Cytokine mRNA Expression Profiles in Rats Infected with the Intestinal ...

3 downloads 0 Views 265KB Size Report
increase in mice infected with Heligmosomoides polygyrus (35), and these findings suggest that the IL-10 gene is regulated differently from the other Th2 ...
INFECTION AND IMMUNITY, Dec. 1995, p. 4653–4660 0019-9567/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 63, No. 12

Cytokine mRNA Expression Profiles in Rats Infected with the Intestinal Nematode Nippostrongylus brasiliensis SHINJI MATSUDA,* RYUICHI UCHIKAWA, MINORU YAMADA,

AND

NAOKI ARIZONO

Department of Medical Zoology, Kyoto Prefectural University of Medicine, Kawaramachi, Hirokoji, Kyoto 602, Japan Received 17 April 1995/Returned for modification 20 June 1995/Accepted 12 September 1995

Although the immune responses to intestinal nematode infection have been well studied and have been shown to be strongly driven by Th2-associated cytokines in mice, such information has been limited with respect to rats. We investigated changes in levels of the mRNAs encoding interleukin-2 (IL-2), IL-3, IL-4, IL-5, IL-10, and gamma interferon in the mesenteric lymph nodes of rats infected with Nippostrongylus brasiliensis by reverse transcription-PCR in comparison with immunoglobulin E (IgE)/IgG2a antibody, eosinophil, basophil, and mucosal mast cell responses. In the two rat strains used, Brown Norway and Fischer-344, which show different responses to allergens, serum IgE increased to much higher levels in the former than in the latter 2 weeks after infection. Intestinal mastocytosis was observed much earlier and more intensely in Brown Norway rats than in Fischer-344 rats, but the degrees of peripheral eosinophilia and basophilia did not differ between the two strains. In both strains, IL-3, IL-4, and IL-5 mRNA expression increased and peaked around 7 to 14 days after infection, while expression of IL-2, IL-10, and gamma interferon mRNAs did not change notably throughout the experimental period. The highest IL-4 mRNA expression was observed slightly earlier in Brown Norway than in Fischer-344 rats, but levels of IL-3 and IL-5 mRNAs peaked synchronously in both strains. The amounts of mRNAs encoding these three cytokines were always higher in Brown Norway than in Fischer-344 rats. It is suggested that in rats, Th2 or Th2-like cells are also induced after nematode infection, and IgE elevation is mainly related to increased IL-4 gene expression. Intestinal infection with nematodes induces various immune responses in humans and animals. The most striking changes are represented as elevated serum immunoglobulin E (IgE), eosinophilia, and intestinal mucosal mastocytosis, which occur 2 weeks or more after primary infection. It has been shown in humans and mice that these responses are mediated by Th2 cytokines: interleukin-4 (IL-4), IL-5, IL-9, and IL-10 (reviewed in references 14 and 32). In fact, mouse CD41 T cells have been shown to be polarized into Th2 cells during nematode infection (reviewed in reference 21). Furthermore, recent studies have revealed that Th2 cytokines, especially IL-4, play a crucial role in determining the outcome of nematode infection. For example, the administration in mice of anti-IL-4 antibody, anti-IL-4 receptor antibody, or IL-12, which suppresses IL-4 production, prolongs nematode infection (7, 39). In rats, despite some observations of remarkable responses in IgE, eosinophils, and mast cells after nematode infection (10, 11, 20, 27), it has not yet been clarified whether intestinal nematode infection generates Th2 or Th2-like cells in vivo. This may partly be due to the lack of reliable assay systems for Th2 cytokines in rats. Diaz-Sanchez et al., using a reverse transcription-PCR (RT-PCR) method, found that a high level of IL-4 mRNA was detected in splenocytes stimulated in vitro with phorbol myristate acetate plus ionomycin from rats immunized with bee venom phospholipase A2 plus ricin, which induced CD81 T-cell depletion and a marked IgE response (3). In another study, levels of both IL-4 and IL-5 mRNAs increased in rat splenocytes cultured with concanavalin A and IL-4 or with concanavalin A and anti-gamma interferon (IFN-g) antibody (25). These results were obtained from studies using rats immunized experimentally and suggest that Th2

or Th2-like cells can also be generated in rat splenocytes under certain conditions. It was also indicated that a T-cell subset collected from thoracic duct lymph of Trichinella spiralis-infected rat showed significant secretion of IL-4 and IL-5 but little IFN-g following stimulation in vitro with T. spiralis antigens, although the levels of expression of IL-2, IL-3, IL-4, IL-5, IL-10, and IFN-g genes increased together (29). Previously, we reported that IgE production in mesenteric lymph nodes (MLN) was significantly higher than that in the spleen in Nippostrongylus brasiliensis-infected rats (37) and that mastocytosis was induced when adult worms were transplanted into the small intestine but not into subcutaneous tissues (1), suggesting that the gut-associated lymphoid tissues, including MLN, are the most important sites for production of cytokines responding to intestinal nematode infection. In the present study, using RT-PCR, we investigated alterations (0 to 6 weeks) in gene expression of some Th1- and Th2-associated cytokines in MLN cells from two inbred rat strains, Brown Norway (BN) and Fischer-344 (F344), which show different IgE responses to allergens. N. brasiliensis infection strongly induced IL-3, IL-4, and IL-5 mRNA expression in both rat strains, which was reflected well in IgE antibody, eosinophil, basophil, and mast cell responses, while expression of IL-2, IL-10, and IFN-g mRNAs showed no notable changes throughout infection. These observations suggest that polarized generation of Th2 or Th2-like cells also occurs in vivo in rats with intestinal nematode infection. MATERIALS AND METHODS Animals and N. brasiliensis infection. Specific-pathogen-free male Brown Norway/Sea (Japan SLC Inc.) and Fischer-344/NS1c (Seiwa Experimental Animals Co.) rats were used at 8 weeks of age. These rats were injected subcutaneously with 2,000 N. brasiliensis larvae as described previously (20). Quantification of serum IgE and IgG2a. After infection, four rats from each strain were bled periodically from the orbital venous plexus. Total serum IgE and IgG2a were measured by enzyme-linked immunosorbent assay (ELISA)

* Corresponding author. Phone: 75-251-5325. Fax: 75-251-5328. Electronic mail address: [email protected]. 4653

4654

MATSUDA ET AL.

INFECT. IMMUN. TABLE 1. Primer sequences for amplification of cytokine and b-actin cDNAs

Cytokine or b-actin

IL-2 IL-3 IL-4 IL-5 IL-10 IFN-g b-Actin

Sense or antisense

Sequence

Bases spanneda

Amplified band (bp)

No. of cycles

Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense

59-CATGTACAGCATGCAGCTCGCATCC-39 59-CCACCACAGTTGCTGGCTCATCATC-39 59-TGCCTTGGAGATTTTGGTGAAGCTCCC-39 59-CAGGTCCTTAAGATGGATCACGTAGA-39 59-ACCTTGCTGTCACCCTGTTCTGC-39 59-GTTGTGAGCGTGGACTCATTCACG-39 59-TGCTTCTGTGCTTGAACGTTCTAAC-39 59-TTCTCTTTTTGTCCGTCAATGTATTTC-39 59-TGCCAAGCCTTGTCAGAAATGATCAAG-39 59-GTATCCAGAGGGTCTTCAGCTTCTCTC-39 59-ATGAGTGCTACACGCCGCGTCTTGG-39 59-GAGTTCATTGACAGCTTTGTGCTGG-39 59-AGAAGAGCTATGAGCTGCCTGACG-39 59-CTTCTGCATCCTGTCAGCGATGC-39

14–38b 423–399b 1514–1540 3660–3635 872–894 2384–2361 516–540 813–787 239–265b 365–339b 1–25c 42–18d 2502–2525 2873–2825

410

31

18, 19

295

38

2, this paper

352

30

25, 30

298

38

25, 38

127

28

8, 24

405

33

4, 23

236

15

25, 26

References

a

Based on genomic DNA sequences unless noted otherwise. Based on cDNA sequences. Numbered from the first nucleotide in the first exon. d Numbered from the first nucleotide in the fourth exon. b c

as described previously (36). Briefly, each well of Immunoplates (Nunc) was coated with 100 ng of anti-rat IgE monoclonal antibody MARE-1 (Zymed, South San Francisco, Calif.) or of anti-rat IgG2a monoclonal antibody MARG2a-1 (Zymed) and then blocked with bovine milk proteins (Block Ace; Dainihon Pharmaceuticals). Serial dilutions of sera were applied, and the plates were incubated for 2 h, then incubated with goat antiserum against rat IgE or IgG2a (1:5,000 dilution; Bethyl Laboratories, Montgomery, Tex.), and finally incubated with affinity-purified biotin-conjugated rabbit IgG against goat IgG (1:1,000 dilution; MBL Inc.). After subsequent incubation with alkaline phosphataseconjugated streptavidin (1:5,000 dilution; Boehringer, Mannheim, Germany), 10 mM 4-nitrophenylphosphate dissolved in diethanolamine buffer (pH 9.8) containing 0.5 mM MgCl2 was added as a substrate. The optical density at 405 nm was measured with a microplate reader (Tosoh). Total IgE and IgG2a levels of test sera were determined by reference to the optical densities of known amounts of IgE and IgG2a myeloma protein, respectively. This assay system was capable of detecting IgE and IgG2a concentrations as low as 1 ng/ml. Eosinophil, basophil, and mast cell counts. After infection, four rats from each strain were bled periodically from the tail vein. The numbers of eosinophils and basophils were counted after staining with Hinkelmann’s solution and alcian blue, respectively, as reported previously (12, 28). To estimate mast cell numbers, animals were sacrificed with an overdose of ether, and a segment of jejunum 20 cm distal to the pyloric sphincter was removed and fixed in Carnoy’s fixative, and the number of mucosal mast cells per villous crypt unit in the jejunal mucosa was determined on the alcian blue-stained sections as described by Miller and Jarrett (20). Preparation of MLN cells. MLN were removed from four rats from each strain on different days after infection and crushed mechanically in Hanks solution containing 5% fetal bovine serum, and cell suspensions were obtained by passage through wire mesh and then nylon mesh with a pore size of 50 mm. After being washed in Hanks solution, 5 3 106 cells were used for RNA extraction. The viability of cells was always above 95%. Isolation of total RNA and synthesis of cDNA. MLN cells obtained from each rat were homogenized in 1 ml of ISOGEN solution (Nippongene), and then RNA was extracted with chloroform and precipitated with ethanol. Four-microgram aliquots of the extracted RNA were reverse transcribed in 20 ml of reverse transcription buffer (50 mM Tris-HCl [pH 8.3], 75 mM KCl, 3 mM MgCl2) containing 0.4 mg of oligo(dT) primer (Pharmacia), 40 U of RNase inhibitor (Toyobo), 1 mM (each) dATP, dTTP, dGTP, and dCTP, 10 mM dithiothreitol, and 100 U of Superscript II reverse transcriptase (Gibco BRL, Gaithersburg, Md.). The reaction mixtures were preheated for 5 min at 708C, chilled quickly on ice, and then incubated for 30 min at 428C. The absence of RNA degradation was confirmed by electrophoresis on 1% agarose gels containing formaldehyde (31). Cytokine cDNA amplification. Aliquots (2.5 ml each) of the synthesized cDNA solution were added to reaction buffer [20 mM Tris-HCl (pH 8.8), 2 mM MgSO4, 10 mM KCl, 10 mM (NH4)2SO4, 0.1% Triton X-100] containing 200 nM (each) dATP, dTTP, dGTP, and dCTP, 0.5 U of Perfect Match DNA polymerase enhancer (Stratagene), 1.25 mM sense and antisense primers, 1 kBq of [a-32P]dCTP, and 1 U of Vent DNA polymerase (New England Biolabs, Beverly, Mass.) in a final volume of 50 ml. The primers used for amplification are shown in Table 1. After initial incubation (5 min at 978C; 2 min at 608C; 1 min at 728C), thermal cycling was begun according to the following profile: 1 min at 978C, 1 min at 608C, and 1 min at 728C in a MiniCycler (MJ Research, Inc., Watertown, Mass.) under the control of an in-sample probe sensor. In a series of experiments, the optimal number of cycles for amplification of each cytokine cDNA

was determined so that the amplified cDNA was well below the plateau level (see Table 1 and Fig. 3C). Three-microliter aliquots of the amplified samples were electrophoresed on 5% polyacrylamide gels, which were then fixed in 10% methanol–10% acetic acid and dried. The amounts of amplified products were determined from the densities (photostimulated luminescence) of the corresponding bands in autoradiograms, using a Bioimage analyzer (Fuji Photo Film), and the results were arbitrarily normalized to the signals of b-actin cDNA, which were amplified in the same tubes for cytokine cDNA amplification. In more detail, PCR was started with primers for a given cytokine species alone, and then the primers for b-actin were added to the reaction tube at the denaturation step (978C) in an appropriate round of the cycling so that the fixed 15 cycles for b-actin and the predetermined optimal cycles (Table 1) for the cytokine could be carried out at the end of the amplification reaction. To compare the patterns of changes in cytokine expression between the two rat strains, each mean value of the amplified signals calculated on each sampling day was converted to fold increase over the preinfection level by dividing by the mean value of four rats on day 0. The specificities of amplified products were confirmed by direct cycle sequencing based on the dideoxy-mediated chain termination method, using a DTth Sequencing PRO kit (Toyobo) after amplified bands were cut out and extracted with glass milk powder (GeneClean II; Bio-101, La Jolla, Calif.). Statistical analysis. Unpaired Student’s t test was used to analyze the significance of differences.

RESULTS Antibody responses in BN and F344 rats following N. brasiliensis infection. IgE antibodies were not detectable (,1 ng/ ml) in naive F344 rats, while the level of IgE in BN rats before infection was 31.3 6 10.6 ng/ml (mean 6 standard error of the mean, four individuals). In both strains, IgE level showed a marked increase and reached a peak level 2 weeks after infection; these levels were 32.5 6 7.1 mg/ml (about 1,000-fold over the preinfection level) in BN rats and 2.5 6 0.5 mg/ml (far greater than 103-fold) in F344 rats (Fig. 1A). Thereafter, the IgE levels decreased slowly but still remained elevated over the baseline even 6 weeks after infection. The changes in IgG2a antibody levels were almost the same in the two strains (Fig. 1B): IgG2a antibody began increasing 2 weeks after infection, but the degree of increase was smaller than for IgE. Cellular responses in BN and F344 rats after N. brasiliensis infection. The number of peripheral eosinophils was much greater in BN than in F344 rats before infection and increased to the highest level 14 days after infection, but the peak values were not significantly different in the two strains (Fig. 2A). In both strains, the number of basophils showed a transient but remarkable increase around 14 days after infection, without significant differences between the two strains (Fig. 2B). The time courses of changes in mast cell numbers in the jejunal

VOL. 63, 1995

CYTOKINE mRNA IN N. BRASILIENSIS-INFECTED RATS

4655

Changes in levels of IL-3, IL-4, and IL-5 mRNAs. All of these cytokine mRNAs showed increases following N. brasiliensis infection in both strains. IL-3 gene expression was strongly elevated by day 7 after infection, remained at high levels until at least 14 days (Fig. 4A and B), and decreased gradually thereafter but did not decline below levels in the preinfection controls even after 6 weeks. The expression of IL-4 mRNA showed a clear peak at day 10 postinfection in BN rats and at day 14 postinfection in F344 rats (Fig. 4C and D). IL-4 mRNA was expressed at a very low level in naive F344 rats, but its expression was amplified markedly, reaching a peak at 14 days. As observed for IL-3 at 6 weeks after infection, IL-4 mRNA levels remained elevated over the baseline. IL-5 gene expression showed similar patterns in the two strains. It peaked around day 14 and declined to nearly the baseline at around day 28 but then showed a further increase (Fig. 4E and F). For all of these cytokines, mRNA levels estimated as a value relative to the amplified b-actin signal after the fixed 15

FIG. 1. Serum IgE (A) and IgG2a (B) levels after N. brasiliensis infection in BN rats (closed circles) and F344 rats (open circles). The data represent the means of four individuals 6 standard errors of the means. The arrows and arrowheads indicate significant differences between BN and F344 rats on the days indicated at P , 0.05 and P , 0.01, respectively.

mucosa were different in the two strains. Mucosal mast cells (MMC) increased remarkably by day 14 after infection in BN rats, while a delayed increase in MMC was observed in F344 rats. Moreover, the plateau level was clearly lower in F344 than in BN rats, and these elevated levels remained long after infection (for at least 6 weeks) (Fig. 2C). It was confirmed that the degrees of the infection were almost the same in the two strains; the maximum levels of eggs per gram index and of worm burden showed no significant differences between the two strains, although the majority of worms were expelled approximately 2 days earlier in BN than in F344 rats (35a). Reliability of cytokine cDNA amplification. The amplification of b-actin cDNA reached saturation rapidly if performed for more than 18 cycles (Fig. 3A), possibly because b-actin mRNA was over 3 orders of magnitude more abundant than any of the cytokine mRNAs. The amplified cDNA signal of b-actin after the fixed 15 cycles, which was used as an internal control throughout the experiments, did not show changes greater than 61.5-fold between the sample tubes. In addition, the expression levels of the b-actin gene did not differ significantly between the two rat strains (an example is shown in Fig. 3B). The 15 PCR cycles with b-actin primers did not inhibit the amplification of cytokine cDNAs in our system (Fig. 3D). As shown in Fig. 3C, the optimal number of cycles for amplification of each cytokine cDNA was determined experimentally.

FIG. 2. Numbers of eosinophils (A) and basophils (B) in peripheral blood per microliter and numbers of mast cells per villous crypt unit (VCU) in the jejunal mucosa (C) after N. brasiliensis infection in BN rats (closed circles) and F344 rats (open circles). The data represent the means of four individuals 6 standard errors of the means. The arrows and arrowheads indicate significant differences between BN and F344 rats on the days indicated at P , 0.05 and P , 0.01, respectively.

4656

MATSUDA ET AL.

INFECT. IMMUN.

FIG. 3. (A) PCR amplification of b-actin cDNA. The template was obtained from reverse-transcribed cDNA from the MLN of BN rats without infection. One microliter of the PCR solution was isolated after the indicated cycles from the same tube and then electrophoresed. After the gel was dried, autoradiograms were scanned with a Bioimage analyzer, and the results are expressed as photostimulated luminescence (PSL) units. (B) Variation of amplified b-actin signals as an internal control between rat strains and between reaction tubes. Two examples are shown: one with primers for b-actin plus IL-4, and the other for b-actin plus IFN-g. The numbers under the abscissa indicate the number of days after infection on which RNA was sampled. (C) Amplification curves of cytokine cDNAs after different PCR cycles. Two examples (IL-3 and IL-5) to determine the optimal numbers of cycles (Table 1; 38 cycles in these examples) of PCR are shown. One microliter of the PCR solution was isolated after the indicated cycles from the same tube, and the amplified signals for each cytokine cDNA were measured as described for panel A. BN0 and BN10 indicate the RNA sources used, which were obtained from naive BN rats or BN rats after 10 days of infection, respectively. (D) The simultaneous amplification of b-actin (fixed 15 cycles) had little effect on the amplification of cytokine cDNA. Three-microliter aliquots of each PCR solution were used for assessment of the cytokine signals after PCR for each cytokine species was carried out for the predetermined number of cycles with or without b-actin primers. The RNA samples were obtained from BN rats 10 days after infection.

PCR cycles, which could be regarded as a reflection of the total amount of transcribed cytokine mRNA, were always higher in BN than in F344 rats throughout the experimental period (Fig. 4A, C, and E), while the degrees of amplitude at any peaks estimated as fold increase over the preinfection control level were higher in F344 rats (Fig. 4B, D, and F). In addition, we investigated gene expression levels of these three cytokines in 0, 12, 24, 48, and 72 h postinfection, and none of them exhibited a twofold or more increase or decrease relative to the preinfection levels (data not shown). Changes in levels of IL-2, IL-10, and IFN-g mRNAs. Levels of all of these cytokine mRNAs did not increase after N. brasiliensis infection. While IL-2, IL-10, and IFN-g gene expression showed a tendency to be suppressed after infection in BN rats, no such tendency was seen in F344 rats except for IL-2 (Fig. 5). The preinfection levels of IL-2 and IL-10 mRNAs in BN rats were about three and two times, respectively, those in F344 rats, but IFN-g mRNA levels were almost the same in both strains on day 0. The gene expression of these three cytokines also showed no significant changes within 3 days of infection, as observed for IL-3, IL-4, and IL-5. DISCUSSION In this study, we investigated mRNA expression patterns of some cytokines in MLN cells of rats infected with N. brasiliensis and whether these patterns showed any differences between two rat strains, BN and F344, which show different IgE responses to nematode infection. The results showed two types of cytokine mRNA expression. One, which included IL-3, IL-4, and IL-5, showed a significant increase in mRNA expression

around 7 to 14 days after infection, while the other showed no notable changes throughout the period of infection; IL-2, IL10, and IFN-g belonged to the latter type. The significant increases in expression of IL-4 and IL-5, the pivotal cytokines of Th2 cells in mice and humans, suggest that primary infection with N. brasiliensis induced Th2 or Th2-like cells in rats also. IL-10, another cytokine classified as Th2 type in mice, did not increase in our rat model. This observation is consistent with a previous report that IL-10 gene expression showed little or no increase in mice infected with Heligmosomoides polygyrus (35), and these findings suggest that the IL-10 gene is regulated differently from the other Th2 cytokine genes. The mRNA expression of IL-2 and IFN-g, Th1-associated cytokines in mice and humans, exhibited a weak depression, if any, in BN rats, but that of IFN-g was rather constant in F344 rats. These results coincide with a previous report of a study using mice (35) and suggest that Th1 or Th1-like cell induction is weak, if it occurs at all, only after intestinal nematode infection in rats. Ramaswamy et al. (29) reported that a certain subset of T cells in rat thoracic duct lymph 3 days after T. spiralis infection showed a secreted cytokine profile resembling Th2 or Th2-like cells when they were properly stimulated in vitro with the parasite antigens, whereas this stimulation increased the concentration of message for all cytokines tested (IL-2, IL-3, IL-4, IL-5, IL-10, and IFN-g). They speculated that the mRNA profile resembles that of Th0 cells more than that of Th1 or Th2 cells and that these cells have not reached their full ability to elaborate the cytokines. In the present study, MLN cells did not show the mRNA profile of Th0 cells even in the early period of infection. This may be due to the difference in lymphocyte populations in thoracic duct lymph and MLN cells or

VOL. 63, 1995

CYTOKINE mRNA IN N. BRASILIENSIS-INFECTED RATS

4657

FIG. 4. Changes in mRNA expression levels of IL-3 (A and B), IL-4 (C and D), and IL-5 (E and F). All data were individually normalized by dividing by the value for b-actin cDNA amplified after the fixed 15 cycles. (A, C, and E) Means of four individuals 6 standard errors of the means. The ordinates represent arbitrary values based on normalization. These normalized values could be regarded as a reflection of the absolute amounts of each cytokine mRNA (because the b-actin expression levels did not differ greatly between individual experiments or between strains, as shown in Fig. 3B). (B, D, and F) Fold increases in level of each cytokine mRNA over the preinfection level. For this determination, mean normalized values were divided by the mean values for day 0, arbitrarily given a value of 1, for each rat strain. The arrows and arrowheads indicate significant differences between BN and F344 rats on the days indicated at P , 0.05 and P , 0.01, respectively. Closed circles, BN rats; open circles, F344 rats.

may result from the difference in our method from that of Ramaswamy et al.; we did not stimulate MLN cells with antigens or mitogens in vitro, and we investigated the amount of cytokine mRNAs in MLN cells at the moment the lymph nodes were removed. Concerning the dissociation between gene expression and protein secretion, we also observed dissociation between the levels of IFN-g mRNA detected by RT-PCR and secreted IFN-g; IFN-g measured by an ELISA decreased significantly 2 weeks after infection (35a), while the levels of the mRNA exhibited no notable changes, as shown in Fig. 5. Certainly, there are some posttranscriptional regulation mechanisms in some cytokines, and such unknown regulation mechanisms may play an important role in cellular interactions in which cytokines are key mediators. Gene expression showed

no significant changes in any cytokines tested within 3 days of infection. Since N. brasiliensis larvae migrate to the lungs within 2 to 3 days of infection before they reach their final habitat, the small intestine, this observation indicates that gene expression in these cytokines is not activated in MLN cells during the lung stage of N. brasiliensis infection. In H. polygyrus infection in mice, in which larvae inoculated orally reach the intestine directly and immediately invade the mucosa, IL-3, IL-5, and IL-9 but not IL-4 messages were increased in Peyer’s patches and MLN by 12 h after inoculation (35). These data suggest that the activation of cytokine gene expression in localized tissues is greatly influenced by the mode of infection. The differences between the BN and F344 strains were originally characterized by the responses to immunization with

4658

MATSUDA ET AL.

INFECT. IMMUN.

FIG. 5. Changes in mRNA expression levels for IL-2 (A and B), IL-10 (C and D), and IFN-g (E and F). Values normalized relative to that for amplified b-actin cDNA (A, C, and E) and fold increases over preinfection levels (B, D, and F) are shown as described in the legend to Fig. 4. The arrows and arrowheads indicate significant differences between BN and F344 rats on the days indicated at P , 0.05 and P , 0.01, respectively. Closed circles, BN rats; open circles, F344 rats.

ovalbumin in Al(OH)3: BN rats produced a high, sustained level of specific IgE against ovalbumin, while that in F344 rats was significantly lower and transient (22). In the present study, the total IgE level before infection was 31.3 6 10.6 ng/ml in BN rats, while it was ,1 ng/ml in F344 rats. N. brasiliensis infection induced marked IgE antibody production in both strains, and the peak IgE level in BN rats was more than 13 times higher than that in F344 rats. We also examined N. brasiliensis-specific IgE responses in these two strains; specific IgE was detectable in BN rats at 2 weeks of infection and thereafter, whereas it was not detectable in F344 rats throughout the observation period (35a). The fold increase of total IgE over the preinfection level was, however, higher in F344 than in BN rats, suggesting that at least total IgE responsiveness is not defective in F344 rats. IL-4 is a strong inducer of IgE production, while IFN-g has an inhibitory effect (5, 6, 33). IL-4 gene expression was remarkably intensified in both BN and F344 rats, although

peak expression occurred a little later in F344 rats. On the other hand, the changes in IFN-g gene expression were only slight. The magnitude of increase in IL-4 mRNA expression as shown by fold increase over the preinfection level was larger in F344 rats (Fig. 4D), although this might be an overestimation resulting from the very low expression levels in preinfection controls of F344 rats (Fig. 4C). In both strains, IL-4 gene expression did not decline to the baseline even 6 weeks after infection, which corresponded to the high IgE levels sustained until late after infection. These findings show that the elevation of IgE induced by N. brasiliensis infection was controlled mainly by IL-4, at least as far as transcriptional events are concerned. In contrast, total IgG2a began increasing slowly 2 weeks after infection, but our findings did not reveal a definitive relationship between IgG2a response and the changes in IL-4 and/or IFN-g gene expression. Although the detection of mRNA by RT-PCR does not

VOL. 63, 1995

CYTOKINE mRNA IN N. BRASILIENSIS-INFECTED RATS

allow strict quantification, this method has been reported to be reproducible and applicable to observe the changes in gene expression as long as experiments are well controlled (15, 34). Under our experimental conditions, in which the levels of expression of the b-actin gene were not very different between BN and F344 rats (Fig. 3B), IL-4 gene expression relative to that of b-actin was higher in BN than in F344 rats not only after but also before infection (Fig. 4C). Thus, although the temporal patterns of IL-4 gene expression did not differ significantly between the two strains, the amount of IL-4 mRNA expressed at each time point was probably greater in BN than in F344 rats. To confirm this observation, more strictly quantitative studies should be carried out; it is becoming clear that there is a strong correlation between the data obtained by the present system based on RT-PCR and those obtained by a quantitative RNase protection assay in our laboratory. The time courses of changes in MMC were also considerably different in the two strains. The MMC number reached a plateau at least 1 week earlier in BN rats than in F344 rats, and the level remained more than twice as high in BN as in F344 rats, although the numbers before infection were almost the same. IL-3, IL-4, IL-9, and IL-10 are known to have mast cell growth and differentiation activities. When bone marrow cells of mice are cultured in the presence of IL-3, mast cells continue proliferating after differentiation. The addition of IL-4 or IL-10 augments the proliferation of mast cells, but either added alone has only trivial effects on cultured mast cells (reviewed in reference 16). IL-9, which has been shown to have mast cell growth-enhancing activity (9), seems to play a more important costimulatory role than IL-10 in the induction of mastocytosis in mice infected with N. brasiliensis, since IL-9 mRNA was increased after infection (7) whereas IL-10 gene expression was relatively constant (17). The results of the present study showed that the level of IL-3 mRNA increased similarly and peaked 7 days after infection in both strains. On the other hand, peak IL-4 gene expression was at least 4 days earlier in BN than in F344 rats. Thus, the time lag observed in the increase in number of MMC may have been related in some way to the timing of maximal increase in mRNA expression of IL-4. The changes in levels of IL-9 mRNA could not be investigated in this study because the molecule has not been cloned yet in rat species, and it is possible that there are some differences between the two rat strains in IL-9 gene expression after infection, which may cause the difference observed in levels of MMC. Besides having a role in inducing differentiation and proliferation of eosinophils, IL-5 has been reported to sustain the survival of matured eosinophils and to also be an eosinophil chemotactic factor (40, 41). The maximum increase in peripheral eosinophils occurred around day 14 after infection in both strains, and this coincided well with the time course of changes in IL-5 mRNA, which exhibited similar patterns in the two strains. Basophils, which are regulated by IL-3 and IL-4, increased transiently around 14 days after infection, showing similar time courses in the two strains, and this followed the peaks in expression of IL-3 and IL-4 mRNAs. Contrary to the gut mast cell responses, strain differences in eosinophil and basophil responses were not observed. It has been shown that bone marrow-derived mast cell precursors invade the intestine from blood and proliferate in the tissue site in response to locally produced cytokines (13). Eosinophils and basophils, however, complete their proliferation and differentiation in the bone marrow and then enter circulating blood. Thus, it is suggested that the numbers of eosinophils and basophils in the peripheral blood are regulated by more factors engaged in the

4659

events in the marrow than those contributing to mucosal mastocytosis. Taken together, the present results indicate that IgE, eosinophil, basophil, and mast cell responses reflect expression of the Th2-type cytokines IL-4 and IL-5 and the Th1/Th2-type cytokine IL-3, which are induced strongly by N. brasiliensis infection in rat MLN cells. These findings suggest that polarized generation of Th2 or Th2-like cells also occurs in vivo in rats following an intestinal nematode (N. brasiliensis) infection. We are now trying to define the cellular sources of the Th2type cytokines by separating CD41 and CD81 T-cell populations and also to measure IL-12 gene expression to study the regulation of generation of Th1 and Th2 (or Th1- or Th2-like) cells in detail in rats infected with the nematode. ACKNOWLEDGMENTS This work was supported by the grants 06770190 and 06670266 from the Ministry of Education, Science, and Culture of Japan, the Tray Research Institute, the Shionogi Research Institute, and the Kyoto Medical Science Laboratory. REFERENCES 1. Arizono, N., M. Yamada, T. Tegoshi, M. Okada, R. Uchikawa, and S. Matsuda. 1994. Mucosal mast cell proliferation following normal and heterotopic infection of the nematode Nippostrongylus brasiliensis in rats. Acta Pathol. Microbiol. Immunol. Scand. 102:589–596. 2. Cohen, D. R., A. J. Hapel, and I. G. Young. 1986. Cloning and expression of the rat interleukin-3 gene. Nucleic Acids Res. 14:3641–3658. 3. Diaz-Sanchez, D., A. Noble, D. Z. Staynov, T. H. Lee, and D. M. Kemeny. 1993. Elimination of IgE regulatory rat CD81 T cells in vivo differentially modulates interleukin-4 and interferon-g but not interleukin-2 production by splenic T cells. Immunology 78:513–519. 4. Dijkema, R., P. H. van der Meide, P. H. Pouwels, M. Caspers, M. Dubbeld, and H. Schellekens. 1985. Cloning and expression of the chromosomal immune interferon gene of the rat. EMBO J. 4:761–767. 5. Finkelman, F. D., I. M. Katona, T. R. Mosmann, and R. L. Coffman. 1988. IFN-g regulates the isotypes of Ig secreted during in vivo humoral immune responses. J. Immunol. 140:1022–1027. 6. Finkelman, F. D., I. M. Katona, J. F. Urban, Jr., J. Holmes, J. Ohara, A. S. Tung, J. G. Sample, and W. E. Paul. 1988. IL-4 is required to generate and sustain in vivo IgE responses. J. Immunol. 141:2335–2341. 7. Finkelman, F. D., K. B. Madden, A. W. Cheever, I. M. Katona, S. C. Morris, M. K. Gately, B. R. Hubbard, W. C. Gause, and J. F. Urban, Jr. 1994. Effects of interleukin 12 on immune responses and host protection in mice infected with intestinal nematode parasites. J. Exp. Med. 179:1563–1572. 8. Goodman, R. E., J. Oblak, and R. G. Bell. 1992. Synthesis and characterization of rat interleukin-10 (IL-10) cDNA clones from the RNA of cultured OX82, OX222 thoracic duct T cells. Biochem. Biophys. Res. Commun. 189:1–7. 9. Hu ¨ltner, L., C. Druez, J. Moeller, C. Uyttenhove, E. Schmitt, E. Ru ¨de, P. Do ¨rmer, and J. van Snick. 1990. Mast cell growth-enhancing activity (MEA) is structurally related and functionally identical to the novel mouse T cell growth factor P40/TCGFIII (interleukin 9). Eur. J. Immunol. 20:1413–1416. 10. Jarrett, E. E. E., and H. Bazin. 1974. Elevation of total serum IgE in rats following helminth parasite infection. Nature (London) 251:613–614. 11. Jarrett, E. E. E., D. M. Haig, and H. Bazin. 1976. Time course studies on rat IgE production in N. brasiliensis infection. Clin. Exp. Immunol. 24:346–351. 12. Kasugai, T., M. Okada, M. Morimoto, N. Arizono, K. Maeyama, M. Yamada, H. Tei, K. Dohmae, H. Onoue, G. F. J. Newlands, T. Watanabe, Y. Nishimune, H. R. P. Miller, and Y. Kitamura. 1993. Infection of Nippostrongylus brasiliensis induces normal increase of basophils in mast cell-deficient Ws/Ws rats with a small deletion at the kinase domain of c-kit. Blood 81:2521–2529. 13. Kasugai, T., H. Tei, M. Okada, S. Hirota, M. Morimoto, M. Yamada, A. Nakahata, N. Arizono, and Y. Kitamura. 1995. Infection with Nippostrongylus brasiliensis induces invasion of mast cell precursors from peripheral blood to small intestine. Blood 85:1334–1340. 14. King, C. L., and T. B. Nutman. 1992. Biological role of helper T-cell subsets in helminth infections. Chem. Immunol. 54:136–165. 15. Kinoshita, T., J. Imamura, H. Nagai, and K. Shimotohno. 1992. Quantification of gene expression over a wide range by the polymerase chain reaction. Anal. Biochem. 206:231–235. 16. Kitamura, Y., T. Kasugai, N. Arizono, and H. Matsuda. 1993. Development of mast cells and basophils: processes and regulation mechanisms. Am. J. Med. Sci. 306:185–191. 17. Madden, K. B., J. F. Urban, Jr., F. D. Finkelman, W. C. Gause, and I. M. Katona. 1992. Cytokine regulation of helminth-induced intestinal mastocytosis. FASEB J. 6:A1723.

4660

MATSUDA ET AL.

18. McKnight, A. J., A. N. Barclay, and D. W. Mason. 1991. Molecular cloning of rat interleukin 4 cDNA and analysis of the cytokine repertoire of subsets of CD41 T cells. Eur. J. Immunol. 21:1187–1194. 19. McKnight, A. J., D. W. Mason, and A. N. Barclay. 1989. Sequence of rat interleukin 2 and anomalous binding of a mouse interleukin 2 cDNA probe to rat MHC class II-associated invariant chain mRNA. Immunogenetics 30:145–147. 20. Miller, H. R. P., and W. F. H. Jarrett. 1971. Immune reactions in mucous membranes. I. Intestinal mast cell response during helminth expulsion in the rat. Immunology 20:277–288. 21. Mosmann, T. R., and R. L. Coffman. 1989. Th1 and Th2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 7:145–173. 22. Murphey, S. M., S. Brown, N. Miklos, and P. Fireman. 1974. Reagin synthesis in inbred strains of rats. Immunology 27:245–253. 23. Nagai, Y., M. Inobe, K. Kikuchi, and T. Uebe. 1993. Functional and phenotypical analysis of subsets of rat CD41 T cells. Microbiol. Immunol. 37:623– 632. 24. Noble, A., D. Z. Staynov, D. Diaz-Sanchez, T. H. Lee, and D. M. Kemeny. 1993. Elimination of IgE regulatory rat CD81 T cells in vivo increases the co-ordinate expression of Th2 cytokines IL-4, IL-5 and IL-10. Immunology 80:326–329. 25. Noble, A., D. Z. Staynov, and D. M. Kemeny. 1993. Generation of rat Th2-like cells in vitro is interleukin-4-dependent and inhibited by interferon-g. Immunology 79:562–567. 26. Nudel, U., R. Zakut, M. Shani, S. Neuman, Z. Levy, and D. Yaffe. 1983. The nucleotide sequence of the rat cytoplasmic b-actin gene. Nucleic Acids Res. 11:1759–1771. 27. Ogilvie, B. M., P. M. Hesketh, and M. E. Rose. 1978. Nippostrongylus brasiliensis: peripheral blood leukocyte response of rats, with special reference to basophils. Exp. Parasitol. 46:20–30. 28. Platt, W. R. 1979. Color atlas and textbook of hematology, 2nd ed., p. 170. Lippincott, Philadelphia. 29. Ramaswamy, K., R. E. Goodman, and R. G. Bell. 1994. Cytokine profile of protective anti-Trichinella spiralis CD41 OX222 and non-protective CD41 OX221 thoracic duct cells in rats: secretion of IL-4 alone does not determine protective capacity. Parasite Immunol. 16:435–445. 30. Richter, G., T. Blankenstein, and T. Diamantstein. 1990. Evolutionary aspects, structure and expression of the rat interleukin 4 gene. Cytokine 2:221– 228. 31. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed., p. 7.43–7.45. Cold Spring Harbor Laboratory,

Editor: J. M. Mansfield

INFECT. IMMUN. Cold Spring Harbor, N.Y. 32. Sher, A., and R. L. Coffman. 1992. Regulation of immunity to parasites by T cells and T cell-derived cytokines. Annu. Rev. Immunol. 10:385–409. 33. Snapper, C. M., and W. E. Paul. 1987. Interferon-g and B cell stimulatory factor-1 reciprocally regulate Ig isotype production. Science 236:944–947. 34. Staynov, D. Z., and T. H. Lee. 1992. Expression of interleukin-5 and granulocyte-macrophage colony-stimulating factor in human peripheral blood mononuclear cells after activation with phorbol myristate acetate. Immunology 75:196–201. 35. Svetic, A., K. B. Madden, X. di Zhou, P. Lu, I. M. Katona, F. D. Finkelman, J. F. Urban, Jr., and W. Gause. 1993. A primary intestinal helminthic infection rapidly induces a gut-associated elevation of Th2-associated cytokines and IL-3. J. Immunol. 150:3434–3441. 35a.Uchikawa, R., M. Yamada, S. Matsuda, T. Tegoshi, M. Nishida, I. Kamata, A. Kuroda, and N. Arizono. Dissociation of early and late protective immunity to the nematode Nippostrongylus brasiliensis in Brown Norway and Fischer-344 rats. Parasitology, in press. 36. Uchikawa, R., M. Yamada, S. Matsuda, and N. Arizono. 1993. IgE antibody responses induced by transplantation of the nematode Nippostrongylus brasiliensis in rats: a possible role of nematode excretory-secretory product in IgE production. Immunology 80:541–545. 37. Uchikawa, R., M. Yamada, S. Matsuda, A. Kuroda, and N. Arizono. 1994. IgE antibody production is associated with suppressed interferon-g levels in mesenteric lymph nodes of rats infected with the nematode Nippostrongylus brasiliensis. Immunology 82:427–432. 38. Ueberla, K. T., W. Q. Li, Z. H. Qin, G. Richter, T. Raabe, T. Diamantstein, and T. Blankenstein. 1991. The rat interleukin-5 gene: characterization and expression by retroviral gene transfer and polymerase chain reaction. Cytokine 3:72–81. 39. Urban, J. F., Jr., I. M. Katona, W. E. Paul, and F. D. Finkelman. 1991. Interleukin 4 is important in protective immunity to a gastrointestinal nematode infection in mice. Proc. Natl. Acad. Sci. USA 88:5513–5517. 40. Yamaguchi, Y., Y. Hayashi, Y. Sugama, Y. Miura, T. Kasahara, S. Kitamura, M. Torisu, S. Mita, A. Tominaga, K. Takatsu, and T. Suda. 1988. Highly purified murine interleukin 5 (IL-5) stimulates eosinophil function and prolongs in vitro survival. IL-5 as an eosinophil chemotactic factor. J. Exp. Med. 167:1737–1742. 41. Yamaguchi, Y., T. Suda, J. Suda, M. Eguchi, Y. Miura, N. Harada, A. Tominaga, and K. Takatsu. 1988. Purified interleukin 5 supports the terminal differentiation and proliferation of murine eosinophilic precursors. J. Exp. Med. 167:43–56.