Altered cytokine expression in mesenteric lymph nodes in a rat strain (Matsumoto Eosinophilic Shinshu) that spontaneously develops hypereosinophilia

Abstract

The Matsumoto Eosinophilic Shinshu (MES) rat is an inbred mutant strain that spontaneously develops systemic hypereosinophilia with eosinophilic inflammatory lesions similar to those associated with hypereosinophilic syndrome in humans and other mammals. To elucidate the pathogenic mechanisms that underlie these features of MES rats, we examined the pattern of cytokine gene expression in mesenteric lymph nodes (MLNs), the thymus, and peripheral blood mononuclear cells as well as the blood clinicopathology and MLN lymphocytic subsets of these animals. MES rats exhibited both leucocytosis, attributable in large part to hypereosinophilia and neutrophilia, and immunoglobulin M (IgM) and IgA gammaglobulinaemia, with increased titres of IgM autoantibodies to nuclear antigens. Reverse transcription and polymerase chain reaction analysis revealed that the amounts of interleukin (IL)-5, IL-4, eotaxin, and interferon-γ mRNAs were increased in the MLN lymphocytes of MES rats compared with the corresponding values for Sprague-Dawley rats. Intraperitoneal administration of a monoclonal antibody specific for IL-5 resulted in an immediate suppression of hypereosinophilia and a delayed suppression of neutrophilia in MES rats. Flow cytometry revealed an increased percentage of CD3⁺ CD4^– CD8^– T lymphocytes in MLNs of MES rats. Our results suggest that the hypereosinophilia of MES rats results from an increased production of IL-5, and that the eosinophilic inflammatory lesions of these animals, which are largely restricted to the gut, may be related both to cytokine overexpression in MLNs and to T helper 1 and 2 immunological responses.

Introduction

Eosinophilia develops in reactive diseases such as allergy, parasitic infection, certain malignancies, immunodeficiency and idiopathic hypereosinophilic syndrome (HES). It is accompanied by a variety of inflammatory reactions as a consequence of the release from activated eosinophils of toxic inflammatory mediators, such as cationic proteins, proteases and oxygen radicals.^1,2 In particular, systemic hypereosinophilia, as in HES, results in severe organ damage.^1,3 Despite its potential destructiveness, the pathogenesis of eosinophilia remains largely uncharacterized.

Several cytokines have been shown to play important roles in the development of eosinophilia.⁴ Interleukin (IL)-3, IL-5 and granulocyte–macrophage colony-stimulating factor (GM-CSF) promote eosinophilopoiesis, and their enhanced production results in eosinophilia. These eosinophilopoietic cytokines are released from dysregulated T cells together with T helper (Th) cytokines in various types of eosinophilia. Th1 cytokines [interferon (INF)-γ and IL-2] and Th2 cytokines (IL-4, IL-5, IL-6 and IL-10) elicit cellular and humoral immunity, respectively.^1,2,4 In addition, eosinophil recruitment to inflammatory lesions is regulated by chemotactic cytokines (chemokines) such as eotaxin and RANTES (regulated on activation, normal T-cell expressed and secreted). These chemokines have been detected locally in eosinophilic inflammatory lesions in various kinds of eosinophilia.⁵ Characterization of patterns of cytokine expression might thus be expected to provide insight into the pathogenic mechanisms of eosinophilia.

The Matsumoto Eosinophilic Shinshu (MES) rat is an inbred mutant strain that originated from specific-pathogen-free Sprague-Dawley (SD) rats.⁶ MES rats spontaneously develop systemic hypereosinophilia from 8 weeks of age onwards without evidence of malignant neoplasia. They manifest many severe eosinophilic inflammatory lesions characterized by infiltration of numerous eosinophils with formation of severe fibrosis and granulomas, such as gastroenteritis, mesenteric lymphadenitis, and aortitis (at a high frequency) as well as hepatic fibrosis (at a lower frequency). In addition, lymphatic abnormalities such as mesenteric lymphocytosis, plasmacytosis, and peripheral T lymphocytosis are found. In bone marrow, there are no abnormalities in myeloid cells except for the proliferation of mature eosinophils. These features of MES rats are similar to those of HES, especially HES with gut involvement, in humans and other mammals. The MES rat has therefore been proposed as an animal model for studies of hypereosinophilia.^7,8

To shed light on the pathogenic mechanisms of these various features of MES rats, we have now determined the abundance of various cytokine mRNAs in the mesenteric lymph nodes (MLNs), thymus, and peripheral blood mononuclear cells (PBMCs) of these animals during hypereosinophilia. We also investigated blood clinicopathology and lymphocyte subsets in their MLNs.

Materials and methods

Animals

Male specific-pathogen-free MES and SD rats were obtained from Japan SLC (Shizuoka, Japan). The animals were housed individually in aluminium breeding cages in a barrier-system room controlled for temperature (21–25°), humidity (45–65%) and lighting (light/dark 12/12 hr; lights on at 08:00 hr). They were fed a commercial diet and provided with water ad libitum. This study was performed in accordance with the Guidelines for Animal Care and Experimentation of Kissei Pharmaceutical Co., Ltd (Nagano, Japan).

Blood sampling and cell purification

Blood samples were collected from the jugular vein with disposable syringes in the presence (for haematology and cell purification) or absence (for serological examination) of ethylenediaminetetraacetic acid (EDTA) (dipotassium salt). The cells used for analysis of mRNA abundance or for flow cytometry were purified by centrifugation with the use of Lympholyte-Rat (Cedarlane, Hornby, Ontario, Canada) from cell suspensions derived from MLNs, thymus, or peripheral blood. The cells counted as lymphocytes with an automated haematology analyser constituted 94·3–100% of the purified preparations.

Clinicopathology

Blood samples were analysed with an automated haematology analyser (Technicon H*1E system and software) suitable for rat specimens (Bayer Medical, Tokyo, Japan). The counts for total leucocytes, neutrophils, lymphocytes, monocytes, eosinophils and basophils were determined. Serum immunoglobulin (Ig) levels were measured by enzyme-linked immunosorbent assay (ELISA) with rat IgG, IgM and IgA quantification kits (Bethyl Laboratories, Montgomery, TX) and a rat IgE quantification kit (Shibayagi, Gunma, Japan). Serum titres of Ig specific for single-stranded DNA (ssDNA) or double-stranded DNA (dsDNA) were also measured by ELISA as described previously, but with a slight modification.⁹ In brief, the wells of 96-well plates that had been coated with poly-l-lysine were coated overnight at 4° with ssDNA or dsDNA (Sigma, St. Louis, MO). They were then incubated consecutively at room temperature for 2 hr with bovine serum albumin, for 1 hr with test serum (1 : 100 dilution), and for 1 hr with horseradish peroxidase-labelled goat antibodies specific for rat IgM or IgA (Bethyl Laboratories). The wells were washed with phosphate-buffered saline (PBS) between incubations. Immune complexes were detected with 3,3′,5,5′-tetramethylbenzidine as substrate and by measurement of absorbance at 450 nm with a microplate reader (MTP-300; Corona Electric, Ibaraki, Japan).

Semiquantitative reverse transcriptase–polymerase chain reaction (RT-PCR) analysis

Total RNA was extracted from purified cells with the use of the TRIzol protocol (Invitrogen, Carlsbad, CA). The yield and quality of the isolated RNA were assessed by spectrophotometry and electrophoresis on a 1% agarose gel; the yield ranged from 0·6–2·9 µg of total RNA per 1 × 10⁶ cells. The RNA (1·5 µg) was subjected to reverse transcription with an oligo(dT) primer and SuperScript II reverse transcriptase (Invitrogen) in a volume of 20 µl. The resulting cDNA (2 µl) was amplified by polymerase chain reaction (PCR) with Platinum Taq DNA polymerase (Invitrogen) in PCR buffer containing MgCl₂, deoxynucleoside triphosphates and gene-specific forward and reverse primers in a volume of 50 µl. The primers for IL-2, IL-4, IL-5, IL-6, interferon (IFN)-γ and β-actin were obtained from Maxim Biotech (South San Francisco, CA); the sizes of the PCR products were 297, 381, 328, 453, 283 and 474 bp, respectively. The primers for IL-3, GM-CSF and eotaxin were designed as described,^10,11 and were as follows: IL-3 forward, 5′-TCCTGATGCTCTTCCACC-3′; IL-3 reverse, 5′-TCGCAGCTGCAGGAATACAACACT-3′; GM-CSF forward, 5′-TGCTCTGGAGAACGAAAAGAACG-3′; GM-CSF reverse, 5′-ATTGAGTTTGGTGAGGTTGCCCC-3′; eotaxin forward, 5′-AACTTCCTGCTGCTTTACCATGA-3′; eotaxin reverse, 5′-TTTTGGAGTTTGGAGTTTTTGGT-3′. The sizes of the corresponding PCR products were 302, 132 and 202 bp, respectively. The amplification protocol comprised an initial denaturation step of 1 min at 96°; various numbers of cycles consisting of 1 min at 94°, 1 min at 58°, and 1 min at 72°; and a final extension step of 10 min at 72°. The cycle numbers were 30 for β-actin, 34 for IL-6 and GM-CSF, and 35 for the other targets, which had been determined to be within the linear amplification range in preliminary PCR runs. The reaction was performed in a thermal cycler (iCycler; Bio-Rad, Tokyo, Japan). Positive and negative controls (PCR mixtures with or without standard cDNA, respectively) were included in preliminary PCR runs. The PCR products were detected by electrophoresis on a 2% agarose gel and staining with ethidium bromide. Polaroid photographs of the gels were scanned and the intensity of the bands was measured by densitometry with National Institutes of Health (NIH) Image software. The intensity of cytokine bands was normalized by that of the corresponding β-actin band.

Administration of a monoclonal antibody to IL-5

A monoclonal antibody (mAb) to IL-5 (TB13, kindly provided by M. Takamoto, Shinshu University, Nagano, Japan) was administered intraperitoneally to 10-week-old MES rats in a single dose of 15 mg/body, which was considered to be approximately equivalent to that (2 mg/body) previously shown to inhibit eosinophilia in mice infected with Toxocara canis.¹² As a control, rats were injected with physiological saline in the same way. Blood samples were collected before and 1, 2, 3, 5, 7, 10, 14, 21 and 28 days after injection of the mAb or physiological saline and were subjected to haematological analysis. In addition, MLNs and stomachs were collected on day 28 and were subjected to histological analysis. Tissue samples were fixed with 10% neutral buffered formalin, embedded in paraffin, and sectioned at 3-µm thickness. The sections were stained with haematoxilin and eosin. To detect eosinophils as described previously,¹³ the sections were also stained with Congo red and were finally counterstained with haematoxylin. The areas of eosinophil infiltration in the total MLN or stomach areas were quantitatively analysed with an image processor (LUZEX-3; Nikon, Tokyo, Japan), and were expressed as a percentage.

Flow cytometry

Two- or three-colour flow cytometric analysis was performed with the following mouse mAbs: fluorescein isothiocyanate (FITC)-labelled anti-pan-T cells (OX-52) and phycoerythrin (PE)-labelled anti-CD45RA (OX-33) from BD Pharmingen (San Jose, CA); FITC-labelled anti-CD3 (1F4) from Caltag (Burlingame, CA); and PE-labelled anti-CD4 (W3/25) as well as PE-Cy5-labelled anti-CD8α (OX-8) from Serotec (Oxford, UK). Cells (1 × 10⁶) were incubated with specific mAbs or non-specific isotype control antibodies for 30 min at 4°, washed twice with PBS, and suspended in PBS for analysis with a FACSCalibur flow cytometer and CELLQuest software (Becton-Dickinson, Mountain View, CA). Forward and side scatter were monitored, and the percentages of each cell subset were determined for 1 × 10⁴ gated cells.

Statistics

Data are expressed as means ± standard deviations. Statistical analysis was performed with Student's or Aspin–Welch's t-test. A P-value of <0·05 was considered statistically significant.

Results

Clinicopathology

Clinicopathological features were examined in MES rats at 14 weeks of age, when eosinophilia had already developed.^7,8 The counts for all types of leucocytes were higher in MES rats than in SD rats, the counts for neutrophils, lymphocytes, monocytes, eosinophils and basophils being 7·8-, 1·5-, 1·9-, 18·2- and 7·0-fold higher, respectively (Table 1). MES rats also showed higher levels of IgM (3·5-fold) and IgA (2·1-fold) in serum than did SD rats. Serum electrophoresis revealed that high levels of IgM and IgA were diagnosed as polyclonal gammopathy (data not shown). The serum level of IgG was lower in MES rats than in SD rats, probably as a result of the higher levels of IgM and IgA. There was no difference in serum IgE levels between MES and SD rats.

Table 1.

Blood leucocyte counts and serum immunoglobulin (Ig) levels in Matsumoto Eosinophilic Shinshu (MES) and Sprague-Dawley (SD) rats at 14 weeks of age

Cell or Ig type	MES rats	SD rats
Total leucocytes (109 cells/l)	20·22 ± 2·89**	8·35 ± 2·33
Neutrophils (109 cells/l)	7·06 ± 2·18**	0·90 ± 0·14
Lymphocytes (109 cells/l)	10·70 ± 0·78**	7·19 ± 2·03
Monocytes (109 cells/l)	0·28 ± 0·09*	0·15 ± 0·06
Eosinophils (109 cells/l)	1·82 ± 1·11*	0·10 ± 0·04
Basophils (109 cells/l)	0·07 ± 0·02**	0·01 ± 0·01
IgG (µg/ml)	3712 ± 517**	5566 ± 879
IgM (µg/ml)	2419 ± 1025**	701 ± 201
IgA (µg/ml)	116 ± 31**	56 ± 17
IgE (ng/ml)	14 ± 10	15 ± 12

Data are means ± standard deviations of values from 10 male rats.

^*P < 0·05

^**P < 0·01 versus SD rats.

Cytokine mRNA abundance in MLN and thymic lymphocytes and in PBMCs

The abundance of various cytokine mRNAs was determined by RT-PCR analysis in lymphocytes purified from MLNs and thymus as well as in PBMCs of MES rats at 14 weeks of age. Overall, the amounts of cytokine mRNAs in MES rats were greater in MLN cells than in the thymus or PBMCs (Fig. 1). The abundance of IL-4, IL-5, IFN-γ and eotaxin mRNAs was greater in MLN cells of MES rats than in those of SD rats, with the difference for IL-5 mRNA being the greatest (29·2-fold). The amounts of IL-2, IL-3 and GM-CSF mRNAs in MLNs did not differ significantly between MES and SD rats. The amount of IL-3 mRNA was significantly lower in thymic lymphocytes of MES rats than in those of SD rats, but there were no differences in the amounts of any of the examined cytokine mRNAs in PBMCs between MES and SD rats. Whereas the mRNAs for IL-4, IL-5 and eotaxin were relatively abundant in MLN cells of MES rats, they were not detected or present at low levels in the thymus or PBMCs.

Figure 1.

Abundance of cytokine mRNAs in lymphocytes isolated from mesenteric lymph nodes (MLNs) or from the thymus as well as in peripheral blood mononuclear cells (PBMCs) of Matsumoto Eosinophilic Shinshu (MES) and Sprague-Dawley (SD) rats at 14 weeks of age. (a) Representative agarose gel electrophoresis. (b) Densitometric analyses for MES rats (solid columns) and SD rats (open columns). Data are means ± standard deviations of values from six male rats. *P < 0·05, **P < 0·01 versus SD rats. IL, interleukin; IFN, interferon; GM-CSF, granulocyte–macrophage colony-stimulating factor.

Serum autoantibodies

The serum titres of IgM and IgA specific for ssDNA or dsDNA were examined in MES rats, as the IgM and IgA gammaglobulinaemia and the increased abundance of Th2 cytokine mRNAs found in these animals were suggestive of autoimmunity. MES rats showed higher titres of IgM specific for ssDNA or dsDNA than did SD rats, whereas the titres of IgA specific for ssDNA or dsDNA did not differ significantly between the two strains (Fig. 2).

Figure 2.

Serum titres of immunoglobulin M (IgM) and IgA specific for single-stranded DNA (ssDNA) or double-stranded DNA (dsDNA) in Matsumoto Eosinophilic Shinshu (MES) rats (solid circles) and Sprague-Dawley (SD) rats (open circles) at 14 weeks of age. Each circle corresponds to an individual rat. **P < 0·01 for the indicated comparisons.

In vivo effects of administration of a mAb to IL-5

We next examined the haematological effects of administration of a mAb to IL-5 in MES rats, given the marked increase in the amount of IL-5 mRNA apparent in MLN cells of these animals. Whereas 10-week-old MES rats injected with physiological saline manifested a time-dependent increase in eosinophil count, animals injected with the mAb to IL-5 exhibited a decrease in this parameter that continued for 7 days (Fig. 3). The eosinophil count was thus significantly lower in the rats injected with antibody than in those injected with saline for 10 days after injection. The eosinophil counts of antibody-injected MES rats both 5 days (0·21 × 10⁹ cells/l) and 7 days (0·19 × 10⁹ cells/l) after injection were similar to those of control SD rats (range of 0·06–0·17 × 10⁹ cells/l determined from 180 SD rats). In contrast, the neutrophil counts of MES rats injected with the mAb to IL-5 were significantly lower than those of the saline-treated animals only on days 21 and 28. Lymphocyte, monocyte and basophil counts in MES rats were not affected by administration of the mAb to IL-5 (data not shown). Histologically, all the saline-injected MES rats (three of three rats) exhibited infiltration of numerous eosinophils with formation of granulomas in the medulla of MLNs and formation of fibrosis and granulomas in the lamina propria and submucosa of glandular stomachs. In contrast, all the antibody-injected MES rats (three of three rats) exhibited remarkable suppression of eosinophil infiltration, especially in MLNs. Image analysis revealed that the areas of eosinophil infiltration in the total MLN areas were 3·7, 5·1 and 8·9% in the individual saline-injected rats and 0·5, 2·1 and 2·2% in the individual antibody-injected rats. Furthermore, one of those rats exhibited suppression of eosinophilic inflammation in the stomach (Fig. 4).

Figure 3.

Effects of administration of a monoclonal antibody (mAb) to interleukin (IL)-5 on total leucocyte, eosinophil, and neutrophil counts in Matsumoto Eosinophilic Shinshu (MES) rats. MES rats were injected intraperitoneally with a single dose (15 mg/body) of a mAb to IL-5 (solid circles) or with physiological saline (open circles) at 10 weeks of age. Total leucocyte, eosinophil and neutrophil counts were monitored at the indicated times thereafter. Data are means ± standard deviations of values from three male rats. *P < 0·05, **P < 0·01 versus saline-treated controls.

Figure 4.

Histology of representative Matsumoto Eosinophilic Shinshu (MES) rats injected with physiological saline (a and c) and monoclonal antibody (mAb) to interleukin (IL)-5 (b and d). Mesenteric lymph nodes (MLNs) of saline-injected (a) and antibody-injected (b) rats stained with Congo red and haematoxylin are shown. Note that eosinophil infiltration (8·9%) with formation of granuloma is observed in the medulla of MLNs in (a) but it is greatly suppressed to 0·5% in (b). Stomachs of saline-injected (c) and antibody-injected (d) rats stained with haematoxylin and eosin are also shown. Note that eosinophil infiltration with formation of fibrosis and granuloma is observed in the lamina propria and submucosa of glandular stomach in (c) but it is suppressed in (d). Bar, 200 µm.

Lymphocyte subsets in MLNs

Finally, we characterized the lymphocyte subsets present in MLNs of MES rats, given the altered pattern of cytokine gene expression apparent in the MLN lymphocytes of these animals. Flow cytometry revealed that all MES rats analysed showed similar levels of eosinophilia as detected by clinicopathological examination (data not shown). The percentages of pan-T cells and CD45RA⁺ cells in MLNs of MES rats were 1·6- and 0·6-fold, respectively, higher than those in MLNs of SD rats (Table 2). As for T-cell subsets defined by CD3, CD4 and CD8 expression, MES rats showed a slightly but significantly higher percentage of CD3⁺ CD4^– CD8^– cells than did SD rats. There were no differences in the percentages of CD3⁺ CD4⁺ CD8^–, CD3⁺ CD4^–CD8⁺ and CD3⁺ CD4⁺ CD8⁺ cells between the two strains. The percentages of CD3^– CD4⁺ and CD3^– CD8⁺ cells were negligible in both MES and SD rats (data not shown).

Table 2.

Lymphocyte subsets in mesenteric lymph nodes (MLNs) of Matsumoto Eosinophilic Shinshu (MES) and Sprague-Dawley (SD) rats at 14 weeks of age

Cell type	MES rats	SD rats
Pan-T cells (% of all cells)	49·3 ± 3·3**	31·2 ± 9·5
CD45RA+ (% of all cells)	42·0 ± 5·3**	65·2 ± 9·6
CD4+ CD8– (% of CD3+ cells)	64·4 ± 2·0	67·7 ± 2·8
CD4– CD8+ (% of CD3+ cells)	26·0 ± 2·0	28·0 ± 2·4
CD4+ CD8+ (% of CD3+ cells)	2·8 ± 0·6	2·6 ± 0·1
CD4– CD8– (% of CD3+ cells)	6·8 ± 1·6**	1·8 ± 0·3

Data are means ± standard deviations of values from six male rats.

^**P < 0·01 versus SD rats.

Discussion

Matsumoto Eosinophilic Shinshu (MES) rats at 14 weeks of age were shown to manifest leucocytosis that was attributable in large part to hypereosinophilia, consistent with our previous observations.^6–8 We also showed here that this hypereosinophilia was accompanied by increased expression of the genes for several cytokines in MLN lymphocytes.

The most marked change in cytokine gene expression in MES rats was the increase in the abundance of IL-5 mRNA in MLNs. The amounts of the mRNAs for IL-3 and GM-CSF, two other eosinophilopoietic cytokines, did not differ between the MLNs of MES and SD rats. Furthermore, treatment of MES rats at 10 weeks of age with a mAb to IL-5 resulted in a rapid and pronounced decrease in eosinophil count. These results thus indicate that the hypereosinophilia of MES rats is predominantly attributable to increased production of IL-5, rather than to that of IL-3 or GM-CSF. IL-5 specifically promotes the growth and differentiation of eosinophils, whereas IL-3 and GM-CSF each target not only eosinophils but also other granulocytes.¹⁴ Indeed, IL-5, rather than IL-3 or GM-CSF, plays a pivotal role in the pathogenesis of various types of eosinophilia in humans.^1–3

MES rats develop neutrophilia as well as mild monocytosis and basophilia in addition to hypereosinophilia. We previously proposed that these changes might be a response to progressive eosinophilic inflammation.⁸ This proposal is supported by our present study, given that we did not detect any differences in the abundance of IL-3 or GM-CSF mRNAs between MES and SD rats, and that administration of a mAb to IL-5 did not affect the monocyte or basophil counts of MES rats. The delayed decrease in the neutrophil count observed after injection of the mAb to IL-5 was probably a result of the healing of eosinophilic inflammation as a consequence of the immediate reduction in the number of eosinophils.

In addition to the increase in the amount of IL-5 mRNA, the abundance of IL-4 mRNA was significantly increased and that of IL-6 mRNA tended to be higher in the MLNs of MES rats compared with those of SD rats. The increased expression of these Th2 cytokines is suggestive of a Th2 immunological condition similar to that associated with most types of eosinophilia in humans.^1,2,4 Such Th2 responses are characterized by an impaired cellular response and an increased B-cell activity that manifest as gammaglobulinaemia and autoimmunity.⁴ The increased serum levels of total IgM and IgA as well as of IgM autoantibodies to nuclear antigens (ssDNA and dsDNA) detected in MES rats in the present study are thus consistent with a Th2 response. Furthermore, mesenteric plasmacytosis, which is considered as histological evidence of such a response, was previously demonstrated in MES rats.⁷ The Th2 response of MES rats is probably attributable to the increased expression of IL-5 and possibly that of IL-6. Both IL-5 and IL-6 promote the terminal differentiation of B cells into IgM- or IgA-producing cells.^15–17 Transgenic mice that produce excess amounts of IL-5 manifest features similar to those of MES rats, including systemic hypereosinophilia, IgM and IgA gammaglobulinaemia, and lymphocytosis caused mainly by increased numbers of B cells, especially Ly-1⁺ B cells that produce IgM autoantibodies.¹⁸ Splendore–Hoeppli material, which is thought to comprise deposited antigen–antibody complexes resulting from the host immune response,^19,20 has been detected at eosinophilic inflammatory lesions in MES rats,⁷ although there has been no direct evidence of Ig deposits in these animals. Given that the relevance of the increased levels of IgM and IgA to disease pathogenesis in MES rats remains unclear, the possible participation of these Igs in the development of eosinophilic inflammatory lesions warrants further investigation.

IL-4 functions as a switch for IgE production.²¹ The amounts of IL-4 and IgE increase simultaneously in reactive eosinophilia and in many patients with HES.⁴ However, the increase in the amount of IL-4 mRNA in the MLNs of MES rats was not accompanied by an increase in the serum level of total IgE, possibly because of the associated increase in the expression of IFN-γ, which inhibits production of this Ig.²² It may also be relevant that the SD rat, from which the MES rat was derived, is an IgE low-responder.²³

The increase in IFN-γ expression in MES rats might be caused by the mesenteric lymphadenitis that manifests as eosinophilic granulomas after 10 weeks of age.⁷ Granulomas are formed in response to infectious or non-infectious processes under both Th1 and Th2 immunological conditions.⁴ The expression of IL-2, another Th1 cytokine, did not appear to be changed in MES rats. It is possible that the granulomas in MES rats at 14 weeks of age are still in the early stages, given that IL-2 expression is increased only at the later stages of granuloma development.²⁴

In addition to eosinophilopoietic and Th cytokines, the amount of mRNA for the eosinophilic chemokine eotaxin was increased in the MLNs of MES rats. Among the eosinophilic chemokines, including RANTES, eotaxin is the most selective for local eosinophil recruitment into inflammatory lesions.⁵ The increase in eotaxin expression might therefore reflect the selective recruitment of eosinophils into inflammatory lesions that are apparent mainly around the gut in MES rats.⁷ Furthermore, eotaxin and IL-5 might function together to promote eosinophilic inflammation in these animals, given that these proteins have been shown to act co-operatively and synergistically to regulate the recruitment of eosinophils into tissues.^25,26

The increase of multiple cytokines in MES rats might be a result of impaired induction of transcription factors, especially those responsible for Th2 cytokine production such as GATA-3, Stat6, NF-AT and AP-1.^27,28

Abnormalities in cytokine gene expression were largely restricted to the MLNs of MES rats, suggesting that gut immunity is affected in these animals. MLNs are a component of gut-associated lymphoid tissue and provide a site for antigen-specific lymphocytes to expand and mature.²⁹ Immune abnormalities of MES rats include lymphocytosis, plasmacytosis and eosinophilic inflammatory lesions.⁷ The attribution of these immune abnormalities to MLNs may explain why the eosinophilic inflammatory lesions of MES rats are largely restricted to the gut, rather than to the heart, which is the most affected organ in humans with systemic hypereosinophilia, including those with HES.^1–3

With the exception of eosinophilic leukaemia, most types of eosinophilia, including HES, are thought to result from the release of cytokines, specifically IL-5, from dysregulated T cells.^1–3 Phenotypically abnormal T cells that release large amounts of IL-5, including CD3^– CD4⁺ and CD3⁺ CD4^– CD8^– cells, have been identified with evidence of clonality in the peripheral blood of some patients with HES.^30–33 We have now shown that the percentage of CD3⁺ CD4^– CD8^– cells in MLNs was increased in MES rats compared with that in SD rats, although the clonality of these cells was unclear. Although extrathymic T lymphopoiesis has been shown to occur in the MLNs of athymic mice, it is totally suppressed in euthymic mice.³⁴ It is thus possible that abnormal mesenteric T cells of extrathymic origin might be responsible for the increased production of cytokines in MES rats, which warrants further investigation.

In conclusion, we have shown that the amount of IL-5 mRNA is increased specifically in MLNs of MES rats and we have demonstrated the existence of both Th1 and Th2 immunological responses in these animals. Both of these findings may be relevant to the hypereosinophilia and eosinophilic inflammatory lesions that develop in MES rats. Our results further support the use of the MES rat as an animal model for studies of hypereosinophilia, especially that with gut involvement.

Abbreviations

dsDNA

double-stranded DNA

ELISA

enzyme-linked immunosorbent assay

FITC

fluorescein isothiocyanate

GM-CSF

granulocyte–macrophage colony-stimulating factor

HES

hypereosinophilic syndrome

IFN-γ

interferon-γ

Ig

immunoglobulin

IL

interleukin

mAb

monoclonal antibody

MES

Matsumoto Eosinophilic Shinshu

MLN

mesenteric lymph node

PBMC

peripheral blood mononuclear cell

PBS

phosphate-buffered saline

PE

phycoerythrin

RANTES

regulated on activation, normal T-cell expressed and secreted

RT-PCR

reverse transcriptase–polymerase chain reaction

SD

Sprague-Dawley

ssDNA

single-stranded DNA

Th

T helper