A mutant strain with defective thymic selection of the Long-Evans Cinnamon (LEC) rat was found to spontaneously develop inflammatory bowel disease (IBD)-like colitis. The secretion of Th1-type cytokines including IFN-γ and IL-2 from T cells of mesenteric lymph node cells (MLNs) and lamina propria mononuclear cells, but not spleen cells, in LEC rats was significantly increased more than that of the control Long-Evans Agouti rats through up-regulated expression of T-bet and phosphorylation of STAT-1 leading to NF-κB activation. In addition, the number of CD4+CD25+Foxp3+ regulatory T (Treg) cells of the thymus, MLNs, and lamina propria mononuclear cells from LEC rats was significantly reduced, comparing with that of the control rats. Moreover, bone marrow cell transfer from LEC rats into irradiated control rats revealed significantly reduced CD25+Foxp3+ Treg cells in thymus, spleen, and MLNs compared with those from control rats. Indeed, adoptive transfer with T cells of MLNs, not spleen cells, from LEC rats into SCID mice resulted in the development of inflammatory lesions resembling the IBD-like lesions observed in LEC rats. These results indicate that the dysfunction of the regulatory system controlled by Treg cells may play a crucial role in the development of IBD-like lesions through up-regulated T-bet, STAT-1, and NF-κB activation of peripheral T cells in LEC rats.
Inflammatory bowel disease (IBD)3 in humans has two manifestations including Crohn’s disease and ulcerative colitis (1, 2, 3). Although the precise mechanisms of the diseases are unknown, it has been reported that activation of the intestinal immune system in response to bacterial Ags with pathologic cytokine production of intestinal T cells through various transcription factors or signal molecules such as T-bet, GATA-3, and STATs plays a key role in the pathogenesis of IBD (4, 5, 6, 7). The cytokine production by T cells is known to initiate and develop chronic intestinal inflammation (8, 9, 10). Crohn’s disease is associated with Th1 cytokines such as IFN-γ and TNF-α (11, 12). Meanwhile, ulcerative colitis in human is associated with Th2 cytokine such as IL-5 (13). In addition, it is suggested that Th3 cytokine such as TGF-β has the immunosuppressive effect of IBD in human and animal models (14, 15). Furthermore, it is reported that nucleotide-binding oligomerization domain-containing proteins expressed in intestinal epithelial cells or APCs play a crucial role in the Th1 response in Crohn’s disease (16, 17, 18, 19). However, the pathogenesis based on cytokine balance of peripheral T cells for IBD is still unclear.
CD4+CD25+ regulatory T (Treg) cells have been widely studied in controlling inflammatory diseases including IBD (20, 21, 22, 23). It is well known that the IBD model, by transfer of naive CD45RBhighCD4+ T cells into T cell-deficient mice, can be controlled by coinjection of Treg cells to suppress CD4+ effector T cell functions such as IFN-γ production (24). It has been reported that the cell number of Treg in the periphery from IBD patients was significantly reduced, or soluble IL-2Rα (CD25) of sera from IBD patients was increased (25, 26). However, it is obscure how natural Treg cells generated from thymus influence the pathogenesis of IBD.
Many animal models—such as C3H/HeJBir, IL-2-deficient, IL-2R-deficient, and IL-10-deficient mice, and HLA-B27/β2 transgenic rats—are known to be referred to as human IBD (27, 28, 29, 30, 31). It has been reported that the disease is induced by the hyperresponsiveness of T cells which is shifted to Th1 or Th2 cytokine production in each model (11, 12, 13). Cytokines produced by lamina propria (LP) CD4+ T lymphocytes appear to initiate and perpetuate chronic intestinal inflammation. It is also reported that Th1 or Th2 cytokine production in IBD models can be modulated by the immunosuppressive cytokine such as TGF-β secreted by Th3 cells (22, 23). In addition, some transcription factors including T-bet and GATA-3 or signal molecules such as STATs are well known to regulate immunopathogenic or immunosuppressive cytokine production for controlling IBD (7).
The Long-Evans Cinnamon (LEC) rat was first described to be a naturally occurring mutant with a specific defect in thymocyte development, which has contained T cell differentiation arrest from CD4+CD8+ double-positive (DP) to CD4+CD8− single-positive (SP) but not to CD4−CD8+ SP thymocytes (32). In addition, peripheral CD4+ T cells of LEC rats were shown not to function as Th cells in Ab production against T cell-dependent Ag and IL-2 production by polyclonal stimulation (32, 33). However, it remains unclear whether the T cell dysfunction of LEC rats might have an influence on any disease associated with the immune disorder. In the present study, we found for the first time that IBD-like lesions developed spontaneously in LEC rats. Thus, the pathogenesis and molecular mechanisms of inflammatory bowel lesions in LEC rats as a human IBD model were analyzed.
Materials and Methods
LEC rats and LEA rats as controls were maintained in our laboratory under specific pathogen-free conditions. SCID mice (C.B-17/lcr-scid/scid Jcl) were purchased from CLEA Japan for use as recipients of adoptive transfer. The experiments were approved by an animal ethics board of Tokushima University.
All organs of rats and recipient mice were removed and fixed in 10% phosphate-buffered formaldehyde (pH 7.2), embedded in paraffin, and prepared for histological estimation. The sections were stained with H&E. The disease incidence was evaluated by three independent, well-trained pathologists in a blinded manner.
Flow cytometric analysis
Surface markers on lymphocytes from thymus, spleen, mesenteric lymph node cells (MLNs), and LP were analyzed by using an EPICS flow cytometer (Beckman Coulter) using FITC-, PE-Cy5-, or PE-conjugated anti-CD4, -CD8, -CD44, -CD62L, and -CD25 mAbs (BD Biosciences). To evaluate intracellular Foxp3 expression, lymphocytes were stained with anti-CD4 and -CD25 mAb, fixed and permeabilized with the buffers of a Foxp3 detection kit (BD Biosciences). After staining with anti-Foxp3 mAb, the lymphocytes were analyzed by flow cytometry.
T cells (>90%) were enriched from single-cell suspensions of spleen and lymph node cells from LEC, control rats, and SCID mice with nylon wool (Wako Pure Chemical), and cultured in 96-well flat-bottom microtiter plates (5 × 104 cells/well) in RPMI 1640 containing 10% FCS, penicillin/streptomycin, and 2-ME. Cells were stimulated with plate-coated anti-CD3 (BD Biosciences) and anti-CD28 mAb (BD Biosciences). [3H]Thymidine incorporation during the last 12 h of the culture for 72 h was evaluated using an automated β liquid scintillation counter. In addition, to detect cell proliferation of the T cell subset, after CFSE (Molecular Probes)-labeled T cells were cultured for 48 h, the T cells were stained with anti-CD4 and -CD8 mAb. Cell divisions of CD4+- or CD8+-gated T cells were analyzed by flow cytometry. For T cell suppression assay, CD25−CD4+ or CD25+CD4+ T cells of MLN from control or LEC rats were enriched using biotin-conjugated anti-CD25 mAb, anti-CD4 mAb (BD Biosciences), magnetic beads, and the CELLection Biotin Binder kit (Dynal Biotech). A total of 5 × 104 CD25−CD4+ T cells from control rats were stimulated with plate-coated anti-CD3 mAb (0.5 μg/ml) for 72 h together with 0, 0.625, 1.25, and 2.5 × 104 CD25+CD4+ T cells from control or LEC rats. [3H]Thymidine incorporation during the last 12 h of the culture was evaluated using an automated β liquid scintillation counter.
Cytokine production was tested by a solid-phase sandwich ELISA using a rat IL-2, IL-4, IFN-γ, and IL-10 kit (BioSource International). In brief, culture supernatants from T cells from lymph nodes or spleen stimulated with anti-CD3 and -CD28 mAbs for 24 h and were added to microtiter plates precoated with an each Ab specific for rat L-2, IL-4, IFN-γ, and IL-10. The biotinylated Ab was added and the plate was incubated for 2 h at room temperature. After washing, streptavidin-HRP solution was added to each well and the plate was incubated for 30 min at room temperature. Finally, stabilized chromogen substrate was added to each well, and the absorbance of each well was read at 450 nm using an automatic microplate reader (Bio-Rad). The concentrations of cytokines were obtained according to the standard curves.
Western blot analysis
Isolated T or CD4+ T cells from spleen and lymph nodes using nylon wool (Wako Pure Chemical) or PE-conjugated CD4 mAb with a magnetic PE selection kit (StemCell Technologies) were stimulated with anti-CD3/-CD28 mAbs for 12∼24 h, washed, pelleted, and incubated in 20 mM/L Tris-HCl (pH 8.0), 20 mM/L NaCl, 0.5% Triton X-100, 5 mM/L EDTA, and 3 mM/L MgCl2 lysis buffer including protease inhibitor mixture (Sigma-Aldrich). After centrifugation for 20 min at 12,000 rpm, supernatant was extracted and used as whole cell lysates. The nuclear extracts were purified using a Nuclear/Cytosol Fraction kit (BioVision). A total of 5∼20 μg of each sample per well was applied for each well and electrophoresed in 10% SDS-polyacrylamide gel. Then, the protein was electrophoretically transferred to polyvinylidene difluoride membrane. The membrane was incubated with anti-T-bet, -GATA-3, -STAT-1, -phospho-STAT1, -GAPDH, -histone (Santa Cruz Biotechnology), and -Foxp3 (eBioscience) as the primary Abs. HRP-conjugated rabbit or mouse IgG was used as the second Ab. Protein binding was visualized with ECL Western blotting reagent (Amersham Biosciences). To quantify the protein expression, the chemiluminescence image was analyzed by ChemiDoc XRS (Bio-Rad).
Bone marrow cells (5 × 106) from LEC and control rats were transferred i.v. into irradiated (4 Gy) control rats. Four weeks after the transfer, the host rats were analyzed. As for adoptive transfer to induce IBD-like lesions, purified T cells (5 × 106) from MLNs or spleen cells of LEC and control rats with nylon wool (Wako Pure Chemical) as donors were used and transferred i.p. into SCID mice. Six weeks after the transfer, the recipient SCID mice were sacrificed and all organs were histologically analyzed. In addition, MLN cells of SCID mice were cultured, and cytokine productions of the supernatants of the cells were analyzed by ELISA.
Injection of neutralizing Abs
A total of 50 μg of anti-rat IFN-γ mAb (PBL Biomedical Laboratories), anti-rat IL-4 mAb (R&D Systems), anti-rat CD8 mAb (BD Biosciences), and isotype control Ab were injected i.v. twice a week into LEC rats from 8 to 10 wk of age. Colon sections were stained with H&E and periodic acid Schiff (PAS), and the pathology was scored blindly using a semiquantitative scale of 0–5 as described previously (34). In summary, grade 0 was assigned when no changes were observed; grade 1, minimal inflammatory infiltrates present in the LP with or without mild mucosal hyperplasia; grade 2, mild inflammation in the LP with occasional extension into the submucosa, focal erosions, minimal to mild mucosal hyperplasia and minimal to moderate micin depletion; grade 3, mild to moderate inflammation in the LP and submucosa occasionally transmural with ulceration and moderate mucosal hyperplasia and mucin depletion; grade 4, marked inflammatory infiltrates commonly transmural with ulceration, marked mucosal hyperplasia and mucin depletion, and multifocal crypt necrosis; grade 5, marked transmural inflammation with ulceration, widespread crypt necrosis, and loss of intestinal glands.
Immunofluorescence staining and confocal microscopic analysis
Frozen sections of colon from LEC and control rats were fixed with 3% paraformaldehyde in PBS, and preblocked with 1% BSA-2.5% FCS in PBS for 1 h. Sections were stained with 1 μg/ml of primary Abs against CD4, CD8, CD45R, CD11b/c (BD Biosciences), T-bet, NF-κB, GATA-3, and phospho-STAT1 (Santa Cruz Biotechnology) for 1 h. After three washes in PBS, the sections were stained with Alexa Fluor 488 donkey anti-mouse IgG (H+L) or goat anti-rabbit IgG (H+L) (Molecular Probe) as the second Abs for 30 min and washed with PBS. The section stained with intracellular proteins were stained with PE-labeled anti-CD4 mAb. The nuclei was stained with 4′,6-diamidino-2-phenylindole (DAPI). The sections were visualized with a laser scanning confocal microscope (Carl Zeiss). A 63 × 1.4 oil differential interference contrast objective lens was used. Quick Operation Version 3.2 (Carl Zeiss) for imaging acquisition and Adobe Photoshop CS2 (Adobe Systems) for image processing were used.
Real-time quantitative RT-PCR
Total RNA was extracted from purified lymphocytes of the thymus, spleen, and MLNs in LEC and control rats using Isogen (Wako Pure Chemical), and was reverse transcribed as described (35). Transcript levels of Foxp3 and hypoxanthine phosphoribosyltransferase (HPRT) were performed using PTC-200 DNA Engine Cycler (MJ Research) with SYBR Premix Ex Tag (Takara). The primer sequences were as follows: Foxp3: forward, 5′-CCCAGGAAAGACAGCAACCTT-3′ and reverse, 5′-CTGCTTGGCAGTGCTTGAGAA-3′; and HPRT: forward, 5′-TGTTGGATACAGGCCAGACTTTGT-3′ and reverse, 5′-TCCACTTTCGCTGATGACACA-3′. Results were calculated by a software of DNA Engine Opicon System (Roche Molecular System).
The Student t test was used for statistical analysis. Values of p > 0.05 were considered significant.
Pathology of colon lesion of LEC rats
Histopathological analysis of all organs of LEC rats from 4 to 12 wk of age was performed. Thickened LP of the colon from LEC rats was observed relative to that from control LEA rats (Fig. 1,A). Ulcer formation with fibrin exudates, mononuclear cell, and neutrophil infiltration was observed in the surface of the lesion (Fig. 1,B, upper). In the thickened LP of LEC rats, mononuclear cell infiltration was prominent (Fig. 1,B, lower). Most severe inflammatory lesions were seen in the tissue around the cecum in LEC rats. Although the precise mechanisms as to why the most severe lesion developed in the tissue around the cecum of LEC rats is unclear, it is possible that the differential bacterial distribution from the other segments of colon or unique mucosal immunity of cecum may influence the cecum lesion of LEC rats. No inflammatory lesions in LEC rats were observed in the other organs in general. The colon lesions developed spontaneously in almost 100% of LEC rats at 8 wk of age or more, and the incidence of lesions was consistently increased with aging (Fig. 1 C). There was no sex difference in the colon lesions of LEC rats. LEC rats have also been reported as an animal model for Wilson’s disease because of the genetic copper metabolism disorder (36). It is known that the body weight of LEC rats is significantly reduced compared with that of control rats because of the disease (37). Therefore, it is unclear whether the IBD-like lesions of the LEC rat may influence the loss of body weight.
To clarify the population of the immune cells infiltrated in the LP of IBD-like lesions from LEC rats, immunohistochemical analysis was performed using the cecum sections from LEC rats. We found that CD4+ T cells and CD11b/c+ macrophages or dendritic cells were mainly infiltrated, and a small number of CD45R+ B and CD8+ T cells were also found in the lesions of LEC rats (Fig. 1,D). The total cell number of lymphocytes in the LP of cecum among colon from LEC rats was significantly higher than that from control rats (Fig. 1,E). However, there was no difference in number of LP mononuclear cells (LPMCs) in proximal and distal colon between control and LEC rats (Fig. 1,E). Moreover, flow cytometric analysis with T lymphocytes in the LP of the lesions revealed that the prominent population was CD4+ T cells compared with those of control rats (Fig. 1 F).
Characteristics of T cells from LEC rats
To examine the immunological characteristics in LEC rats, the thymocytes and peripheral T cells were analyzed. We found that the cell number of thymus and spleen from LEC rats was significantly lower than that from control rats, while the cell number of MLNs was significantly higher compared with that from control rats (Fig. 2,A). Significant reduction of CD4+CD8− cells of the thymus and CD4+ T cells of the spleen from LEC rats was observed in contrast to the populations from control rats as described in the previous report (Fig. 2,B) (32). In contrast, there was a greater increase of CD4+ T cell number of MLNs found in LEC rats compared with that from control rats (Fig. 2,B). No significant change for CD8+ T cells was observed in the thymus, spleen, and MLNs (Fig. 2,B). In addition, CD44 expression, one of the activation markers for T cells, on CD4+ T cells of MLNs and LP from LEC rats was significantly higher than that from control rats (Fig. 2,C). Also, CD62L−CD4+ T cells, which are memory phenotype, of MLNs and LP from LEC rats were significantly increased compared with those from control rats (Fig. 2 C). As for CD69 expression, one of early activation markers, there was no difference between control and LEC rats (data not shown). These findings strongly suggest that CD4+ T cells of MLNs and LP from LEC rats may play a crucial role in the development of spontaneous IBD-like lesions of LEC rats.
We next analyzed the proliferative response of T cells from LEC and control rats using stimulation with anti-CD3 and -CD28 mAbs. Purified T cells from spleen and MLNs of LEC and control rats were stimulated with plate-coated anti-CD3 mAb (0∼1 μg/ml) and anti-CD28 mAb (5 μg/ml) for 72 h. It was previously reported that splenic T cell response to Con A of LEC rats was significantly reduced (32). Consistent with the report, the proliferative responses of splenic T cells with CD3-CD28 ligation were significantly lower than that from control rats (Fig. 2,D). However, it was interesting to note that a prominent increase in T cell response with CD3-CD28 ligation in MLNs from LEC rats was observed compared with that from control rats (Fig. 2,D). To clarify which population of T cells indicates higher proliferative response, the purified T cells of MLNs were labeled with CFSE, and stimulated with CD3-CD28 ligation for 72 h. Proliferative response of CD4+ or CD8+ T cells was analyzed by flow cytometry using CFSE dilutions. We found that CD4+ T cells of MLNs from LEC rats were clearly much more proliferative compared with those from control rats, whereas no difference was observed in CD8+ T cells between LEC and control rats (Fig. 2 E). These results indicate that the CD4+ T cells in MLNs from LEC rats are primarily responsible for the development of IBD-like lesions in LEC rats.
Cytokine switching of T cells in LEC rats
To define the function of T cells of spleen and MLNs from LEC rats, cytokine production in the culture supernatants from T cells stimulated with CD3-CD28 ligation was analyzed. IL-2 production in the spleen cells from LEC rats was significantly lower (anti-CD3 mAb: 0.5 and 1 μg/ml) than those in control rats (Fig. 3,A). As for IFN-γ in spleen, no difference was observed between LEC and control rats. In addition, a significant decrease in Th2 cytokine production including IL-4 and IL-10 was observed in the spleen cells from LEC rats. In contrast, Th1 cytokines (IL-2 and IFN-γ) in the T cells of MLNs from LEC rats were significantly increased compared with those in control rats (Fig. 3 A). By contrast, IL-10 production from MLN T cells in LEC rats was significantly lower (anti-CD3 mAb: 1 μg/ml) than that in control rats. No difference in IL-4 production between LEC and control rats was observed. These results indicate that the Th1 response of MLN T cells from LEC rats may influence the pathogenesis of IBD-like lesions in LEC rats.
It is well known that T-bet and GATA-3 are key transcription factors in controlling Th1 or Th2 cytokine production (4, 5). We then analyzed the expression T-bet and GATA-3 using the T cells of MLNs stimulated with CD3-CD28 ligation. T-bet expression of T cells from LEC rats was relatively increased comparing with that from control rats, whereas no difference in GATA-3 expression was observed between LEC and control rats (Fig. 3,B). Moreover, the expression of STAT-1 and STAT-6, which exist upstream of T-bet and GATA-3, respectively, was analyzed. Increased phosphorylation of STAT-1 in T cells of MLNs from LEC rats was detected, but no difference was observed in STAT-1 expression between LEC and control rats. By contrast, there was no difference in the expression and phosphorylation of STAT-6 in either group (Fig. 3,B). To quantify the protein expressions of T-bet, pStat-1, GATA-3, and pSTAT-6, relative expressions to GAPDH as a housekeeping protein were shown in Fig. 3,C. T-bet expression of stimulated MLN T cells from LEC rats was 4∼5-fold higher compared with that of control rats. The pSTAT-1 had an expression that was twice as high relative to that of control rats. Furthermore, increased expression of Th1 cytokines including IL-2 and IFN-γ of CD4+ T cells in LP from LEC rats was observed compared with that from control rats by fluorescence staining (Fig. 3,D). By contrast, IL-4- or IL-10-producing cells were undetectable in the LP from both control and LEC rats (data not shown). To examine the correlation of Th1 cytokine with colitis of LEC rats, anti-IFN-γ-neutralizing Ab, anti-IL-4-neutralizing Ab, or isotype control Ab was injected into LEC rats from 8 to 10 wk of age, and the histology of colitis was evaluated (Fig. 3,E). Histological score of LEC rats administrated with anti-IFN-γ mAb was significantly lower than that with isotype control Ab. No significant change was observed by injection of anti-IL-4 mAb. Moreover, to investigate the role of CD8+ T cells in the development of colitis, anti-CD8-neutralizing Ab to deplete CD8+ T cells was injected into LEC rats. There was no significant change of pathology in anti-CD8 mAb administered LEC rats, compared with that in controls (Fig. 3,E). Ulcer formation, lymphocyte infiltrates, decreased numbers of mucin-producing cells, and mucosal hyperplasia were observed in the sections from isotype control Ab, anti-IL-4 mAb, and anti-CD8 mAb-injected rats, while injection of anti-IFN-γ mAb was able to effectively suppress the colon lesion of LEC rats (Fig. 3, E and F).
NF-κB activation of T cells in IBD-like lesions from LEC rats
To further confirm whether the T cell signal via up-regulated T-bet and phosphorylation STAT-1 in MLN T cells from LEC rats can lead to NF-κB activation, which is a potent transcription factor regulating the target genes essential for T cell activation or proliferation (38, 39, 40, 41), we analyzed NF-κB expression of the nuclear extracts of T cells stimulated with CD3-CD28 ligation. We demonstrated that there was an increased nuclear translocation of NF-κB (p65) of stimulated T cells in LEC rats compared with that in control rats (Fig. 4 A).
We next examined the expression of T-bet, GATA-3, and nuclear translocation of NF-κB of infiltrating CD4+ T cells in IBD-like lesions from LEC rats by confocal microscopy. Increased expression of T-bet together with nuclear transport of NF-κB was observed in CD4+ T cells of inflammatory lesions from LEC rats (Fig. 4, B and C), whereas GATA-3+ CD4+ T cells were almost undetectable in the lesions (Fig. 4 B). These results imply that T cells of LEC rats might be activated as effecter cells through STAT-1, T-bet, and NF-κB, resulting in Th1 cytokine production which affect the development of IBD-like lesions of LEC rats.
Treg cells in LEC rats
It has been reported that regulatory immune cells such as CD4+CD25+ T cells play a crucial role for the pathogenesis of both human IBD and the animal models (21, 22, 24). Thus, we analyzed whether CD4+CD25+ regulatory T cells influence the development of IBD-like lesions via the immune disorder of LEC rats. First, the cell populations of CD4+CD25+ T cells in the thymus, spleen, and MLNs form control and LEC rats were analyzed by flow cytometry. CD4+CD25+ T cells are known to express high levels of Foxp3, a transcription factor that in a normal rat is selectively expressed in CD25+ Treg cells (20). The number of CD25+Foxp3+ Treg cells of the thymus, spleen, and MLN from LEC rats was significantly lower than that of control rats (Fig. 5, A and B). In particular, Treg cells of the thymus and MLNs from LEC rats were reduced in contrast to control rats (thymus; p = 0.0113, MLN; p = 0.0133). Moreover, Foxp3+CD4+ T cells in MLN from LEC rats were significantly reduced relative to that from control rats (Fig. 5,B). Interestingly, Foxp3+ Tregs in MLNs (5.12 ± 0.71%) from control rats were markedly increased relative to that in the thymus (0.43 ± 0.11%) and spleen (2.04 ± 0.34%) as shown in Fig. 5,B. We next assessed the mRNA expression of Foxp3 in the thymus, spleen, and MLNs from LEC and control rats using real-time quantitative RT-PCR. Consistent with the results of flow cytometric analysis (Fig. 5, A and B), mRNA expression of Foxp3 in the thymus and MLNs from LEC rats was significantly reduced relative to that from control rats (Fig. 5,C). The prominent expression of Foxp3 mRNA in MLNs from both LEC and control rats was observed comparing with that of the thymus and spleen from both groups (Fig. 5,C). Furthermore, we confirmed the protein levels of Foxp3 using the nuclear extracts of CD4+ T cell in the thymus, spleen, and MLNs from LEC and control rats by Western blotting. As expected, the protein expression of Foxp3 in the thymus and MLNs from LEC rats was markedly lower, and its expression in the spleen from LEC was slightly decreased, compared with the control rats (Fig. 5,D). Moreover, to clarify the function of Treg cells from LEC rats, T cell suppression assay was performed. Namely, CD25−CD4+ T cells from control rats were stimulated with plate-coated anti-CD3 mAb and cocultured with control or LEC Treg cells. Although control Treg cells suppressed T cell response considerably, the suppressive function of LEC Treg cells was significantly impaired (Fig. 5,E). In addition, when Foxp3+ Treg cells in the each segment of colon including cecum, proximal, and distal colon were analyzed, Foxp3+ cells of LPMCs from all segments in LEC rats were significantly decreased compared with those in control rats (Fig. 5 F). These findings suggest that the decreased CD4+CD25+Foxp3+ Tregs in the thymus, MLNs, and LPMCs might affect the intestinal immunity associated with IBD-like lesions in LEC rats.
Treg cells from bone marrow (BM) cells of LEC rats
To know whether the decreased number of Treg cells in LEC rats is derived from the BM cells or not, the BM cells from LEA or LEC rats were transferred into irradiated LEA rats. The number of Foxp3+CD25+CD4+ Treg cells in thymus, spleen, and MLNs of chimeric rats was analyzed by flow cytometry. In parallel with the decreased numbers of Treg cells from LEC rats, the Foxp3+ Treg cells of thymus, spleen, and MLNs in the chimeric rats transferred with LEC BM cells were significantly reduced compared with those from LEA BM cells (Fig. 6,A). Moreover, as to the expansion of T cells from BM cells, although thymocytes and spleen cells of LEC BM cell transplantation (BMT) rats were slightly decreased compared with those of LEA BMT rats, a significantly increased cell number of MLN from LEC BMT rats was observed (Fig. 6 B). These findings indicate that the precursor of Treg may exist in the BM, and that the generation of Treg precursors in LEC rats might be deficient in the BM in addition to the deficiency of thymic differentiation.
Induction of IBD-like lesions by adoptive transfer of MLN T cells from LEC rats into SCID mice
To determine whether inflammatory bowel lesions of LEC rats could be induced in SCID mice via a T cell-mediated pathway, adoptive transfer of T cells in MLNs from LEC rats into SCID mice was performed. Importantly, inflammatory lesions of transferred SCID mice with MLN T cells were observed restrictedly in the colon, while no lesion was detected in Cont/SCID recipients at this stage (Fig. 7,A and Table I). In addition to the IBD-like lesions, the body weight of LEC/SCID recipients transferred with MLN T cells from LEC rats was significantly decreased after the transfer compared with that from Cont/SCID recipients (Fig. 7,B), whereas there was no change for Cont/SCID recipients. By contrast, adoptive transfer of splenic T cells from LEC rats did not induce any lesions in the SCID recipient (Table I). A large number of CD4+ T cell population was observed within the infiltrating T cells from IBD-like lesions in IBD/SCID recipients (Fig. 7,C). Moreover, it was demonstrated that MLN T cells stimulated with CD3 and CD28 from LEC/SCID recipients could produce higher levels of IL-2 and IFN-γ, not IL-4 and IL-10, than those from Cont/SCID recipients (Fig. 7,D). Furthermore, the increased expressions of NF-κB and T-bet in LPMCs from LEC/SCID mice were detected compared with those from Cont/SCID mice (Fig. 7 E). These results indicate that the IBD-like lesions of LEC rats can be transferred with MLN T cells, not splenic T cells, into SCID mice, and that Th1 response may play a pivotal role in the pathogenesis of IBD-like lesions in LEC rats.
|Transfera .||Incidence %b .||Induction/Total .|
|Transfera .||Incidence %b .||Induction/Total .|
Purified T cells of MLN or spleen from LEC and control rats were transferred into SCID mice. Six weeks later, histological analysis of colon lesions from SCID recipients was performed.
Incidence of IBD-like lesions was evaluated using four to eight SCID recipients.
As for the animal models of IBD, there have been two categories reported previously (1). One is a spontaneous IBD model with any immune dysfunction such as in IL-2−/−, IL-2R−/−, IL-10−/−, and NOD2−/− mice (18, 41, 42, 43). The other is a model manipulated by any drug, bacteria, and cell transfer (44, 45, 46, 47, 48). In this study, we demonstrated that the LEC rat is one of spontaneous IBD models, by which the mechanism is due to the decline of Treg cells and the Th1 shift of the cytokine.
T lymphocytes play a central role for the intestinal immune system (12, 49). Recent studies suggested that the balance between Th1 and Th2 cytokines secreted by T cells appears to regulate IBD (13, 39, 48). In this study, the pattern of cytokine production of MLN cells and LPMCs, but not spleen cells, in LEC rats had clearly shifted to Th1. It was reported that the proliferative response of peripheral T cells in LEC rats against T cell mitogen such as Con A had decreased, and that IL-2 production of T cells in LEC rats had reduced (32). However, we found here that proliferative response and Th1 cytokine production of MLN T cells through the TCR/CD3 and CD28 pathway had increased compared with that of control rats. The response of MLN T cells or LPMCs which may regulate the intestinal immune system seems to be different from that of spleen cells from a point of view of cytokine production, and any other immune functions. Namely, MLNs and LPMCs have unique functions regulating the intestinal immune system with exposure to numerous food, bacterial, and/or endogenous Ags. It is possible that defective thymic selection for differentiation into CD4SP cells form DP cells in LEC rats might influence any function including cytokine production in the peripheral T cells. It is well known that cytokine production of immune cells can be occurred through various stimuli such as microorganisms, inflammations, and mechanical or physiological stresses (45, 50, 51, 52). The balance between Th1 and Th2 cytokine production is controlled by a number of transcriptional factors and signaling molecules (4). Recent studies indicated that GATA-3 and T-bet might play a central role in controlling the balance of cytokine production from Th cells (4, 5, 6, 7). We demonstrated here that Th1 response in MLN T cells and LPMCs through T-bet, STAT-1, and NF-κB might be considerably enhanced, and then the T cell response to numerous intestinal Ags could be responsible for the development of IBD-like lesions in LEC rats.
CD4+CD25+Foxp3+ Treg cells have been widely investigated to be generated from the thymus and regulate the peripheral T cells (20). The defective thymic selection of LEC rat influences positive selection into CD4+ from CD4+CD8+ cells, and results in the dysfunction of peripheral CD4+ T cells (32). It has been reported that Treg cells can prevent IBD-like lesions induced by naive CD4+ T cell transfer into T cell-deficient mice, and regulate Th1 effectors producing IFN-γ (24). In this study, it is speculated that the dysfunction of Treg cells in LEC rats might be associated with the hyperactivation of Th1 effectors in the periphery. In particular, Treg cells in MLNs and LPMCs might play a crucial role for regulating the immune response against numerous intestinal Ags. Therefore, the dysfunction of Treg cells in MLNs and LPMCs would be associated with the deficiency of intestinal tolerance, resulting in the development of IBD-like lesions in LEC rats. In contrast, although Foxp3+CD25−T cells were reported to have suppressive functions similar to those of Foxp3+CD25+ classical Treg cells in normal mice (53), there was no difference in the Foxp3+CD25−T cells (percent) of thymus (control; 0.063 ± 0.011%, LEC; 0.068 ± 0.018%), spleen (control; 0.25 ± 0.04%, LEC; 0.20 ± 0.06%), MLNs (control; 0.47 ± 0.07%, LEC; 0.41 ± 0.06%) between control and LEC rats. The Foxp3+CD25+ Treg cells, but not Foxp3+CD25−T cells, of LEC rats might influence the development of colitis of LEC rats.
Direct in vivo evidence of T cell-dependent IBD-like lesions of LEC rats is the induction of the lesion into SCID mice by adoptive transfer with MLN T cells from LEC rats. Although defective T cell functions have been previously reported using fetal thymus grafts of LEC rats into SCID mice, the inflammatory lesions in general have not been investigated (54). Although it is not unclear whether the precursor of Treg cell is generated in BM, the experiment using BM chimera of LEC and LEA rats may imply the origin of the Treg precursor in BM cells.
The LEC rat has been also reported to be used as an animal model for Wilson’s disease and developed hepatitis from four months after birth because of the genetic copper metabolism disorder. At a later age, chronic hepatitis and hepatocellular carcinoma are observed in LEC rats (36, 55). The LEC rats have been useful in studying mechanisms of spontaneous carcinogenesis. Generally described, it has been shown that the pathogenesis of colon cancer could be associated with chronic inflammation such as IBD (56). Previous studies demonstrated that there was a high frequency of 7 N-methyl-N-nitrosourea-induced colon adenocarcinoma in LEC rats was observed compared with that of control rats (57). It can be speculated that IBD-like inflammatory lesions of LEC rats might affect the carcinogenesis via any indirect mechanism. Although LEC rats were described to have the defective differentiation of CD4 SP cells in the thymus, the decline of peripheral CD4+ T cells, and the failure of proliferative response to T cell mitogen in CD4+ T cells, any inflammatory disease based on these immune disorders has not been investigated. In contrast, the mutation of Th immunodeficiency (thid) gene was previously reported (58), which might be associated with immune disorder of LEC rats. Our study focused on Treg cells and cytokine switching as the mechanisms of IBD-like lesions of LEC rats. It is still unclear whether thid gene regulates the generation of Treg cells and cytokine switching through T-bet and STAT-1 leading to NF-κB in CD4+ T cells from LEC rats.
In summary, we demonstrated for the first time that IBD-like lesions developed spontaneously in LEC rats. The dysfunction of Treg cells in the periphery and Th1 shift of cytokine production might play a crucial role for pathogenesis of IBD-like lesions in LEC rats. Analyzing the molecular mechanism for IBD-like lesions in LEC rats will be useful for understanding human IBD based on immune disorder in central and peripheral tolerance.
We thank Ai Nagaoka and Satoko Yoshida for their technical assistance.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by Grants-in-Aid for Scientific Research (17109016 and 17689049) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
Abbreviations used in this paper: IBD, inflammatory bowel disease; Treg, regulatory T; DP, double positive; SP, single positive; MLN, mesenteric lymph node cell; LP, lamina propria; BM, bone marrow; PAS, periodic acid Schiff; DAPI, 4′,6-diamidino-2-phenylindole; HPRT, hypoxanthine phosphoribosyltransferase; LPMC, LP mononuclear cell.