sense	Sequence of antisense
CCR1	5′-AGGCCCAGTGGGAGTTCAC-3′	5′-TCCACTGCTTCAGGCTCTTGT-3′
CCR2	5′-TCATCCACGGCATACTATCAACA-3′	5′-GTGGCCCCTTCATCAAGCT-3′
CCR3	5′-GCTTTGAGACCACACCCTATGAA-3′	5′-GACCCCAGCTCTTTGATTCTGA-3′
CCR4	5′-AGAAAATTCATGCGAGGGAGAA-3′	5′-TGTCACTCAAGGGTCGTGTCA-3′
CCR5	: 5′-AGGCCATGCAGGCAACAG-3′	5′-TCTCTCCAACAAAGGCATAGATGA-3′
CCR6	5′-GAACTGCCCACTTCCCTTTCT-3′	5′-CAGGCTGGCGTGGTTCTCT-3′
CCR7	5′-GCTCCAGGCACGCAACTTT-3′	5′-GACTACCACCACGGCAATGA-3′
CCR8	5′-TGACCGACTACTACCCTGATTTCTT-3′	5′-GCTGCCCCTGAGGAGGAA-3′
CCR9	5′-TCTCAGTTCCCCTACAACTCCATT-3′	5′-CAGTTGGAGATGAACATGGCATA-3′
CXCR1	5′-CCCTCTTTAAGGCCCACATG-3′	5′-AAGGACGACAGCGAAGATGAC-3′
CXCR2	5′-ACCCTCTTTAAGGCCCACATG-3′	5′-CAAGGACGACAGCGAAGATG-3′
CXCR3	5′-CAGCCTGAACTTTGACAGAACCT-3′	5′-GCAGCCCCAGCAAGAAGA-3′
CXCR4	5′-TCAGCCTGGACCGGTACCT-3′	5′-GCAGTTTCCTTGGCCTTTGA-3′
CXCR5	5′-GCCTGCTCGTGGCCTGTA-3′	5′-CGGCGGTAGGCGTGAAC-3′
CX3CR1	5′-TCAGCATCGACCGGTACCTT-3′	5′-CTGCACTGTCCGGTTGTTCA-3′
XCR1	5′-GCCTTCTCTCATTGCTGTTTCA-3′	5′-TTTAGGTGTCTGCGGAACTTGA-3′
Eotaxin/CCL11	5′-GGCTGACCTCAAACTCACAGAAA-3′	5′-ACATTCTGGCTTGGCATGGT-3′
MIP-1α/CCL3	5′-GATCTGCGCTGACTCCAAAGA-3′	5′-CCAAGACTCTCAGGCATTCAGTT-3′
MIP-1β/CCL4	5′-TGCTCGTGGCTGCCTTCT-3′	5′-CAGGAAGTGGGAGGGTCAGA-3′
RANTES/CCL5	5′-GCAAGTGCTCCAATCTTGCA-3′	5′-GATGTATTCTTGAACCCACTTCTTCTC-3′
γIP-10/CXCL10	5′-GGACGGTCCGCTGCAA-3′	5′-GCTTCCCTATGGCCCTCATT-3′
Mig/CXCL9	5′-TTTTCCTTTTGGGCATCATCTT-3′	5′-AGCATCGTGCATTCCTTATCACT-3′

All unknown cDNAs were diluted to contain equal amounts of β-actin cDNA. The standards, ‘no template’ controls and unknown samples were added in a total volume of 50 µl per reaction. PCR retain conditions were 2 min at 50°, 10 min at 95°, 40 cycles with 15 s at 95°, 60 s at 60°. Potential PCR product contamination was digested by uracil-N-glycosylase (UNG) where dTTP is substituted by dUTP.²³ In the reaction system, UNG and AmpliTaq Gold (Applied Biosystems) were applied according to the manufacturer's instructions.^23,24

Northern blot assay

For mRNA detection, 5 µg of pooled total RNA samples were electrophoresed under denaturing conditions, followed by blotting onto Nytran membranes, and cross-linked by UV irradiation as previously described.²⁷ CCR3 and CXCR3 cDNA probes, labelled by α-[³²P]dCTP, was obtained by PCR amplification of the sequence mentioned above from total RNA from total spleen cells from normal adult mice. The membranes were hybridized overnight with 1 × 10⁶ c.p.m./ml of ³²P-labelled probe, followed by intensively washing with 0.2 × standard saline citrate (SSC) (1 × SSC = 0.15 m NaCl, 0.015 m sodium citrate, pH 7.0) and 0.1% sodium dodecyl sulphate (SDS) before being autoradiographed.

Mouse peritoneal and peripheral eosinophil purification

As previously described with modification²⁸ animals were injected intraperitoneally with 3 ml of thioglycolate (0.5%). Mouse peritoneal exudate cells were obtained at day 4 after thioglycolate injection from lavage of the peritoneal cavity with PBS and were used for eosinophil purification. Eosinophils were obtained by removal of adherent cells (mainly macrophages) after overnight culture in complete RPMI medium. The purities of eosinophils were 60–95% estimated after May–Grünwald staining. For mRNA measurement, neutrophils and remain macrophages were further depleted by the magnetic-activated cell separation system (MACS) using anti-CD16-coated MACS particles (Miltenyi Biotech, Bergisch Gladbach, Germany), so that the purity of eosinophils could be ≥93%.²⁹ The eosinophils from peripheral blood were purified by negative selection procedure as described previously.¹⁷ In most situations, the blood samples from mice were pooled because the volume of blood from each mouse was too limited to carry out following procedures.

Flow cytometry

As previously described³⁰ for detection of CCR3 or CXCR3, peritoneal exudate eosinophils were first incubated with a rat anti-mouse CCR3 PE-labelled monoclonal antibody (mAb) or goat anti-mouse polyclonal antibody (pAb) CXCR3 at 5 µg/ml or 5 µg/ml isotype-matched goat antibody (DAKO) in PBS containing 2% BSA and 0.1% sodium azide for 20 min, followed washing twice in staining buffer. For CXCR3 detection, a secondary swine anti-goat immunoglobulin G (IgG) PE-labelled mAb (DAKO) was subsequently added. The cells were then incubated with a goat anti-mouse CD3 (DAKO), rat anti-mouse CD15 FITC-labelled mAb (BD PharMingen Europe, Heidelberg, Germany) at 5 µg/ml in PBS containing 2% BSA and 0.1% sodium azide for 20 min, followed by washing twice. All procedures were carried out at 4°. The cells were then fixed with 1% paraformaldehyde. The analyses were performed with a flow cytometer (Coulter® XL, Coulter Corporation, Miami, FL).

Immunocytochemistry assay

For detection of chemokine receptors on eosinophils²⁹ the purified cells were spun down on a glass slide at 18 g for 4 min Then the slide was moved into a fixation dish immediately to avoid drying out. The fixation liquid was a mixture of methanol and acetone (1/1, v/v). After 5-min fixation the preparation was washed twice in PBS for 5 min Blocking buffer (PBS with 1% BSA and 0.3% Triton-X-100) was added for 5 min at 20° to avoid unspecific binding, followed by primary antibody (CXCR3 pAb or CCR3 mAb) at concentration of 10 µg/ml. The preparation was incubated overnight at 4°. The next day the preparation was washed twice in PBS for 5 min each time, followed by addition of secondary antibody and was visualized using the alkaline phosphatase staining system (Dako) according to the manufacturer's instructions. Finally, the preparation was sealed and stored in the dark until observation under a microscope.

Chemotaxis assay

The chemotaxis assay was performed in a 48-well microchamber (Neuro Probe, Bethesda, MD) technique.³¹ Briefly, chemokines were diluted in RPMI-1640 with 1% fetal calf serum (FCS) and placed in the lower wells (25 µl). Twenty-five ml of the cell suspension (freshly purified peritoneal exudate eosinophils) at 1 × 10⁴ cells/ml was added to the upper well of the chamber, which was separated from the lower well by a 5-µm pore-size, polycarbonate, polyvinylpyrolidone-free membrane (Nucleopore, Pleasanton, CA). The chamber was incubated for 45 min at 37° in an atmosphere containing 5% CO₂. The membrane was then carefully removed, fixed in 70% methanol and stained for 5 min in 1% Coomassie Brilliant Blue. The cells that migrated and adhered to the lower surface of the membrane were counted by using a light microscopy. Approximately 10% of the cells will migrate spontaneously (known as migrating cells on negative control, MCNC)³² corresponding to between 200 and 300 cells. The results were expressed as chemotactic index (CI), that is the ratios between the numbers of migrating cells in the sample and in the medium control^31,33 plus or minus standard deviation (SD).

Statistical analysis

For pair-wise comparisons, the groups were analysed using a t-test, while for multiple comparisons a one-way anova with Dunnett's method setting the saline group as control was employed. In both cases, P < 0.05 was considered significant.

Results

CXCR3 was highly up-regulated on infiltrating eosinophils in S. japonicum egg-induced granuloma in the mouse liver

We analysed the some chemokine receptor expression on the infiltrating cells around the deposited eggs in hepatic granulomas at different time intervals post infection of S. japonicum. At day 14 postinfection (the early stage), there were almost no hepatic granulomas and few infiltrating cells were neutrophils showing CCR3 positive and CXCR3 negative (data not shown). At day 30 postinfection, there were increased numbers of infiltrating cells in hepatic granulomas, the majority of infiltrating cells were morphologically recognized as eosinophils (determined by eosin Y staining). These cells were expressing strongly CCR3 positive and significant increased CXCR3 positive (Fig. 1b, d). At day 42 postinfection (the late stage), there were heavily increased numbers of infiltrating eosinophils in hepatic granulomas (Fig. 1a). CCR3 expression on the infiltrating cells was significantly decreased, but almost all infiltrating eosinophils were CXCR3 positive (Fig. 1c, e).

Figure 1.

Histological analyses of liver from C57BL/6 (H-2b) mice infected with S. japonicum. Samples were harvested at day 30 (b and d) or day 42 (a, c and e) postinfection, and fixed with 10% buffered formalin. Section was prepared and stained by HE staining (a) as described in Materials and Methods. Frozen sections were prepared and stained with anti-CCR3 mAb (b and c) or isotype antibody controls (figures not shown), or anti-CXCR3 pAb (d and f) or isotype antibody controls (Ba, Ca, Da and Ea) as described in Materials and Methods. The positively staining cells in representative areas of the slides are shown in brown colour. Arrow, S. japonicum egg. Magnification: ×200.

To confirm the observation mentioned above, we had purified total RNA from the liver sections either from uninfected mice or infected with S. japonicum for time intervals, and measured CCR3 and CXCR3 mRNAs by a real-time quantitative RT–PCR technique. CCR3 mRNA was detected in liver sample of uninfected mice (approximately 4.2 × 10³ copies/50 ng cDNA). After infection with S. japonicum CCR3 mRNA expression significantly decreased in the liver samples. By day 7, day 14 and day 28, there were approximately 4.3 × 10³ copies, 1.7 × 10³ copies and 1.2 × 10³ copies/50 ng cDNA, respectively. At day 42, there were approximately 1.1 × 10³ copies/50 ng cDNA (Fig. 2a). CXCR3 mRNA expression after infection with S. japonicum was significantly increased. CXCR3 mRNA was expressed at low level in liver sample of uninfected mice (approximately 0.8 × 10³ copies/50 ng cDNA). By day 7, day 14 and day 28, there were approximately 2.3 × 10³ copies, 5.7 × 10³ copies and 6.4 × 10³ copies/50 ng cDNA in the cells, respectively. At day 42, there were approximately 8.3 × 10³ copies/50 ng cDNA (Fig. 2b). The changes of CCR3 and CXCR3 mRNA expressions had also been confirmed by Northern blot. In upper panels, decreasing CCR3 (Fig. 2c) and increasing CXCR3 (Fig. 2d) expressions of mRNA were observed by day 7, day 14, day 28 and day 42 after infection, respectively. The lower panels were showing that comparable total RNA amounts from different days were added. We had also measured other chemokine receptor expression by real-time quantitative RT–PCR. As shown in Table 1, except for CXCR4, the expression levels of CC, CXC, CX₃C and XC chemokine receptors were not significantly changed during the infection with S. japonicum. CXCR4 expression was significantly up-regulated postinfection. Because the liver samples contain different type of cells, we later examined mRNA expression levels of the different chemokine receptors in purified peritoneal exudate eosinophils during the infection (Table 2, details shown below). To further understand the mechanism of eosinophil migration towards hepatic granulomas induced by egg of S. japonicum, we also examined the expression levels of different chemokine ligands. The expression levels of the several CC ligands examined (eotaxin/CCL11, RANTES/CCL5, MIP-1α/CCL3, MIP-1β/CCL4) were not significantly changed in liver samples of mice at different time intervals (day 7, day 10, day 14, day 30 and day 42) during the infection with S. japonicum. There were 1.3 × 10³ and 1.6 × 10³ copies/50 ng cDNA of γ IP-10/CXCL10 and Mig/CXCL9, respectively, in uninfected liver. We detected transient increased expression peaks of CXC chemokine γ IP-10/CXCL10 and Mig/CXCL9 mRNA at day 10 postinfection. The mRNA expression levels of γ IP-10/CXCL10 and Mig/CXCL9 reached 5.2 × 10³ and 4.6 × 10³ copies/50 ng cDNA, respectively. The expression levels of γ IP-10/CXCL10 and Mig/CXCL9 at day 7, day 14, day 30 and day 42 were similar to that in uninfected mice. The results were indicating that S. japonicum infection also elevated the expression of some CXCLs for CXCR3 in lever. The eosinophils highly up-regulated CXCR3 in later stag of the infection would be allowed to migrate into liver to form eosinophil hepatic granuloma.

Figure 2.

The graphs of the real time detection and amplification of mRNA of CCR3 (a) or CXCR3 (b) in either the healthy or egg granuloma formed livers. The samples were liver section (1.5 g) either from native C57BL/6 (H-2b) mice or C57BL/6 (H-2b) infected with S. japonicum for the time intervals shown. The procedures for total RNA isolation were described in Materials and Methods. All * are indicating significant changes (all P < 0.05). A linear relationship between C_T and log starting quantity of standard DNA template or target cDNA (CCR3 or CXCR3) was detected (data not shown). The correlation coefficients are approximately 0.96–0.99. The showing graphs are representatives of six similar experiments conducted. CCR3 (c) and CXCR3 (d) mRNA Northern blots of the samples were liver sections (2.0 g) either from native BALB/c (H-2^d) mice or BALB/c (H-2^d) infected with S. japonicum for the time intervals shown. Total RNA were isolated, electrophoresed and blotted as described in Materials and Methods. The hybridization signals for CCR3 or CXCR3 mRNA from different tissues were shown in upper panels. The 28S rRNAs in lower panels confirm comparable amounts of loaded total RNA.

Table 2.

The real-time detection and amplification of mRNA of chemokine receptors in liver samples and peritoneal exudate eosinophils

Chemokine R*	Healthy†	Infected†	Healthy‡	Infected‡
CCR1§	5.8 × 103¶	6.1 × 103	4.5 × 103	4.8 × 103
CCR2	4.3 × 102	5.1 × 102	1.1 × 102	1.0 × 102
CCR4	7.3 × 102	6.5 × 102	–**	–
CCR5	4.5 × 102	6.1 × 102	1.2 × 102	1.0 × 102
CCR6	ND	ND	–	–
CCR7	3.5 × 102	4.1 × 102	–	0.5 × 102
CCR8	2.3 × 102	3.0 × 102	–	0.1 × 102
CCR9	1.5 × 102	2.2 × 102	–	0.2 × 102
CXCR1	4.4 × 103	3.8 × 103	2.5 × 103	2.2 × 103
CXCR2	2.7 × 103	3.1 × 103	4.8 × 103	3.7 × 103
CXCR4	1.1 × 103	7.8 × 103***	2.2 × 103	6.9 × 103***
CXCR5	–	–	–	–
CX3CR1	ND	ND	–	–
XCR1	–	–	–	–

^*CC, CXC, CX3C and XC chemokine receptors examined were listed (except CCR3 and CXCR3 shown elsewhere).

^†Liver samples from either healthy native C57BL/6 (H-2b) mice or C57BL/6 (H-2b) infected with S. japonicum for 42 days.

^‡Peritoneal exudate eosinophils from either healthy native C57BL/6 (H-2b) mice or C57BL/6 (H-2b) infected with S. japonicum for 42 days.

^§Procedures for total RNA isolation and real time quantitative RT–PCR were described in Materials and Methods. The data shown are representative of six similar experiments conducted.

^¶mRNA of chemokine receptors was expressed as copies per 50 ng cDNA.

^**Under detectable level.

ND, no determination.

^***P < 0.005. The data are representative of at least four similar experiments conducted for each chemokine receptor.

Thus, highly up-regulated CXCR3 expression was crucial important to cause eosinophil infiltration in S. japonicum egg-induced granuloma in the mouse liver.

CXCR3 was also highly and functionally up-regulated on peritoneal exudate eosinophils in mice infected with S. japonicum

To further investigate the importance of chemokine receptors in the migration and infiltration of eosinophils during the infection with S. japonicum in mice, we purified the peritoneal exudate eosinophils, and examined the expression of chemokine receptors by flow cytometry and real time quantitative RT–PCR. There was a large number of CCR3⁺ cell fractions on freshly isolated peritoneal exudate eosinophils from uninfected mice (more than 88%) (Fig. 3a). The CCR3 expression was decreasing during the infection. There was 67% CCR3⁺ cell fraction by day 7 postinfection, 42% by day 14, 40% by day 28 and 36% by day 42, respectively. CXCR3 expression was significantly increased on peritoneal exudate eosinophils (Fig. 3b). There were less than 5% CXCR3⁺ cell fractions on freshly isolated peritoneal exudate eosinophils from uninfected mice, 28% by day 7 postinfection, 64% by day 14, 90% by day 28 and 91% by day 42, respectively. To be sure that the observations were not affected by the stimulation of thioglycolate during the experimental procedure, we therefore purified eosinophils from peripheral blood from mice with or without infection of S. japonicum. There was large number of CCR3⁺ cell fractions on freshly isolated peripheral blood eosinophils from uninfected mice (more than 90%) (Fig. 4a). There was 74% CCR3⁺ cell fraction by day 7 postinfection, 38% by day 14, 30% by day 28 and 27% by day 42, respectively. CXCR3 expression was significantly increased on peripheral blood eosinophils during the infection (Fig. 4b). There were less than 6% CXCR3⁺ cell fractions on freshly isolated peripheral blood eosinophils from uninfected mice, 25% by day 7 postinfection, 67% by day 14, 92% by day 28 and 93% by day 42, respectively. Thus, the pattern of expression of CCR3 and CXCR3 was very similar to that of peritoneal exudate eosinophils. To confirm that expression of CXCR3 was indeed on the eosinophils, we therefore conducted an immunocytochemical assay on purified peritoneal exudate eosinophils to demonstrate the existence of CCR3 and CXCR3 on eosinophils. Because of autofluorescence in eosinophils, we chose an alkaline phosphatase staining system for visualization in the immunocytochemical assay to be absolutely sure of the observed positive results. The results from the immunocytochemical assay documented that CCR3 was expressed on the mouse peritoneal exudate eosinophils (Fig. 5b). The expression level in mice seems to decrease during infection with S. japonicum (Fig. 5c, d). CXCR3 was expressed very rarely on peritoneal exudate eosinophils in uninfected mice (Fig. 5f). The expression level was significantly decreasing during the mice infection with S. japonicum (Fig. 5g, h). Figure 5(a and b) showed the isotype antibody-negative controls.

Figure 3.

Double colour flow cytometric analysis of the distribution and modulation of CCR3 (a) or CXCR3 (b) on the peritoneal exudate eosinophils. The peritoneal exudate eosinophils were either from native C57BL/6 (H-2b) mice or C57BL/6 (H-2b) infected with S. japonicum for the time intervals shown. The graphs show CCR3 or CXCR3 single histograms after forward or side scatter gating within eosinophils. The dashed lines in graphs are isotype controls. The procedure for cell staining was described in Materials and Methods. The percentages of CCR3⁺ or CXCR3⁺ cells are indicated in Results. Data are from a single experiment, which is representative of six similar experiments performed.

Figure 4.

Double colour flow cytometric analysis of the distribution and modulation of CCR3 (a) or CXCR3 (b) on the peripheral blood eosinophils. The peripheral blood eosinophils were either from native C57BL/6 (H-2b) mice or C57BL/6 (H-2b) infected with S. japonicum for the time intervals shown. The graphs show CCR3 or CXCR3 single histograms after forward or side scatter gating within eosinophils. The dashed lines in graphs are isotype controls. The procedure for cell staining was described in Materials and Methods. The percentages of CCR3⁺ or CXCR3⁺ cells are indicated in the Results. Data are from a single experiment, which is representative of five similar experiments performed.

Figure 5.

Expression of CCR3 (b, c and d) and CXCR3 (f, g and h) on mouse peritoneal exudate eosinophils at day 42 postinfection. The cells were purified, fixed with a mixture of methanol and acetone (1/1, v/v), and immuno-stained as described in Materials and Methods. Immunoreactive cells were visualized using an alkaline phosphatase staining system. (a and e), cells were stained with a negative control isotype antibody. Cells were photographed under ×400 magnification (a–d) or ×200 magnification (e–h). Bars ∼10 µm.

To confirm the observation mentioned above, we purified total RNA from peritoneal exudate eosinophils either from uninfected mice or infected with S. japonicum at different time intervals, and measured CCR3 and CXCR3 mRNAs by real-time quantitative RT–PCR technique. CCR3 mRNA was detected in eosinophils of uninfected mice (approximately 3.8 × 10³ copies/50 ng cDNA), and 4.8 × 10³ copies by day 7 postinfection with S. japonicum. After longer infection with S. japonicum CCR3 mRNA expression significantly decreased in the cell samples. By day 14, day 28 and day 42, there were approximately 2.0 × 10³ copies, 1.5 × 10³ copies and 1.0 × 10³ copies/50 ng cDNA, respectively (Fig. 5a). CXCR3 mRNA expression after infection with S. japonicum was significantly increased. CXCR3 mRNA was expressed at a low level in peritoneal exudate eosinophils of uninfected mice (approximately 0.5 × 10³ copies/50 ng cDNA). By day 7, day 14 and day 28, there were approximately 2.1 × 10³ copies, 6.0 × 10³ copies and 6.5 × 10³ copies/50 ng cDNA in the cells, respectively. At day 42, there were approximately 8.5 × 10³ copies/50 ng cDNA (Fig. 6b). The changes of CCR3 and CXCR3 mRNA expressions had also been confirmed by Northern blot. In upper panels, decreased CCR3 (Fig. 6c) and increased CXCR3 (Fig. 6d) expressions of mRNA were observed by day 7, day 14, day 28 and day 42 after infection, respectively. The lower panels of Fig. 6 show that comparable total RNA amounts from different days were added. We also measured expression of other chemokine receptors in peritoneal exudate eosinophils by real-time quantitative RT–PCR. As shown in Table 2, except for CXCR4, the expression levels of CC, CXC, CX₃C and XC chemokine receptors were mostly very low; there were no significant changes during the infection with S. japonicum. CXCR4 expression was significantly up-regulated postinfection. We also examined the expression levels of different chemokine ligands. The expression levels of several CC and CXC chemokine ligands examined were not significantly changed in peritoneal exudate eosinophils of mice during the infection with S. japonicum (data not shown).

Figure 6.

Graphs of real-time detection and amplification of mRNA of CCR3 (a) or CXCR3 (b) in peritoneal exudate eosinophils from either healthy native C57BL/6 (H-2b) mice or C57BL/6 (H-2b) infected with S. japonicum for time intervals as indicated. The procedures for total RNA isolation were described in Materials and Methods. All * indicate significant changes (all P < 0.05). A linear relationship between C_T and log starting quantity of standard DNA template or target cDNA (CCR3 or CXCR3) was detected (data not shown). The correlation coefficients are approximately 0.96–0.99. The graphs are representative of six similar experiments conducted. CCR3 (c) and CXCR3 (d) mRNA Northern blots of the samples were peritoneal exudate eosinophils either from native C57BL/6 (H-2b) mice or C57BL/6 (H-2b) infected with S. japonicum for time intervals as indicated. Total RNA were isolated, electrophoresed and blotted as described in Materials and Methods. The hybridization signals for CCR3 or CXCR3 mRNA from different tissues are shown in the upper panels. The 28S rRNAs in the lower panels confirm comparable amounts of loaded total RNA.

We examined the abilities of eotaxin/CCL11 (a ligand for CCR3), γ IP-10/CXCL10 and Mig/CXCL9 (ligands for CXCR3) to induce peritoneal exudate eosinophil chemotaxis after infection with S. japonicum, because decreased CCR3 and increased CXCR3 expressions were found in these cells during infection. γ IP-10/CXCL10 and Mig/CXCL9 induced almost no chemotactic migration in freshly purified peritoneal exudate eosinophils (CI about 1.5 ± 0.2), whereas eotaxin/CCL11 induced the significant chemotactic migration of the cells (CI about 2.8 ± 0.4) (Fig. 7a). IP-10/CXCL10 and Mig/CXCL9 induced an increasing chemotactic migration of purified peritoneal exudate eosinophils in mice postinfection with S. japonicum by day 7 (Fig. 7b), day 28 (Fig. 7c) and day 42 (Fig. 7d), meanwhile eotaxin/CCL11 induced a decreasing chemotactic migration in these cells (Fig. 7). SDF-1α/CXCL12 induced a significantly but unchanged chemotactic migration of purified peritoneal exudate eosinophils in mice postinfection with S. japonicum by day 0, day 7, day 28 and day 42 (data not shown), indicating that CXCR4 up-regulation was not chemotactic function related. To be sure that the chemotactic abilities of eosinophils were not affected by the stimulation of thioglycolate during the experimental procedure, we therefore purified eosinophils from peripheral blood from mice with or without infection of S. japonicum. The pattern of migration of peripheral eosinophils towards different chemokines mentioned above was very similar to that of peritoneal exudate eosinophils (Table 3).

Figure 7.

The migration of peritoneal exudate eosinophils from either healthy native C57BL/6 (H-2b) mice (a) or C57BL/6 (H-2b) infected with S. japonicum for 7 days (b), 28 days (c) or 42 days (d) towards eotaxin/CCL11 (□), γ IP-10/CXCL10 (⋄) or Mig/CXCL9 (▪). All results were determined as described in Materials and Methods and expressed as chemotactic index (CI ± SD), and based on triplicate determination of chemotaxis on each concentration of chemokines indicated. The illustrated data are from a single representative experiment of six performed. All * indicate significant differences between healthy native versus S. japonicum-infected animals in each concentration of chemokine applied. All P < 0.05.

Table 3.

The migration of peripheral blood eosinophils^*

Postinfection	γ IP-10/CXCL10 (100 ng/ml)	Mig/CXCL9 (100 ng/ml)	eotaxin/CCL11 (100 ng/ml)
Native	1.4 ± 0.3†	1.5 ± 0.5	3.0 ± 0.7
Day 7	2.5 ± 0.5	2.4 ± 0.8	2.2 ± 0.3
Day 14	2.9 ± 0.6	ND	ND
Day 28	3.0 ± 0.4	2.9 ± 0.9	1.5 ± 0.6
Day 42	3.2 ± 0.4	3.3 ± 0.9	1.3 ± 0.4

^*Cells were purified from either healthy native C57BL/6 mice or C57BL/6 infected with S. japonicum for 7 days, 14 days, 28 days or 42 days.

^†All results were determined as described in Materials and Methods and expressed as chemotactic index (CI), based on triplicate determination of chemotaxis at optimal concentration of chemokines indicated. The listed data are mean values (±SD) of four experiments performed. ND, no determination.

Thus, CXCR3 was highly and functionally up-regulated on peritoneal exudate eosinophils in mice infected with S. japonicum; meanwhile CCR3 was significantly and functionally down-regulated in these cells.

Discussion

Infection with helminth parasites is associated with elevated IgE, systemic eosinophilia, and mast cell proliferation.³⁴ These features reflect a polarized Th2 cell response enriched in IL-4, IL-5, and IL-13 production.^35,36 Helminth infection often induces eosinophilic granulocytes to infiltrate host tissues in intense foci associated with extensive tissue damage.³⁷ Schistosomiasis causes a host response to trapped parasite eggs in the tissue resulting in the formation of granulomatous lesions in the liver and intestine. Migrating larvae are likely to be killed in the lungs by eosinophils through an antibody-dependent cytotoxicity mechanism. Activated eosinophils kill Strongyloides larvae in the presence of specific antibodies in mice.³⁸ Elevation of IL-5 in mesenteric lymph node cells in S. japonicum-infected mice suggested activation of eosinophils in these mice. Because schistosomulae migrate in the lungs, eosinophils might be further stimulated by IL-5 produced by CD4 CD8 double negative T cells in the lungs.³⁹ In spite of the fact that many components and regulatory features of the eosinophil recruitment pathway have been described^34–41 the molecular basis for the massive infiltration of eosinophils during helminth infection, particularly the roles of the chemokines receptors and their ligands, is not fully understood. In the present study, we have explored chemokine receptor expression in eosinophils obtained from the peritoneal lavage exudation fluid of mice infected with S. japonicum. In the present study we have demonstrated that CXCR3 is highly and functionally up-regulated on peritoneal exudate eosinophils in mice infected with S. japonicum; meanwhile CCR3 is significantly and functionally down-regulated in these cells, and CXCR3 is a dominant factor directing eosinophils into S. japonicum egg-induced granuloma in the mouse liver during the process of infection. To our knowledge, this is the first report on CXCR3 overexpression on eosinophils from mice infected with S. japonicum, and provides direct evidence of the biological activity of eosinophils to participate in granuloma formation in acute S. japonicum infection by means of CXCR3 and its ligands γ IP-10/CXCL10 and Mig/CXCL. Thus, a rather complex picture is now beginning to take shape of how eosinophils selectively enter eosinophilic granuloma induced by S. japonicum eggs, and cause physiological and pathophysiological events under continuous interaction with chemokines and cytokines to damage liver and intestine.

CCR3 is expressed in human on eosinophils and basophils^37,40 on a subset of CD4⁺ T cells⁴¹ and on dendritic cells.⁴² CCR3 expression on T cells requires stimulation with both the Th1 cytokine IL-2 and the Th2 cytokine IL-4.³³ Cellular sources of the CCR3 ligand eotaxin/CCL11 are bronchial epithelial cells, T cells, macrophages and eosinophils.⁴³ The importance of this chemokine in allergic diseases is widely recognized because of its activation and attraction of eosinophils, basophils and T cells.^33,41 In the human, CCR3 is dominantly expressed on both bronchoalveolar lavage fluid and peripheral blood eosinophils from patients with eosinophilic lung diseases,²³ suggesting a primary as well as a causal role for this receptor in eosinophil recruitment throughout the process of movement from the circulation to inflamed tissue sites. On the other hand, decreased CCR3 expression on the eosinophils in bronchoalveolar lavage fluid in comparison with that on peripheral blood eosinophils may be due to internalization of the surface receptors caused by binding with ligands such as eotaxin/CCL11.²³ Meanwhile, CXCR4 is functionally increased, with statistical significance, in bronchoalveolar lavage fluid eosinophils.²³ In the present study, we have found that CCR3 expression on the infiltrating eosinophils in both egg-induced hepatic granuloma and peritoneal exudation is significantly decreased in mice during the process of infection with S. japonicum at both protein and mRNA levels, but almost all infiltrating eosinophils are induced the CXCR3 expression during the infection. Meanwhile, CXCR4 expression is significantly up-regulated postinfection, but its function to direct eosinophil migration was shown to be unchanging. In mouse, we have observed decreased CCR3 expression on the eosinophils in situ during the formation of eosinophilic granuloma induced by S. japonicum eggs and on peritoneal exudate eosinophils during infection with S. japonicum, which may be caused by the eosinophil cytotoxicity of the infection of the parasite. The mechanism for this phenomenon should be subjected to further investigation. The increased CXCR3 expression by the eosinophils in situ during the process of infection of S. japonicum, on the other hand, may be caused by the eosinophil activation resulting from the infection by the parasite. Nevertheless, CXCR3 expression plays a major role in the migration of eosinophils to the inflammatory locals.

In summary, we have demonstrated that CXCR3 was highly and functionally up-regulated on infiltrating eosinophils in S. japonicum egg-induced granuloma in the mouse liver and peritoneal exudate eosinophils in mice infected with S. japonicum. These findings may be crucial for better understanding of the mechanism of formation of liver granuloma dominated by eosinophils.

Abbreviations

CI

chemotactic index

CXCL (CC)

CXC (CC) chemokine ligand

CXCR (CC)

CXC (CC) chemokine receptor

γ IP-10

interferon-γ inducible protein-10

MCNC

migrating cells on negative control

Mig

monokine induced by interferon-γ

MIP

macrophage inflammatory protein