We previously demonstrated that IL-7 is produced by intestinal goblet cells and is essential for the persistence of colitis. It is well known, however, that goblet cells are decreased or depleted in the chronically inflamed mucosa of animal colitis models or human inflammatory bowel diseases. Thus, in this study, we assess whether intestinal IL-7 is surely required for the persistence of colitis using a RAG-1/2−/− colitis model induced by the adoptive transfer of CD4+CD45RBhigh T cells in combination with parabiosis system. Surprisingly, both IL-7−/− × RAG-1−/− and IL-7+/+ × RAG-1−/− host mice developed colitis 4 wk after parabiosis to a similar extent of colitic IL-7+/+ × RAG-1−/− donor mice that were previously transferred with CD4+CD45RBhigh T cells. Of note, although the number of CD4+ T cells recovered from the spleen or the bone marrow of IL-7−/− × RAG-1−/− host mice was significantly decreased compared with that of IL-7+/+ × RAG-1−/− host mice, an equivalent number of CD4+ T cells was recovered from the lamina propria of both mice, indicating that the expansion of CD4+ T cells in the spleen or in the bone marrow is dependent on IL-7, but not in the lamina propria. Development of colitis was never observed in parabionts between IL-7+/+ × RAG-1−/− host and noncolitic IL-7−/− × RAG-1−/− donor mice that were transferred with CD4+CD45RBhigh T cells. Collectively, systemic, but not intestinal, IL-7 is essential for the persistence of colitis, suggesting that therapeutic approaches targeting the systemic IL-7/IL-7R signaling pathway may be feasible in the treatment of inflammatory bowel diseases.
Inflammatory bowel disease (IBD)3 are caused by chronic inflammatory responses in the gut wall, commonly take persistent courses, but in some patients relapse after remissions (1, 2, 3, 4, 5, 6). Because the recurrent disease usually mimics the primary disease episode, it is possible that the disease is caused by the repeated activation and expansion of colitogenic effector CD4+ T cells arising from common long-lived colitogenic memory CD4+ T cells, which latently reside in their target tissues or in some reservoirs. Nevertheless, the nature of the colitogenic memory CD4+ T cells over time is not fully understood.
IL-7 is secreted by stromal cells in the bone marrow (BM) and thymus, and epithelial cells including the intestine (7, 8, 9, 10). Recent findings revealed that IL-7 is an important cytokine supporting the survival of resting naive and memory CD4+ T cells, but not effector CD4+ T cells (9, 10, 11, 12, 13, 14, 15, 16). We have previously demonstrated that, 1) IL-7 is constitutively produced by intestinal goblet epithelial cells (8), 2) IL-7 transgenic (Tg) mice, in which IL-7 overexpression was driven by SRα promotor, developed chronic colitis that mimicked histopathological characteristics of human IBD (17), 3) mucosal CD4+IL-7Rαhigh T cells in CD4+CD45RBhigh T cell-transferred colitic mice are colitogenic (18), and 4) IL-7−/− × RAG-1−/− mice transferred with colitogenic lamina propria (LP) CD4+ T cells isolated from colitic CD4+CD45RBhigh T cell-transferred mice did not develop colitis (19).
Somewhat at odds, however, we also found that production of intestinal IL-7 was dramatically decreased in the inflamed mucosa of colitic IL-7 Tg mice in accordance with depletion of goblet cells (17). Because our IL-7 Tg mice were established by expressing IL-7 under regulation of the ubiquitous SRα promoter, it was possible that intestinal IL-7 is indeed decreased at the site of mucosal inflammation due to depletion of goblet cells, which is a feature often seen in the inflamed mucosa of human IBD, but systemic IL-7 of other tissue origin, such as BM (20) and thymus (21), is rather critical for the maintenance of colitogenic memory CD4+ T cells. Based on these complex backgrounds, in this study, we assess the distinct requirement of intestinal or systemic IL-7 in the development and persistence of colitis using a RAG-1/2−/− colitis model (22, 23) induced by adoptive transfer of CD4+CD45RBhigh T cells in combination with parabiosis system.
Materials and Methods
C57BL/6-Ly5.2 mice were purchased from Japan CLEA. C57BL/6-Ly5.1 mice and C57BL/6-Ly5.2-RAG-2-deficient (RAG-2−/−) mice were obtained from Taconic Farms and Central Laboratories for Experimental Animals. C57BL/6-Ly5.2-background RAG-1+/− and IL-7+/− mice were provided from Dr. Rosa Zamoyska (National Institute for Medical Research, London, U.K.) (24). These mice were intercrossed to generate IL-7−/− × RAG-1−/− and IL-7+/+ × RAG-1−/− littermate mice in the Animal Care Facility of Tokyo Medical and Dental University (TMDU). Mice were maintained under specific pathogen-free conditions in the Animal Care Facility of TMDU. Donors and recipients were used at 6–12 wk of age. All experiments were approved by the regional animal study committees and were done according to institutional guidelines and Home Office regulations.
Parabiosis experimental design
To assess the specific requirement of mucosal or systemic IL-7 in the development of colitis, we performed adoptive transfer experiment in combination with a parabiosis system using IL-7+/+ × RAG-1−/− and IL-7−/− × RAG-1−/− littermate recipients (Fig. 1 A). For adoptive transfer, CD4+ T cells were first isolated from SP cells of C57BL/6-Ly5.2 mice using the anti-CD4 (L3T4)-MACS system (Miltenyi Biotec) according to the manufacturer’s instruction. Enriched CD4+ T cells (96–97% pure, as estimated by FACSCalibur (BD Biosciences)) were then labeled with PE-conjugated anti-mouse CD4 (RM4–5; BD Pharmingen) and FITC-conjugated anti-mouse (16A; BD Pharmingen). CD4+CD45RBhigh cells were purified using a FACSAria (BD Biosciences). This population was >98.0% pure on reanalysis. IL-7+/+ × RAG-1−/− mice (n = 18) and IL-7−/− × RAG-1−/− mice (n = 6) were then injected i.p. with 3 × 105 splenic CD4+CD45RBhigh T cells from normal C57BL/6-Ly5.2 mice. After 6 wk post transfer, IL-7+/+ × RAG-1−/− mice, but not IL-7−/− × RAG-1−/− mice, transferred with CD4+CD45RBhigh T cells developed a wasting disease and colitis as previously reported (19).
We then conducted parabiosis surgery according to institutional guidelines and Home Office regulations. In brief, sex-matched mice were anesthetized before surgery, and incisions were made in the skin on the opposing flanks of the donor and recipient animals. Surgical sutures were used to bring the body walls of the two mice into direct physical contact. The outer skin was then attached with surgical staples. For this parabiosis experiment, we divided colitic IL-7+/+ × RAG-1−/− (n = 18) mice that were previously transferred with CD4+CD45RBhigh T cells into three groups: Group 1, colitic IL-7+/+ × RAG-1−/− mice joined with normal C57BL/6-Ly5.1 mice (n = 6); Group 2, colitic IL-7+/+ × RAG-1−/− mice joined with new IL-7+/+ × RAG-1−/− mice (n = 6); Group 3, colitic IL-7+/+ × RAG-1−/− mice joined with new IL-7−/− × RAG-1−/− mice (n = 6). As Group 4, noncolitic IL-7−/− × RAG-1−/− mice previously transferred with CD4+CD45RBhigh T cells were joined with new IL-7+/+ × RAG-1−/− mice (n = 6). All mice were observed for clinical signs, such as hunched posture, piloerection, diarrhea, and blood in the stool. At autopsy, mice were assessed for a clinical score (25) that is the sum of three parameters as follows: hunching and wasting, 0 or 1; colon thickening, 0–3 (0, no colon thickening; 1, mild thickening; 2, moderate thickening; 3, extensive thickening); and stool consistency, 0–3 (0, normal beaded stool; 1, soft stool; 2, diarrhea; 3, bloody stool) (25).
Tissue samples were fixed in PBS containing 10% neutral-buffered formalin. Paraffin-embedded sections (5 μm) were stained with H&E. Three tissue samples from the proximal, middle, and distal parts of the colon were prepared. The sections were analyzed without prior knowledge of each mouse. The area most affected was graded by the number and severity of lesions. The mean degree of inflammation in the colon was calculated using a modification of a previously described scoring system (25) as follows: mucosa damage, 0; normal, 1; 3–10 intraepithelial cells (IEL)/high power field (HPF) and focal damage, 2; >10 IEL/HPF and rare crypt abscesses, 3; >10 IEL/HPF, multiple crypt abscesses and erosion/ulceration, submucosa damage, 0; normal or widely scattered leukocytes, 1; focal aggregates of leukocytes, 2; diffuse leukocyte infiltration with expansion of submucosa, 3; diffuse leukocyte infiltration, muscularis damage, 0; normal or widely scattered leukocytes, 1; widely scattered leukocyte aggregates between muscle layers, 2; leukocyte infiltration with focal effacement of the muscularis, 3; extensive leukocyte infiltration with transmural effacement of the muscularis.
Single cell suspensions were prepared from SP, LP, and BM as previously described (18). To isolate LP CD4+ T cells, the entire length of the colon was opened longitudinally, washed with PBS, and cut into small pieces. The dissected mucosa was incubated with Ca2+, Mg2+-free HBSS containing 1 mM DTT (Sigma-Aldrich) for 45 min to remove mucus and then treated with 3.0 mg/ml collagenase (Roche) and 0.01% DNase (Worthington Biochemical) for 2 h. The cells were pelleted two times through a 40% isotonic Percoll solution, and then subjected to Ficoll-Hypaque density gradient centrifugation (40/75%). Enriched LP CD4+ T cells were obtained by positive selection using anti-CD4 (L3T4) MACS magnetic beads. The resultant cells when analyzed by FACSCalibur contained >95% CD4+ cells. BM cells were obtained by flushing two femurs with cold RPMI 1640. For in vitro assay, only live cells were counted by using trypan blue staining method, and confirmed that the viability of cells was almost the same (>96% live) among the sample groups.
Reverse transcription polymerase chain reaction
Total RNA was isolated by using Isogen reagent (Nippon Gene). Aliquots of 5 μg total RNA were used for complementary DNA synthesis in a reaction volume of 20 μl using random primers. One microliter of reverse transcription product was amplified with 0.25 U of rTaq DNA polymerase (Toyoba) in a 50 μl reaction. Sense and antisense primers and the cycle numbers for the amplification of each gene were as follows: sense IL-7, 5′-GCCTGTCACATCATCTGAGTGCC-3′ and antisense IL-7, 5′-CAGGAGGCATCCAGGAACTTCTG-3′ for IL-7 (35 cycles); and sense G3PDH, 5′-TGAAGGTCGGTGTGAACGGATTTGGC-3′ and antisense G3PDH, 5′-CATGTAGGCCATGAGGTCCACCAC-3′ for G3PDH (30 cycles). The amplification for each gene was logarithmic under these conditions. PCR products were separated on 1.8% agarose gels, stained with ethidium bromide, and visualized with a Lumi-Imager F1 (Roche).
We used consecutive cryostat colon sections in all studies. Immunohistochemistry using purified mAb against mouse CD4 (RM4–5; BD Pharmingen) or biotin-conjugated polyclonal IL-7 Ab (BAF407; R&D Systems) was performed. In brief, O.C.T. compound-embedded tissue samples were cut into serial sections 6-μm thick, placed on coated slides, and fixed with 4% paraformaldehyde phosphate buffer solution for 10 min. Slides were then incubated with the primary Ab at 4°C overnight, followed by staining with AlexaFluor 488 goat anti-rat IgG for CD4 detection or AlexaFluor 488 streptavidin (Molecular Probes) for IL-7 detection at room temperature for 60 min. All slides were counterstained with 4′, 6′ -diamidino-2-phenylindole (DAPI; Vector Laboratories) and observed under a confocal microscopy (LSM510 Carl Zeiss).
To measure cytokine production, 1 × 105 LP CD4+ T cells were cultured in triplicate of 200 μl culture medium at 37°C in a humidified atmosphere containing 5% CO2 in 96-well plates (Costar) precoated with 5 μg/ml hamster anti-mouse CD3ε mAb (145-2C11, BD Pharmingen) and hamster 2 μg/ml anti-mouse CD28 mAb (37.51, BD Pharmingen) in PBS overnight at 4°C. Culture supernatants were collected after 48 h and assayed for cytokine production. Cytokine concentrations were determined by specific ELISA following the manufacturer’s recommendation (R&D Systems).
To detect the surface expression of a variety of molecules, isolated SP, BM, or LP mononuclear cells were preincubated with an FcγR-blocking mAb (CD16/32; 2.4G2, BD Pharmingen) for 20 min followed by incubation with specific FITC-, PE-, PerCP-, allophycocyanin-labeled Abs for 30 min on ice. The following mAbs were obtained from BD Pharmingen: anti-CD4 mAb (RM4–5), anti-CD45RB mAb (16A), anti-CD45.1 (Ly5.1; A20), and anti-CD45.2 (Ly5.2; 104). Standard four-color flow cytometric analyses were obtained using the FACSCalibur and analyzed by CellQuest software. Background fluorescence was assessed by staining with control irrelevant isotype-matched mAbs.
The results are expressed as mean ± SEM. Groups of data were compared by the Mann-Whitney U test. Differences were considered to be statistically significant when p < 0.05.
IL-7−/− × RAG-1−/− host mice joined with colitic IL-7+/+ × RAG-1−/− donor mice develop a wasting disease
We have previously demonstrated that IL-7 is essential for the development and the persistence of colitis as a survival factor for colitogenic CD4+ memory T cells (19). Furthermore, we have found that IL-7 Tg mice, in which IL-7 was systemically overproduced, develop colitis spontaneously, but production of intestinal IL-7 was conversely decreased in the inflamed mucosa because of depletion of the goblet cells. Based on such paradoxical findings, in this study, we assess whether intestinal or systemic IL-7 is essential for the perpetuation of colitis, by adoptive transfer experiment in combination with parabiosis system using IL-7+/+ × RAG-1−/− and IL-7−/− × RAG-1−/− littermate recipients (Fig. 1,A). To this end, we first induced chronic colitis by adoptive transfer of splenic CD4+CD45RBhigh T cells from normal C57BL/6-Ly5.2 mice into IL-7+/+ × RAG-1−/− mice (Fig. 1 A). Consistent with our previous report (19), the transferred IL-7+/+ × RAG-1−/− mice manifested progressive weight loss from 3 wk after transfer and clinical symptoms of colitis 6 wk after transfer (data not shown). In contrast, the CD4+CD45RBhigh T cell-transferred IL-7−/− × RAG-1−/− mice showed no clinical signs of colitis and weight loss (data not shown) (19), indicating that IL-7 is essential for the development of colitis.
At 6 wk after transfer, we next generated four groups of parabionts (Fig. 1,A). In parabionts between colitic IL-7+/+ × RAG-1−/− donor mice that has been previously transferred with Ly5.2+CD4+CD45RBhigh T cells and normal C57BL/6-Ly5.1 host mice (Group 1) (Fig. 1,A), clinical symptoms, such as diarrhea, anorectal prolapse, and hunched posture, gradually decreased over time in IL-7+/+ × RAG-1−/− donor mice as compared with the mice at the time of surgery, and completely disappeared at 4 wk after surgery by assessing the clinical score (Fig. 1,B). C57BL/6-Ly5.1 host mice were consistently healthy during the observed period (Fig. 1,B). In parabionts between colitic IL-7+/+ × RAG-1−/− donor mice and new IL-7+/+ × RAG-1−/− host mice (Group 2) (Fig. 1,A), all the IL-7+/+ × RAG-1−/− donor mice were consistently diseased (Fig. 1,B), and clinical symptoms of colitis gradually increased in new IL-7+/+ × RAG-1−/− host mice, which reached to the equal level of the paired IL-7+/+ × RAG-1−/− donor mice at 4 wk after surgery (Fig. 1,B). In parabionts between colitic IL-7+/+ × RAG-1−/− donor mice and new IL-7−/− × RAG-1−/− host mice (Group 3) (Fig. 1,A), IL-7+/+ × RAG-1−/− donor mice remained diseased to a similar level of IL-7+/+ × RAG-1−/− donor mice in Group 2 (Fig. 1,B), and notably, IL- 7−/− × RAG-1−/− host mice, albeit with the absence of intestinal IL-7, were gradually sick and clinical symptoms of colitis reached to the equal level of paired IL-7+/+ × RAG-1−/− donor mice and the IL-7+/+ × RAG-1−/− host mice in Group 2 at 4 wk after surgery (Fig. 1,B). In sharp contrast, in parabionts between the nondiseased IL-7−/− × RAG-1−/− donor mice that were transferred with CD4+CD45RBhigh T cells and new IL-7+/+ × RAG-1−/− host mice (Group 4) (Fig. 1,A), both IL-7−/− × RAG-1−/− donor and IL-7+/+ × RAG-1−/− host mice were consistently healthy during the observed period (Fig. 1 B), indicating that CD4+CD45RBhigh T cell-transferred IL-7−/− × RAG-1−/− mice never retained colitogenic CD4+ T cells.
IL-7−/− × RAG-1−/− host mice parabiosed with colitic IL-7+/+ × RAG-1−/− donor mice develop Th1-mediated colitis
Four wk after surgery, the colons from parabionts between IL-7+/+ × RAG-1−/− donor mice and C57BL/6-Ly5.1 host mice in Group 1 and parabionts between IL-7−/− × RAG-1−/− donor mice and IL-7+/+ × RAG-1−/− host mice in Group 4 were macroscopically normal (data not shown). In contrast, the colon from all mice in Groups 2 and 3, regardless of IL-7+/+ × RAG-1−/− or IL-7+/+ × RAG-1−/− mice and as donors or hosts, were equally enlarged and had a greatly thickened wall (data not shown). In addition, the enlargement of spleen was also present in donors and hosts of Groups 2 and 3 mice (data not shown). Histological examination showed that in colons from Group 1, donor IL-7+/+ × RAG-1−/− mice, which initially had clinical symptoms of colitis, exhibited no pathological change 4 wk after surgery, and were indistinguishable from the colons of C57BL/6-Ly5.1 host mice (Fig. 2,A, left). In turn, we could not detect any pathological finding in Group 4 parabionts between IL-7−/− × RAG-1−/− donor mice and IL-7+/+ × RAG-1−/− host mice. In contrast, all the donor and host mice in Groups 2 and 3 parabionts showed prominent epithelial hyperplasia with glandular elongation and a massive infiltration of mononuclear cells (Fig. 2,A, left). This difference was also confirmed by histological scoring of colon sections (Fig. 2,B), showing that the host mice in parabionts in Groups 2 and 3 developed colitis comparable to the paired diseased donor mice that had sustained colitis, while all the donor and host mice in Groups 1 and 4 did not develop colitis. Furthermore, acid mucin production examined by Alcian blue staining revealed a marked decrease of mucin-producing goblet cells in all colitic mice in Groups 2 and 3 in contrast to mice in Groups 1 and 4 (Fig. 2 A, right).
To clarify that newly developed colitis in host mice of Groups 2 and 3 was surely mediated by the infiltration of immigrant CD4+ T cells from donor mice, but not by innate immune cells such as granulocytes and macrophages, we next assessed colonic infiltration of CD4+ T cells by immunohistochemistry. Fig. 3 clearly demonstrated marked infiltration of CD4+ T cells in the colon of host mice as well as in donor mice in parabionts of Groups 2 and 3. In contrast, only a small population of CD4+ T cells was found in the host and donor mice in Groups 1 and 4 (Fig. 3). Especially, although the IL-7+/+ × RAG-1−/− host mice in Group 1 had severe wasting disease with symptoms of colitis before surgery, there were only a few infiltrated CD4+ T cells observed in colonic LP, indicating that the previous colitis was suppressed and cured by certain immigrant suppressor cells derived from normal host mice.
We next examined the cytokine production by LP CD4+ T cells from each mouse in Groups 1–4. As shown in Fig. 4, LP CD4+ T cells from donor and host mice in Groups 2 and 3 produced significantly higher amounts of IFN-γ and TNF-α as compared with those from mice in Groups 1 and 4, indicating that colitic LP CD4+ T cells in IL-7−/− × RAG-1−/− host mice or IL-7+/+ × RAG-1−/− host mice of Groups 2 and 3 have functions of Th1-mediated immune responses. Importantly, the elevated production of these cytokines in Groups 2 and 3 was dependent on the presence of colitis, but not on the expression of IL-7 in the colon.
Expansion of CD4+ T cells is dependent on IL-7 in the SP or BM but is independent of IL-7 in the LP
We have previously reported that BM retaining colitogenic CD4+ T cells in colitic mice might play a critical role as a reservoir for persisting colitis (18). Furthermore, BM is physiologically a major source of IL-7, contributing to the development of B cells (24). To further investigate the role of intestinal and/or systemic IL-7 in consecutive immunopathology of the parabiosis model, we next compared the composition of CD4+ T cells in the LP, BM, and SP of donor and host mice in each parabiont using flow cytometry at 4 wk after surgery. The recovered cell numbers of CD3+CD4+ T cells from the donor and host LP in Groups 2 and 3 were significantly higher as compared with those of the paired donor and host colitic mice in Groups 1 and 4 parabionts, respectively (Fig. 5,A). Furthermore, the recovered cell numbers of CD3+CD4+ T cells in the donor and host BM (Fig. 5,B) and SP (Fig. 5,C) in Groups 2 and 3 were significantly higher as compared with those of the paired donor and host colitic mice in Group 4, but not in Group 1, parabionts, respectively. In contrast, IL-7+/+ × RAG-1−/− donor mice that were previously transferred with CD4+CD45RBhigh T cells and C57BL/6-Ly5.1 host mice in Group 1 sustained a normal number of cells in the BM and SP (Fig. 5, data not shown). Most importantly, although the number of CD3+CD4+ T cells recovered from the SP or BM of the IL-7−/− × RAG-1−/− host mice in Group 3 was significantly decreased compared with that of the IL-7+/+ × RAG-1−/− host mice in Group 2, an equivalent number of CD4+ T cells was recovered from the LP of both host mice in Groups 2 and 3, indicating that the expansion of CD4+ T cells in the SP and BM is dependent on IL-7, but is independent in the LP.
Further analysis of Group 1 mice using a four-colored CD3/CD4/Ly5.1/Ly5.2 FACS staining revealed that >95% of total CD4+ T cells were derived from Ly5.1+ cells and most resident Ly5.2+CD4+ T cells decreased to only 5–10% of total CD4+ T cells in SP and BM in Group 1 IL-7+/+ × RAG-1−/− donor mice (Fig. 5). Interestingly, although the absolute number of LP CD4+ T cells was significantly decreased in Group 1 IL-7+/+ × RAG-1−/− donor mice as compared with those of IL-7+/+ × RAG-1−/− donor mice in Groups 2 and 3 colitic parabionts, ∼50% of total LP CD4+ T cells remained to be Ly5.2+, suggesting that 1) colitogenic LP Ly5.2+CD3+CD4+ T cells were resistant to the suppression by Ly5.1-derived cells as compared with Ly5.2+CD3+CD4+ T cells in other sites and/or 2) they remained in the intestine, and in other words could not exit, and redistribute outside the intestine. Furthermore, small but substantial percentages (1–5%) of total CD4+ T cells in each tissue of host C57BL/6-Ly5.1 mice were donor-derived Ly5.2+ cells, indicating that two-way recirculation of CD4+ T cells from the donor to the host and vice versa had been established and most of Ly5.2+ colitogenic CD4+ T cells in both donor and host mice had undergone the contraction under a certain suppressive mechanism including suppression by CD4+CD25+Foxp3+ regulatory T cells derived from host C57BL/6 mice.
IL-7 is not detected in host IL-7−/− × RAG-1−/− host mice after parabiosis
Studies showing engraftment of BM-derived cells to various nonhemopoietic tissues including epithelial cells after BM transplantation are now on topic (26, 27), and we have previously demonstrated that human BM cells have a potential to repopulate the gastrointestinal epithelia by detecting Y-chromosomes in female cases that have undergone BM transplantation using male donor cells (28). It was thus needed to assess whether this was the case with our parabiosis setting, and if so, it was interesting to know whether IL-7 was produced by engrafted colonic epithelial cells derived from the BM of IL-7+/+ × RAG-1−/− donor mice in IL-7−/− × RAG-1−/− host mice after surgery in Group 3. As shown in Fig. 6,A, immunohistochemistry revealed that IL-7 is detected in uninflamed colonic epithelia of both IL-7+/+ × RAG-1−/− donor and C57BL/6 host mice in Group 1 and IL-7+/+ × RAG-1−/− host, but not in IL-7−/− × RAG-1−/− donor, mice in Group 4. Consistent with previous findings (17), IL-7 expression was detectable, but markedly decreased in inflamed colonic epithelia in Groups 2 and 3 of IL-7+/+ × RAG-1−/− mice along with the decreased goblet cells, in both host and donor mice (Fig. 2,A, right). In contrast, IL-7 was not detected in the inflamed colonic epithelia of Group 3 IL-7−/− × RAG-1−/− host mice (Fig. 6,A). Consistent with these results, further RT-PCR analysis for IL-7 mRNA expression showed that IL-7 mRNA was not detected in colitic IL-7−/− × RAG-1−/− host mice in Group 3, and was markedly decreased in colitic IL-7+/+ × RAG-1−/− donor and host mice in Groups 2 and 3 in clear contrast to that of control C57BL/6 mice (Fig. 6 B).
In this study, we demonstrated that intestinal IL-7 is not essential for the development and perpetuation of colitis by showing that IL-7−/− × RAG-1−/− host mice parabiosed with colitic IL- 7+/+ × RAG-1−/− donor mice develop a wasting disease and severe colitis. Because we previously demonstrated that IL-7 is needed to develop and sustain colitis by showing a lack of colitis development in IL-7−/− × RAG-1−/− mice transferred with CD4+CD45RBhigh T cells or colitogenic LP CD4+ T cells (19), in this study, we suggest that IL-7 production from tissues other than the intestine, such as BM, is sufficient, or rather may be essential to develop and sustain the chronic colitis.
Before starting this study, we confronted a paradox between two facts. The first fact is that IL-7-producing goblet cells are easily decreased or depleted in patients with severe ulcerative colitis (29), colitic IL-7 Tg mice (17) and in the present model of colitis (Fig. 2 A, right) resulting in the decreased IL-7 production in the intestine, and the second fact is that IL-7 appeared to be indispensable for the development and persistence of chronic colitis by adoptive transfer experiment using IL-7−/− × RAG-1−/− mice (19). Based on these backgrounds, we hypothesized that intestinal IL-7 is indeed important to establish GALT, such as Payer’s patches and cryptopatches, and also to maintain IELs (30), but not needed to develop and sustain colitis, since many Ags, such as intestinal bacterial Ags, may be sufficient to stimulate colitogenic CD4+ T cells in the intestinal LP without stimuli from IL-7. To prove it, we performed a combinational experiment using adoptive transfer and parabiosis systems in the present study. Although the parabiosis system seems to be somewhat artificial and problematic on some level as two mice, host and donor, are forced to have a surgical stress and behavioral limitation (Groups 3 and 4), mice laboring colitogenic CD4+ T cells are surgically joined, resulting in prompt development of anastomoses of blood vessels within a few days. Even in the present setting, it is noteworthy that IL-7−/− × RAG-1−/− host mice joined with colitic IL-7+/+ × RAG-1−/− donor mice developed a wasting disease and colitis to the similar level of colitic IL-7+/+ × RAG-1−/− donor mice over time.
In this parabiosis system, however, it was also possible that certain stem cells that are committed to differentiate into IL-7-producing mesenchymal cells or epithelial cells homed to the intestine, and might have been involved in the development and persistence of colitis in IL-7−/− × RAG-1−/− host mice joined with colitic IL-7+/+ × RAG-1−/− donor mice (Group 3). To rule out this possibility, we also demonstrated that IL-7 expression was not detected in the colon of the IL-7−/− × RAG-1−/− host mice both at the protein and mRNA levels (Fig. 6). Consistent with the present result, another group demonstrated that restoring intestinal IL-7 expression to IL-7−/− mice did not result in the development of colitis (31). Collectively, the current results clearly indicate that intestinal IL-7 is not essential, but systemic IL-7 from extraintestinal sites is essential, for the development and sustainment of colitis.
It is also very important to know why IL-7 is decreased in the inflamed mucosa of colitis in terms of pathogenesis of chronic colitis. In other words, it is possible that the lack or decrease of IL-7 production in inflamed mucosa of colitis is pathologically needed to maintain chronic colitis. Consistent with this hypothesis, we previously demonstrated that although IL-7 promoted proliferation of human LP IL-7Rα-expressing CD4+ T cells, double stimuli by IL-7 and anti-CD3 mAb conversely suppressed it (21). In addition, Fluur and colleagues (32) very recently reported that IL-7 induces Fas-mediated T cell apoptosis by inducing Fas expression on CD4+ T cells. Thus, it appears that intestinal IL-7 physiologically plays a key role in the elimination of pathological LP CD4+ T cells activated by intestinal bacteria. Further studies will be needed to address this issue.
Interestingly, the recovered cell number of LP CD4+ T cells was equivalent between host and donor mice both in Group 2 and 3, although it was likely that total production of IL-7 in Group 3 parabionts between one IL-7+/+ mouse and one IL-7−/− mouse was approximately half compared with that in Group 2 parabionts between two IL-7+/+ mice. Because it seems that the production of IL-7 is maintained at a constant rate and is uninfluenced by extrinsic stimuli (33, 34), this result indicates that factors other than IL-7, such as stimulation by commensal bacteria might control the homeostasis of cell number in the LP, but not in the BM and SP. Further studies will be needed to address this issue.
BM is a major source of IL-7 in the body (26). In contrast to the LP, it is noteworthy that the number of CD4+ T cells recovered from the BM and SP of the colitic IL-7−/− × RAG-1−/− host mice (Group 3) was significantly decreased compared with that of the IL-7+/+ × RAG-1−/− host mice. Regarding this result, we recently demonstrated that CD4+ effector-memory-like T (TEM-like) cells reside in the BM of colitic SCID and RAG-1/2−/− mice induced by adoptive transfer of CD4+CD45RBhigh T cells (20). Importantly, these resident BM CD4+ TEM-like cells are closely attached to IL-7-producing stromal cells in the colitic BM. Most importantly, the accumulation of BM CD4+ TEM-like cells was significantly decreased in IL-7-deficient recipients reconstituted with the colitogenic LP CD4+ TEM-like cells. Together with the present study, these findings suggest that the BM CD4+ TEM-like cells residing in mice with chronic colitis play a critical role as a reservoir for lifelong persisting colitis in an IL-7-dependent manner. However, it is still possible that IL-7 produced by sites other than intestine or BM, such as skin, liver, eye, lymph nodes (LN), and SP, also contribute to the development and perpetuation of colitis. In this regard, we very recently demonstrated that splenectomized LN-null lymphotoxin α−/− × RAG-2−/− mice transferred with colitogenic LP CD4+ T cells develop colitis (35), suggesting that IL-7 production at least by LN and SP does not appear to be essential. To further clarify the role of IL-7 produced by BM mesenchymal cells in the pathogenesis of chronic colitis, BM chimera of IL-7−/− × RAG-1−/− mice, which are lethally irradiated and transplanted with the BM cells from IL-7+/+ × RAG-1−/− mice, may be quite beneficial. Interestingly, however, it is also well known that extraintestinal complications of IBD patients such as skin, liver, and mucocutaneous manifestations (36) appears to be closely associated with sites of local IL-7 production by keratinocytes, hepatocytes, and uvea cells. Although no inflammation was not observed at least in liver and skin in the present model of colitis (data not shown), further studies will be needed to address this issue.
Clinicopathologically, IBD is characterized by chronic intestinal inflammation. Surgery does not cure IBD, especially Crohn’s disease, as relapse is a rule after remission, suggesting that IBD is not a circumscribed disease, but rather a systemic disease mediated by colitogenic memory CD4+ T cells distributing throughout the body via the bloodstream, which may hide in their reservoir, such as BM. Consistent with this hypothesis, recent findings showing usefulness of leukocytoapheresis, which removes peripheral blood cells for the treatment of refractory IBD patients (37, 38), suggests that recirculation of colitogenic memory CD4+ T cells from the gut to some reservoir and vice versa, may play a role in the perpetuation of chronic colitis. Furthermore, we have recently demonstrated that FTY720 that has an ability to inhibit circulation of lymphocytes prevents the development of SCID/RAG-1/2−/− colitis induced by adoptive transfer of LP colitogenic CD4+ TEM-like cells (39). Together with the current results, it would be possible that the circulation of colitogenic CD4+ TEM-like cells is quite active in IBD, making them continue to circulate in the blood and migrate to IL-7-producing reservoir from the IL-7-depleted LP.
In summary, in this study, we demonstrated that systemic IL-7, but not intestinal IL-7, is essential for the development and perpetuation of colitis, suggesting that therapeutic approaches targeting systemic IL-7 using the biologics against IL-7 may be feasible in the treatment of IBD.
We are grateful to R. Zamoyska for providing mice and K. Suzuki for his 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 study was supported in part by grants-in-aid for Scientific Research, Scientific Research on Priority Areas, Exploratory Research and Creative Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science and Technology; the Japanese Ministry of Health, Labor and Welfare; the Japan Medical Association; Foundation for Advancement of International Science; Terumo Life science Foundation; Ohyama Health Foundation; Yakult Bio-Science Foundation; and Research Fund of Mitsukoshi Health and Welfare Foundation.
Abbreviations used in this paper: IBD, inflammatory bowel disease; BM, bone marrow; LP, lamina propria; SP, spleen; Tg, transgenic; IEL, intraepithelial cell; HPF, high power field; DAPI, 4′, 6′-diamidino-2-phenylindole; LN, lymph node.