Gastroenterology
Volume 132, Issue 3 , Pages 966-981, March 2007

Identification of a Chemokine Network That Recruits FoxP3+ Regulatory T Cells Into Chronically Inflamed Intestine

  • Seung G. Kang

      Affiliations

    • Laboratory of Immunology and Hematopoiesis, Department of Comparative Pathobiology, Purdue Cancer Center, Purdue University Life Science Program, Purdue University, West Lafayette, Indiana
    • ,
  • Ronald J. Piniecki

      Affiliations

    • Laboratory of Immunology and Hematopoiesis, Department of Comparative Pathobiology, Purdue Cancer Center, Purdue University Life Science Program, Purdue University, West Lafayette, Indiana
    • ,
  • Harm Hogenesch

      Affiliations

    • Laboratory of Immunology and Hematopoiesis, Department of Comparative Pathobiology, Purdue Cancer Center, Purdue University Life Science Program, Purdue University, West Lafayette, Indiana
    • ,
  • Hyung W. Lim

      Affiliations

    • Laboratory of Immunology and Hematopoiesis, Department of Comparative Pathobiology, Purdue Cancer Center, Purdue University Life Science Program, Purdue University, West Lafayette, Indiana
    • ,
  • Eric Wiebke

      Affiliations

    • Department of Surgery, Indiana University School of Medicine, Indianapolis, Indiana
    • ,
  • Stephen E. Braun

      Affiliations

    • Division of Immunology, New England Primate Research Center, Harvard Medical School, Southborough, Massachusetts
    • ,
  • Satoshi Matsumoto

      Affiliations

    • Yakult Central Institute for Microbiological Research, Tokyo, Japan
    • ,
  • Chang H. Kim

      Affiliations

    • Laboratory of Immunology and Hematopoiesis, Department of Comparative Pathobiology, Purdue Cancer Center, Purdue University Life Science Program, Purdue University, West Lafayette, Indiana
    • Corresponding Author InformationAddress requests for reprints to: Chang H. Kim, PhD, VPTH 126, 725 Harrison Street, Purdue University, West Lafayette, Indiana 47907. fax: (765) 494-9830.

Received 16 August 2006; accepted 7 December 2006. published online 08 January 2007.

Article Outline

Background & Aims: It has been unclear which chemokine network is involved in migration of T-cell subsets to chronically inflamed lesions of the intestine. SAMP1/YP mice develop a spontaneous chronic transmural intestinal lesion specifically in the ileum. Using these mice, we investigated the gut chemokine network involved in specific migration of T-cell subsets to the inflamed lesion of the intestine. Methods: We performed expression analyses of chemokines and their receptors, chemokine receptor blocking studies, and migration studies in vitro and in vivo to identify the gut chemokine network induced in intestinal inflammation and to determine its role in migration of conventional and FoxP3+ suppressor T cells to the inflamed intestine. Results: The expression of homeostatic chemokines was largely unchanged in the inflamed lesion of SAMP1/YP mice compared with control mice. However, an additional chemokine axis (CCL5-CCR5) was up-regulated in the inflamed intestine of SAMP1/YP mice compared with control mice. Activated T cells of SAMP1/YP mice compared with control mice were hyperresponsive to CCL5 in chemotaxis. CCR5+ T cells preferentially migrated to the inflamed lesion, which can be blocked by a CCR5 antagonist. Importantly, the FoxP3+ regulatory T cells of the inflamed lesion of SAMP1/YP mice highly expressed CCR5. CCR5 blockade suppressed the migration of FoxP3+ T cells into the inflamed intestine and significantly exacerbated the intestinal inflammation. Conclusions: The CCL5-CCR5 chemokine axis is involved in preferential recruitment of FoxP3+ regulatory T cells, which prevents further exacerbation of chronic inflammation in the intestine.

Abbreviations used in this paper: CFSE, 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester, ELISA, enzyme-linked immunosorbent assay, IL, interleukin, MAdCAM-1, mucosal addressin cell adhesion molecule 1, MLN, mesenteric lymph node, PCR, polymerase chain reaction, SCID, severe combined immunodeficient, VCAM-1, vascular cell adhesion molecule 1

 

Homeostatic chemokines are expressed often in a tissue-specific manner to guide lymphocyte migration for development and effector function.1, 2, 3 In the intestine, CCL25 and CCL28 are expressed to regulate the migration of lymphocytes. CCL25 is expressed in the crypts of the small intestine4, 5 and attracts immunoglobulin A+ plasma cells6 and T cells.7 CCL28 is expressed by epithelial cells in various mucosal tissues8, 9 and attracts immunoglobulin A+ lymphocytes.10, 11, 12 CCL11 is implicated in the migration of eosinophils into the intestine.13

It has been reported that other chemokines are up-regulated in intestinal tissues with inflammatory bowel diseases (IBD). Chemokines such as CCL2, CCL4, CCL5, CCL9, CCL17, CCL19, CCL20, CXCL1, CXCL5, and CL1 are up-regulated in chronically inflamed colon tissues of interleukin (IL)-10 (−/−) mice.14 In Rag2 (−/−) mice reconstituted with CD4+CD45RBhigh T cells, CCL2, CCL5, CCL9, CCL17, CCL22, and CXCL10 are up-regulated in the mouse intestine.14 Expression of CCL17 and CCL22, along with others, has been reported in Crohn’s disease (CD) in humans.15 The functions of these proinflammatory chemokines in T-cell migration to the intestine have yet to be elucidated.

SAMP1/Yit mice spontaneously develop intestinal inflammation similar to that of human CD in many aspects: a discontinuous lesion localized in the terminal ileum and heavily infiltrated with T cells, neutrophils, and macrophages.16, 17 The localized transmural ileitis is a unique feature not shared with other mouse models of IBD.18, 19 The intestinal lesion of SAMP1/Yit mice has increased crypts and decreased numbers of intraepithelial lymphocytes.16, 20, 21 There is an increase in Th1 cell numbers, expression of IL-5, and infiltration of eosinophils.20, 22 A substrain called SAMP1/YitFc derived after more than 20 generations of inbreeding displays more accelerated ileitis, muscularis hypertrophy, ulceration, and fistulae formation.23 An initial genetic study has identified peroxisome proliferator-activated receptor γ as one of the potential susceptibility genes in SAMP1/YitFc mice.24

In this study, we characterized the altered gut chemokine network in the SAMP1/Yit mice maintained at Purdue University (referred to as “SAMP1/YP mice” to differentiate them from other mouse colonies) and investigated the role of this additional chemokine network in migration of T-cell subsets to the intestinal lesion and in intestinal inflammation. While the 2 housekeeping chemokines CCL25 and CCL28 were expressed unchanged in a segment-specific manner in the SAMP1/YP mice compared with control AKR/J mice, we found a new gut chemokine axis (CCL5-CCR5) induced in the chronically inflamed intestine. Our results suggest that this newly induced gut chemokine axis plays an important role in migration of T cells, particularly FoxP3+ regulatory T cells, to the chronically inflamed intestine.

Back to Article Outline

Materials and Methods 

Animals 

SAMP1/YP (H-2k) mice, originally derived as previously described,16 and AKR/J mice, purchased from Jackson Laboratories (Bar Harbor, ME), were maintained for up to ∼18 generations at Purdue University in conventional cages. Severe combined immunodeficient (SCID; C3HSmn.C-Prkdcscid/J) mice were purchased from Jackson Laboratories and maintained in barrier cages under a specific pathogen-free condition. All the experiments with animals in this study were approved by the Purdue Animal Care and Use Committee.

Assessment of Spontaneous Inflammation in SAMP1/YP Mice 

SAMP1/YP mice and AKR/J control mice were killed at indicated ages, and the intestine was histologically evaluated. In a subset of experiments, AKR/J and SAMP1/YP mice were treated once every 2 days for 11 weeks (from 4 to 15 weeks of age) with intraperitoneal injections of 100 μg of TAK-779 (NIH AIDS Research & Reference Reagent Program) in 100 μL of 5% d-mannitol (control medium) or the control medium only. Mice were killed, and the small intestine was fixed in neutral-buffered formalin, embedded in paraffin, and stained with H&E. For histologic assessment, the severity of tissue infiltration by mononuclear cells and polymorphonuclear cells, numbers of goblet cells, and thickness of crypts and muscularis propria in affected lesions of SAMP1/YP mice and control tissues of AKR/J mice were scored in 10 randomly selected high-power fields to calculate the average scores for each lesion.

Assessment of Induced Intestinal Inflammation in SCID Mice 

For the adoptive transfer model, adult female major histocompatibility complex–matched SCID mice were used as recipients in the following experiments. Mesenteric lymph node (MLN) cells (1 × 106 cells/mouse) of SAMP1/YP mice (∼30 weeks old) were injected intraperitoneally into 6-week-old SCID mice to induce intestinal inflammation. Some mice were injected with control medium. TAK-779 was injected as described previously. Mice were monitored for weight change, and all mice were killed 6 weeks after cell transfer or when mice with severe inflammation became moribund and/or lost >20% of their initial weight. The intestines were removed and evaluated histologically. The intestinal inflammation in SCID mice was scored by combining the severity scores (0–4; 0, no inflammation; 1, slight infiltration of inflammatory cells in the lamina propria; 2, moderate infiltration with inflammatory cells with mild mucosa hyperplasia; 3, marked infiltration with inflammatory cells with disturbed mucosal architecture, crypt abscesses, and marked hyperplasia; and 4, massive infiltration with inflammatory cells with severe mucosa hyperplasia) and the area scores (0–4; 0, no area affected; 1, 1%–25%; 2, 26%–50%; 3, 51%–75%; and 4, 76%–100%).

RNA Expression Analyses of Chemokines and Chemokine Receptors 

Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA) from the terminal ileum, Peyer’s patches, and MLNs of 30-week-old AKR/J or SAMP1/YP mice. The RNA was reverse transcribed using SuperScript First-Strand Synthesis System (Invitrogen) according to the manufacturer’s protocol. Serially diluted complementary DNA samples were analyzed for expression of a panel of chemokines by real-time polymerase chain reaction (PCR) using SYBR Green (BMA, Rockland, ME) on an ABI PRISM 5700 sequence detection system (Applied Biosystems, Foster City, CA). The expression levels of chemokines in each sample were assessed after normalization for β-actin. For conventional PCR, complementary DNA samples were serially diluted and first analyzed by PCR (20 cycles) for β-actin to determine appropriate dilution factors for each sample and then PCR analyzed for genes of interest. A total of 5–20 μg of total RNA was assayed by the RiboQuant Multi-Probe RNase Protection Assay System (BD Biosciences, San Diego, CA).

In Situ Fluorescent Immunohistochemistry 

Terminal ileum tissues from 20- to 30-week-old AKR/J or SAMP1/YP mice frozen in Tissue-Tek freezing medium (Miles, Elkhart, IN) were made into 5-μm sections. Frozen sections were fixed in cold acetone and blocked with 10% rabbit serum for 20 minutes at room temperature. Sections were stained with goat anti-mCCL5 antibody (10 μg/mL, AF478; R&D Systems, Minneapolis, MD) or goat control immunoglobulin G antibody (Sigma-Aldrich, St Louis, MO) for 1 hour at 4°C. The sections were further incubated with biotinylated anti-goat immunoglobulin G (Vector Laboratories, Burlingame, CA). Sections were blocked with 5% goat serum and further stained with streptavidin-phycoerythrin, anti–CD4/fluorescein isothiocyanate (L3T4), or anti–CD8/fluorescein isothiocyanate (53-6.7). The sections were further stained with Hoechst 33342 and examined with a Nikon E400 microscope (Tokyo, Japan) equipped with epifluorescence.

Assessment of CCL5 Expression by Enzyme-Linked Immunosorbent Assay 

CD4+ or CD8+ T cells were isolated from MLNs by magnetic beads (Miltenyi Biotec, Auburn, CA; ∼90% pure). The MLN CD4+ or CD8+ T cells (0.4 × 106 cells/well) were cultured in 96-well plates (BD Biosciences, San Jose, CA) coated with 10 μg/mL of anti-CD3 (17A2; BD Biosciences) for 2 days in a co2 incubator. When indicated, anti-CD28 (37.51; BD Biosciences; 2 μg/mL) was added to provide costimulation signals. The conditioned culture medium was examined by CCL5 enzyme-linked immunosorbent assay (ELISA). Additionally, tissue proteins were extracted from the terminal ileum tissues of 30-week-old SAMP1/YP and age-matched control AKR/J mice by grinding the tissues in liquid nitrogen using a mortar and a pestle followed by homogenization using zirconium beads in phosphate-buffered saline buffer. A protease inhibitor cocktail (Sigma-Aldrich) was added to prevent degradation of proteins during and after homogenization. Serially diluted homogenates were examined for CCL5 protein levels by a 3-step sandwich ELISA (R&D Systems). The total protein concentrations were determined by a bicinchoninic acid method (Pierce, Rockford, IL).

Chemotaxis Assay 

The chemotaxis assay was performed by a 2-chamber Transwell migration assay system (pore size, 5.0 μm; Corning, Corning, NY) as previously described.25, 26, 27 MLN cells were isolated from 30-week-old AKR/J or SAMP1/YP mice and activated with concanavalin A (2.5 μg/mL; Sigma Chemical Co) for 2 days and then cultured in murine IL-2 (5 ng/mL) for an additional 2 days. A total of 0.5 × 106 cells/well was added to the upper chambers. In some experiments, TAK-779 (1.0 μg/mL) was added to both chambers. After 3 hours, the cells that migrated to the lower chambers were harvested, stained with anti–CD4/APC and anti–CD8/cychrome, and counted by FACScalibur (BD Biosciences).

Retroviral Expression of Chemokine Receptors and Chemokine–Chemokine Receptor Interaction 

Chemokine receptor–overexpressing T cells were generated by a retroviral transfer method.28 CD4+ T cells were activated for 36 hours with concanavalin A and IL-2 (5 ng/mL; R&D Systems). CD4+ T cells were then transduced with retrovirus made in Phoenix-ecotropic retrovirus packaging cells (a gift from Dr Gary Nolan, Stanford University). Retroviral gene transfer rates were assessed based on green fluorescent protein expression by fluorescence-activated cell sorter analysis. A pCDNA3.1 vector for expression of CCL5-Fc fusion protein was constructed12 and transfected into HEK-293 cells for production of CCL5-Fc fusion protein in 3-day-old culture supernatant. Biotinylated anti-human immunoglobulin G and streptavidin-APC were used to detect the binding of CCL5-Fc proteins to cell surface chemokine receptors.

Expression of FoxP3 and CCR5 by Intestinal T Cells 

The lamina propria lymphocytes were isolated after removing the epithelial cells with 5 mmol/L EDTA (5 times), digestion (3 times, 45 minutes each) with 300 U/mL collagenase (type 3; Worthington, Lakewood, NJ) and 100 μg/mL deoxyribonuclease I (Worthington), filtration through a nylon mesh, and centrifugation in a 40/75% Percoll gradient. The cells, isolated from the intestine, Peyer’s patches, and MLNs, were stained with antibodies to CCR5 (C34-3448; BD Biosciences, San Jose, CA), CD4, and FoxP3 (FJK-16s; eBioscience) for flow cytometric determination of CCR5 expression and frequency of FoxP3+ cells.

T-Cell Homing Into Inflamed Intestine 

Test CCR5+ CD4+ T cells were generated by retroviral gene transfer as described previously from the MLN cells of SAMP1/YP mice. The test T cells and control T cells (untransduced) were cultured in the presence of retinoic acid (10 nmol/L; Sigma-Aldrich) to induce the gut-homing phenotype.29 Control T cells were labeled with the red fluorescent dye tetramethylrhodamine-5-(and-6)-isothiocyanate, while CCR5+ CD4+ T cells were labeled with the green dye 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester (CFSE). The 2 T-cell subsets (107 cells per mouse) were coinjected into SAMP1/YP mice via a tail vein. T cells were isolated from various tissues of the SAMP1/YP or AKR/J mice 20–24 hours after injection and analyzed by flow analysis. We also generated gut-homing FoxP3+ (and FoxP3−) CD4+ T cells by retroviral transduction of murine FoxP3 gene into spleen T cells in the presence of retinoic acid. For this, spleen T cells from AKR/J mice were activated for 36 hours as described previously and infected with FoxP3 retrovirus. Cells were further cultured for 4 days in the presence of retinoic acid and cytokines (5 ng/mL of IL-2 alone or IL-2 and 2 ng/mL of IL-12). Naive CD4+ T cells become regulatory T cells when they are transduced with the FoxP3 gene.30, 31 The cells (∼2 × 107 cells per mouse) were labeled with CFSE and injected intravenously into SAMP1/YP mice for 20- to 24-hour homing. Lamia propria cells were isolated and stained for CD4, FoxP3, and CCR5 as described previously. Normalized homing indexes were calculated by the formula HI = a ÷ b, where a represents [Test Cells Migrated Into the Tissue Site] ÷ [Control Cells Migrated Into the Tissue Site] and b represents [Test Cells in Input] ÷ [Control Cells in Input].

In Vitro Suppression of Target T Cells by CD4+CD25+ Cells 

CD4+CD25− T cells (responders) and CD4+CD25+ T cells (suppressors) were isolated from MLNs of AKR/J and SAMP1/YP mice and cocultured for 72 hours at indicated ratios in the presence of anti-CD3 (5 μg/mL; 145-2C11; eBioscience) and 1 × 105 irradiated splenocytes as antigen-presenting cells in U-bottomed 96-well plates. Cells were further incubated with 1 μCi/well of 3H-thymidine for 8 hours, and 3H-thymidine incorporation was measured by a beta scintillation counter (Packard Instruments, Meriden, CT).

Statistical Analyses 

Student t test (paired or unpaired; 2-tailed) was used to determine the significance of the differences between 2 sets of related data. Statistical significance was set at P values less than .05.

Back to Article Outline

Results 

The Gut Chemokine Network of SAMP1/YP Mice 

As previously reported,16, 20 the SAMP1/YP mice develop a CD-like lesion with the thickening of the lamina propria infiltrated with immune cells in the ileum segment (Figure 1A and B). The ileitis started early at 4–5 weeks of age and became more severe with age. Most mice had advanced ileitis at 10 weeks of age. The ileitis was characterized by severe discontinuous transmural inflammation, localized in the terminal ileum (the last 7–8 cm before the ileocecal junction), with mild inflammation extending into the proximal ileum in some mice. Bowel wall thickening and constriction of the lumen in the terminal ileum, caused by inflammation and hyperplasia of the mucosa, submucosa, and muscularis externa, were readily detected grossly. The lesions were more opaque than uninflamed segments and devoid of digested food content due to bowel thickening and constriction. Inflamed lamina propria of SAMP1/YP ileum was heavily infiltrated with mononuclear and polymorphonuclear leukocyte cells. Infiltration of polymorphonuclear leukocytes and mononuclear cells in the submucosa and muscularis externa was relatively mild compared with that in the lamina propria. Hyperplasia of the crypts and fusion of villi were observed in the lesions, and goblet cells were greatly increased in number (Figure 1B). Mice with intestinal fistulae were rare. Histologic analysis of intestinal segments other than ileum revealed no apparent inflammation (Figure 1C). MLNs of SAMP1/YP mice were enlarged and fused together. The results indicate that the inflammation was more advanced than in the mice described in the original report16 and similar or slightly less advanced than in the SAMP1/YitFc mice described by Rivera-Nieves et al.23

  • View full-size image.
  • Figure 1. 

    Kinetics and localization of chronic intestinal inflammation in SAMP1/YP mice. (A) The histopathology of SAMP1/YP mice at various ages (original magnification 100×). (B) Infiltration of mononuclear cells and polymorphonuclear cells, hyperplasia of goblet cells, and thickening of muscularis externa in the terminal ileum were measured with age. Combined data were obtained from 10 high-power fields per affected tissue site and mouse, and data from 4–6 mice were combined for each time point. (C) The inflammation is limited to the ileum of ∼30-week-old SAMP1/YP mice. Intestinal tissue sections were stained with H&E (original magnification 100×), and representative areas are shown from 4–6 mice for each group.

SAMP1/YP mice were originally derived from AKR/J mice. Therefore, AKR/J mice have commonly been used as normal controls for SAMP1/YP mice. We assessed the basal expression levels of gut chemokines in AKR/J mice and SAMP1/YP mice. Intestinal tissues were divided into 4 parts: jejunum, ileum, cecum, and colon. A quantitative reverse-transcription polymerase chain reaction technique was used to examine the message levels of 19 major CC and CXC chemokines (Figure 2A). Conventional PCR was also performed for selected chemokines highly expressed in the intestine (Figure 2B and C). The most notable chemokine, highly expressed in the intestine of the control mice, was CCL25. CCL25 is specifically expressed in the jejunum and ileum but not expressed (or expressed only at low levels) in the cecum and colon. The expression of another chemokine, CCL28, was largely restricted to the colon. In addition, the expression of CXCL13 and CXCL12 at the RNA level was observed in the colon (Figure 2A). However, the expression of inflammatory chemokines such as CXCL11, CCL17, CXCL16, CXCL10, CCL4, CCL1, CCL5, and CCL3 was minimal in all 4 parts of the normal intestine of AKR/J mice. In SAMP1/YP mice, CCL25 expression was observed in the jejunum and ileum but not in the cecum and colon, an expression pattern similar to that of AKR/J mice (Figure 2B). Also, CCL28 expression in the control and SAMP1/YP mice was similar to each other, with the highest expression in the colon. CCL5 was the chemokine specifically up-regulated in the jejunum, ileum, and cecum of the SAMP1/YP mice compared with AKR/J control mice, with the highest expression in the ileum and terminal ileum, where the lesion is localized. Increased expression of CCL5 in the inflamed ileum of SAMP1/YP mice was not detected at 3 weeks but detected at 5 weeks of age. Overall, the CCL5 expression was gradually increased with age (Figure 2C). We also used a ribonuclease protection assay method to detect CCL5 expression, and CCL5 was highly up-regulated in the intestinal lesion of SAMP1/YP versus AKR/J mice (Figure 2D). However, CCL5 expression in gut-associated lymphoid tissues such as Peyer’s patches and MLNs was similar in the 2 mouse strains.

  • View full-size image.
  • Figure 2. 

    The gut chemokine network of SAMP1/YP mice. AKR/J control mice and SAMP1/YP mice were examined by (A) real-time reverse-transcription PCR, (B and C) conventional reverse-transcription PCR, and (D) ribonuclease protection assay. Thirty-week-old mice were used for A, B, and D. A representative set of data from at least 3 independent experiments using different mice is shown.

Intestinal CD8+ T Cells Produce CCL5 in SAMP1/YP Mice in Response to T-Cell Activation Signals 

We further examined the expression of CCL5 at the protein level using an ELISA technique (Figure 3A). Terminal ileum tissues from SAMP1/YP and control mice were made into protein extracts and examined for CCL5 protein levels by a 3-step sandwich ELISA. The CCL5 expression in the terminal ileum tissues was up-regulated by 10- to 30-fold in SAMP1/YP mice compared with control mice.

  • View full-size image.
  • Figure 3. 

    CCL5 protein is overexpressed in the inflamed intestinal lesion of SAMP1/YP mice. (A) CCL5 protein is up-regulated in the terminal ileum of SAMP1/YP mice compared with AKR/J mice (n = 4). Terminal ileum tissues were examined for CCL5 protein by ELISA. CCL5 expression levels were normalized per milligram of total protein. (B) Frozen sections of terminal ileum tissues from SAMP1/YP mice (30 weeks old) were examined for CCL5 expression by immunohistochemistry. Representative data from 8 separate experiments are shown.

We next examined the expression of CCL5 in the CD-like lesion of SAMP1/YP mice by an immunofluorescence histochemistry technique. Frozen sections of terminal ileum tissues were stained with fluorescently labeled antibodies to CD4 or CD8 and with anti-CCL5 antibody (Figure 3B). CCL5 expression was detected in the mucosa layer of the CD-like lesion. It was particularly evident in the lamina propria (Figure 3B). Approximately half of the CCL5+ cells were CD8+ cells, but not all CD8+ cells were CCL5+. Only small numbers of CCL5+ CD4+ cells or CCL5+ F4/80+ cells (not shown) were detected in the lamina propria. These results suggest that CD8 T cells constitute a major source of the CCL5 up-regulated in the inflammatory lesion.

To more definitely determine that intestinal T cells produce CCL5, we isolated T cells from the MLNs of SAMP1/YP mice and AKR/J control mice. We cultured them in a T-cell activation condition and measured the concentration of CCL5 secreted by CD4+ or CD8+ T cells (Figure 4). The CD8+ T cells isolated from SAMP1/YP mice were highly efficient in production of CCL5, while their counterparts from control AKR/J mice were relatively inefficient (Figure 4A and B). The efficient production capacity of CCL5 by CD8+ T cells was evident over a 3-day culture period. However, there was no difference in the CCL5 production capacity of the CD4+ T cells isolated from SAMP1/YP mice and AKR/J mice (Figure 4C). The CCL5 production by CD8+ T cells was dependent on T-cell receptor activation, suggesting that CD8+ T cells can produce CCL5 in the intestinal lesion of SAMP1/YP mice upon antigenic stimulation.

  • View full-size image.
  • Figure 4. 

    CCL5 protein is overproduced by intestinal CD8+ cells in SAMP1/YP mice upon T-cell activation. MLN T cells were cultured for 48 hours, and the culture media were assayed for CCL5 by ELISA. T cells were cultured in the presence of (A) the polyclonal activator concanavalin A for indicated time or (B and C) anti-CD3 and/or anti-CD28 for 24 hours followed by ELISA to quantitate CCL5 produced in culture media. (A) One representative kinetic datum is shown, or (B and C) data from 3 independent experiments were combined and averages and SE are shown.

Activated T Cells of SAMP1/YP Mice Efficiently Migrate to CCL5 

We assessed the chemotactic behavior of the T cells activated after isolation from the MLNs of SAMP1/YP and control mice to determine their chemotactic responsiveness to CCL5. The CD4+ T cells originated from SAMP1/YP mice were more responsive to CCL5 in chemotaxis than those from AKR/J mice (Figure 5). CD8+ T cells from SAMP/YP mice were also more responsive to CCL5 than their counterparts from control AKR/J mice. MLN T cells isolated from the control and SAMP1/YP mice were cultured in the presence of IL-2 + IL-12 under a T-cell activation condition to assess the development of their migration potential to CCL5. When cultured in the presence of IL-2 + IL-12, both AKR/J and SAMP1/YP CD4+ T cells migrated to CCL5 at a level similar to each other, suggesting that there is no intrinsic difference in the development of the migration potential in the 2 mouse strains.

  • View full-size image.
  • Figure 5. 

    Activated T cells of SAMP1/YP mice are more responsive to CCL5 in chemotaxis than their counterparts from control mice. (A) In vitro chemotaxis assays were performed using the Transwell system (5-μm pores) with T cells isolated from MLNs of SAMP1/YP mice and AKR/J mice and activated for 4 days in the presence of IL-2. (B) T cells were cultured in the presence of IL-2 and IL-12 for 4 days and examined for their chemotactic responses to CCL5 (100 ng/mL). Data from 3 independent experiments were combined, and averages and SE are shown. *Significant differences between T cells from AKR/J and SAMP1/YP mice.

CCR5 Is a Major Receptor for CCL5, and TAK-779 Specifically Blocks Their Interaction 

Because CCL5 is highly expressed in the intestinal lesion of SAMP1/YP mice, we examined the expression of its receptors in the intestine. CCR5 was up-regulated in the small intestine of SAMP1/YP mice compared with control mice, while CCR1 was comparably expressed at the RNA level in the diseased and control mice (Figure 6A).

  • View full-size image.
  • Figure 6. 

    CCR5 is the main receptor for CCL5, and the CCR5-CCL5 interaction is specifically blocked by TAK-779. (A) Reverse-transcription PCR analysis of CCR1 and CCR5. Representatives from 3 independent experiments are shown. (B) The TAK-779 effect on CCL5 binding to chemokine receptors. Chemokine receptors were overexpressed in MLN CD4+ T cells prepared from AKR/J mice after retroviral gene transfer. CCL5-Fc protein was examined for its binding capacity to the chemokine receptors. TAK-779 was tested for its blocking effect on interactions between the CCL5-Fc protein and chemokine receptors. (C) TAK-779 was also tested for its blocking effect on the migration of CCR5+ CD4+ T cells generated by retroviral gene transfer. Representative data from 3 independent experiments (error bars are differences of 2 measurements).

CCL5 is known to bind multiple chemokine receptors at varying affinities. We examined the specificity and affinity of CCL5 binding to the chemokine receptors using CD4+ T cells overexpressing a particular chemokine receptor and chemokine-Fc proteins in an effort to determine the major CCL5 receptor on T cells. As positive binding controls for the chemokine receptors, other chemokines such as CXCL9, CCL25, CCL21, and CCL27 were tested along with CCL5. CCL5 did not bind CCR3, CCR7, CXCR3, CCR9, and CCR10 at detectable levels (Figure 6B or not shown). However, clear binding of CCL5 to CCR5 was observed. Weak binding to CCR1 was also noted but relatively insignificant.

TAK-779 is the small molecule compound that blocks CCL5 binding to CCR5 but has minimal or no effect on CCR2b, CCR1, CCR3, or CCR4.32 It binds specifically to a cavity formed between transmembrane helices 1, 2, 3, and 7 on the extracellular surface of CCR5.33 We tested the specificity and efficacy of TAK-779 before we used this small molecule antagonist for in vivo experiments. TAK-779 inhibited the binding of CCL5 to CCR5 but not to other chemokine receptor systems such as CXCL9-CXCR3, CCL27-CCR10, CCL21-CCR7, and CCL25-CCR9 (Figure 6B or not shown). TAK-779 also showed a dose-dependent blocking effect on the CCL5-mediated migration of CCR5+ CD4+ T cells (Figure 6C). When TAK-779 was added to T-cell culture, it did not change the T-cell proliferation induced by T-cell receptor activation and IL-2, activation-induced T-cell death, and T-cell polarization driven by T-cell receptor activation and IL-12 (not shown). The optimal concentration of TAK-779 to block the interaction of CCL5 and CCR5 was ∼0.5 μg/mL. These results show that CCR5 is the major receptor for CCL5 and is specifically inhibited by TAK-779. Based on this information, we examined the blocking effect of TAK-779 on T-cell migration to the inflamed gut lesion versus other tissue sites as described in the following text.

CCR5 Drives T-Cell Migration to the CCL5-Expressing Intestinal Lesion 

It is practically impossible to isolate enough numbers of gut-homing CCR5+ T cells from the intestine or secondary lymphoid tissues to determine the role of CCR5 in T-cell migration in vivo. Therefore, we generated the test CCR5+ and control CCR5− T cells by retroviral gene transfer (Figure 7A). The T-cell subsets were cultured in the presence of retinoic acid in a T-cell activation condition to induce the gut-homing phenotype in T cells.29 Both the test CCR5+ and control CCR5− T cells were α4β7+ and CCR9+ (Figure 7A). We injected the test T cells intravenously along with the control T cells into SAMP1/YP mice. Labeled T cells were allowed to migrate for 20–24 hours. Both the test and control T cells migrated to the MLNs, peripheral lymph nodes, the spleen, the lamina propria of healthy intestinal segment, and the peritoneal cavity at similar rates, thus giving homing indexes close to 1 (Figure 7A and B). However, the test T cells migrated significantly better than control T cells to the inflamed intestinal lesion of SAMP1/YP mice (Figure 7B). The migration to the adjacent noninflamed jejunum segment was detected but relatively inefficient. In contrast, the increased CCR5+ T-cell migration to the intestine did not occur in AKR/J mice. The test T cells that migrated to the lesion and other tissue sites were examined for CCR5 expression (Figure 7C). A high level of CCR5+ T-cell enrichment was noted in the inflamed terminal ileum but not in the normal intestinal segment or in other organs. This enrichment of CCR5+ T cells was effectively suppressed by the CCR5 blocker TAK-779.

  • View full-size image.
  • Figure 7. 

    CCR5+ T cells preferentially migrate to the inflamed lesion. (A) CCR5+ CD4+ T cells, generated by retroviral gene transfer from the MLN cells of SAMP1/YP mice, were labeled with CFSE, while CCR5− control CD4+ T cells were labeled with tetramethylrhodamine-5-(and-6)-isothiocyanate. The 2 T-cell subsets were coinjected into 25- to 30-week-old SAMP1/YP or AKR/J mice via a tail vein. TAK-779 (100 μg/mouse) or control medium (5% d-mannitol) was injected intraperitoneally 12 hours before and right after the cell injection. T cells were isolated from various tissues of the mice 20–24 hours postinjection and analyzed by flow cytometry analysis. The tissues examined were spleen, MLNs, peripheral lymph nodes, Peyer’s patches (PP), lamina propria of a normal intestinal segment (the distal part of jejunum right before ileum [LP(N)]), lamina propria of the inflamed terminal ileum lesion (the last 7–8 cm of ileum before the ileocecal junction [LP(I)]), and peritoneal cavity. (B) Homing indexes were calculated according the formula described in Material and Methods. Migration of CCR5+ CD4+ T cells in SAMP1/YP and AKR/J mice is compared. Representative data from 3 independent experiments. Significant differences (P < .05) from TAK-779 (*) and AKR/J controls (**). (C) CD4+ T cells isolated from various tissue sites were stained for CCR5 surface expression.

Exacerbation of the Intestinal Inflammation by CCR5 Blockade 

Next, we treated SAMP1/YP mice with TAK-779 for 11 weeks (from 4 to 15 weeks of age) and examined the effect on intestinal inflammation (Figure 8). We started the injection at 4 weeks of age because SAMP1/YP mice start to develop mild inflammation at this age. Surprisingly, the intestinal inflammation in SAMP1/YP mice (indicated by infiltration of polymorphonuclear cells and mononuclear cells, numbers of goblet cells, thickness of muscularis externa) was significantly exacerbated compared with the SAMP1/YP mice injected with the control medium (Figure 8A and B). In control AKR/J mice, the TAK-779 treatment did not cause significant inflammation. Because the results with SAMP1/YP mice were unexpected, we used another animal model to investigate the effect of CCR5 blockade on intestinal inflammation. MLN cells of SAMP1/YP mice, when transferred to major histocompatibility complex–matched SCID mice, can induce intestinal inflammation and weight loss.20 Although induced by SAMP1/YP mouse cells, this model is different from the SAMP1/YP mouse model in that the large intestine is more inflamed than the small intestine. We induced the inflammation in SCID mice by transferring SAMP1/YP cells and treated the SCID mice with long-term (6 weeks) intraperitoneal injection with TAK-779 or control medium. Then, we examined the severity of weight loss and intestinal inflammation. TAK-779 treatment significantly exacerbated the weight loss (Figure 8C) and the inflammation in the large intestine (Figure 8D and E). Although less pronounced, the exacerbation of inflammation was also detected in the small intestine (not shown).

  • View full-size image.
  • Figure 8. 

    CCR5 blockade exacerbates the intestinal inflammation. (A and B) SAMP1/YP mice were injected intraperitoneally with TAK-779 (n = 8) or control medium (n = 7) for 11 weeks starting from 4 weeks of age. Similar numbers of age-matching AKR/J mice were also treated in the same manner. (C and D) SCID mice were injected with MLN cells of SAMP1/YP mice to induce inflammation and then treated with TAK-779 or control medium every 2 days for 6 weeks. Control mice did not receive MLN cells. Weight loss due to inflammation was examined (C). The average number of mice (n) for each group was 10. Tissue sections (distal colon) were stained with H&E and scored for inflammation (D and E). *Significant differences (P < .05) from controls.

FoxP3+ T Cells in the Chronically Inflamed Lesion Highly Express CCR5 

The unexpected results described in Figure 8 prompted us to examine CCR5 expression by FoxP3+ suppressor T cells (Figure 9A), because it is possible that FoxP3+ suppressor T cells would preferentially use the CCL5-CCR5 chemokine axis to migrate into the inflamed lesions for suppression. Approximately 10% of FoxP3+ T cells of MLNs and Peyer’s patches of young AKR/J and SAMP1/YP mice expressed CCR5. While only ∼25% of FoxP3+ T cells in the intestine of AKR/J mice expressed CCR5, the majority (∼70%) of FoxP3+ T cells in the inflamed lesion of SAMP1/YP mice were CCR5+. Moreover, ∼50% of FoxP3+ T cells in the histologically normal jejunum of SAMP1/YP mice expressed CCR5.

  • View full-size image.
  • Figure 9. 

    CCR5+FoxP3+ T cells are greatly enriched in the inflamed intestine of SAMP1/YP mice. (A) The majority of FoxP3+ cells in the inflamed intestine of SAMP1/YP mice are CCR5+. FoxP3+ and Foxp3− T cells, harvested from various organs of SAMP1/YP mice (total, n = 14) and AKR/J mice (total, n = 9) at 3, 7, and 30 weeks of age, were examined for expression of CCR5. (B) FoxP3+ cell frequencies among the CD4+ T cells of the intestine and MLNs of SAMP1/YP and AKR/J mice were determined. (C) CD4+CD25+ T cells of the MLNs of SAMP1/YP and AKR/J mice are equally competent in their suppressive activity in vitro on proliferation of CD4+CD25− T cells. Indicated numbers of CD4+CD25+ T cells were added to the cultures of CD4+CD25− T cells. A representative set of data obtained from 3 independent experiments are shown. *Significant differences (P < .05) from controls.

While the proportion of CCR5+ cells among FoxP3+ T cells was increased in the inflamed tissues of SAMP1/YP mice as described previously, the frequency of total FoxP3+ T cells among CD4+ T cells (or the relative ratio of FoxP3+ to FoxP3− CD4+ T cells) was significantly reduced in the inflamed lesion of SAMP1/YP mice but not in the intestine of AKR/J mice (Figure 9B). The FoxP3+ cell frequency was also transiently reduced in a histologically normal intestinal segment (the distal part of the jejunum right before the ileum) of SAMP1/YP mice at 7 weeks but was regained later at 30 weeks of age. FoxP3+ T-cell numbers were not decreased in MLNs and other lymphoid tissues. While the relative ratio of FoxP3+ to FoxP3− T cells was decreased in the intestine of SAMP1/YP mice, the FoxP3+ cells of SAMP1/YP mice were not functionally defective. MLN CD4+ CD25+ T cells (most of these cells were FoxP3+ cells) isolated from AKR/J and SAMP1/YP mice were similar to each other in their in vitro suppressive function for CD4+ CD25− T cells (Figure 9C).

FoxP3+ T Cells Are Recruited to the Inflamed Intestine of SAMP1/YP Mice in a CCR5-Dependent Manner 

We next determined if the transcription factor FoxP3 is important for expression of CCR5 and if FoxP3+ T cells use the CCL5-CCR5 chemokine axis to migrate into inflamed intestine of SAMP1/YP mice. For this, the mFoxP3 gene was transferred into CD4+ T cells with a retroviral vector. On average, 5.6% of FoxP3+ (FoxP3-transduced) and 1.8% of FoxP3− (FoxP3-nontransduced) CD4+ T cells, cultured in IL-2, expressed CCR5 (n = 6, significance difference between FoxP3+ and FoxP3− cells, P = .002). Thus, many more FoxP3+ than FoxP3− CD4+ T cells expressed CCR5. When cultured in IL-12 (a Th1 inflammatory signal), this difference between FoxP3+ and FoxP3− CD4+ T cells became greater (Figure 10A). When injected intravenously, FoxP3+ T cells migrated better into the intestine, particularly the inflamed lesion, than FoxP3− CD4+ T cells. However, the migration rates of FoxP3+ and FoxP3− T cells into MLNs and peripheral lymph nodes of SAMP1/YP mice were comparable to each other (Figure 10B). The CCR5 antagonist, TAK-779, effectively suppressed the migration of FoxP3+ T cells into the inflamed intestine of SAMP1/YP mice, suggesting that FoxP3+ T cells use the CCR5-CCL5 axis to migrate into the inflamed intestine (Figure 10C).

  • View full-size image.
  • Figure 10. 

    Preferential migration of FoxP3+ T cells into the inflamed intestine in a CCR5-dependent manner. (A) FoxP3 conditions T cells to overexpress CCR5 in response to IL-12. The mFoxP3 gene was retrovirally transferred into MLN CD4+ T cells. The T cells were cultured in the presence of IL-12 and retinoic acid, and expression of CCR5 was determined on the surface of transduced FoxP3+ and untransduced FoxP3− CD4+ T cells. (B) FoxP3+ CD4+ T cells migrate better than FoxP3− CD4+ T cells to inflamed intestine. Retrovirally generated FoxP3+ T cells were labeled with CFSE and injected intravenously into SAMP1/YP mice (6–7 weeks old). At 20 hours after cell injection, migration of FoxP3+ T cells into various organs was examined. FoxP3+ and FoxP3− CD4+ T cells, migrated into various tissue sites, were identified by their staining patterns of CFSE and FoxP3. Homing index (HI) for this panel was calculated with the formula based on frequencies of the cells of interest determined by flow cytometry: HI = a ÷ b, where a represents (Migrated CFSE+FoxP3+ Cells Into Tissue Site A) ÷ (Migrated CFSE+ FoxP3− Cells Into Tissue Site A) and b represents (Injected FoxP3+ Cells in Input) ÷ (Injected FoxP3− Cells in Input). Data from 4 independent experiments (n = 4 in each group) were combined. (C) CCR5 blockade by TAK-779 suppresses FoxP3+ T cell migration to inflamed intestine. TAK-779 (or control medium) was injected intraperitoneally into SAMP1/YP mice 12 hours before and at the time of the injection of the cells to block the CCR5 pathway. Data from 3 separate experiments were combined (n = 3), and averages and SEM are shown. Percent HI was calculated using the formula: % HI = (a ÷ b) × 100, where a represents (Migrated CFSE+FoxP3+ Cells Into Tissue Site A of TAK-779–Injected Animals) ÷ (Resident CD4+ Cells in Tissue Site A of TAK-779–Injected Animals) and b represents (Migrated CFSE+FoxP3+ Cells Into Tissue Site A of Control-Injected Animals) ÷ (Resident CD4+ Cells in Tissue Site A of Control-Injected Animals).

Back to Article Outline

Discussion 

Unlike other IBD mouse models, SAMP1/YP mice develop fairly specific intestinal lesions in the ileum, mimicking CD in humans. This provided us the opportunity to study the differences between the inflamed and normal intestinal segments in chemokine expression profiles and T-cell recruitment characteristics. Our results show the presence of an inflammatory chemokine network in SAMP1/YP mice in addition to the largely undisturbed homeostatic gut chemokine network composed of CCL25 and CCL28. CCL5 was a major chemokine constituting the inflammatory gut chemokine network in SAMP1/YP mice. T cells of SAMP1/YP mice, particularly FoxP3+ T cells in lamina propria of the inflamed intestinal lesion, highly express CCR5 and migrate to CCL5. Using a gene transfer strategy, we also found that the transcription factor FoxP3 makes T cells highly efficient in up-regulation of CCR5 in response to an inflammatory cytokine signal (IL-12). CCR5 efficiently drives T-cell migration to the inflamed intestine but poorly to other tissue sites. CCR5 blockade in vivo exacerbated intestinal inflammation in SAMP1/YP and SCID mice, and the CCL5-CCR5 chemokine axis recruits FoxP3+ T cells to the lesions, suggesting that this recruitment of FoxP3+ T cells has significant suppressive effects on the chronic intestinal inflammation.

As previously reported,4, 5, 6, 8, 9, 10, 11, 12 the intestinal tract can be divided into distinct chemokine domains. Based on CCL25 expression, the intestine can be divided into CCL25high and CCL25−/low intestinal segments. The CCL25high domain includes the jejunum and ileum segments, while the CCL25−/low domain includes the colon and other parts of the large intestine. CCL28 expression is largely restricted to the large intestine. The cecum does not belong to the CCL25high or CCL28high domain. CCL11, an eosinophil-specific chemokine, is expressed throughout the entire intestinal tract. While the expression of the homeostatic chemokines, CCL25 and CCL28, is largely unaltered in SAMP1/YP mice, CCL11 is elevated in the intestine of SAMP1/YP mice compared with control AKR/J mice. Up-regulation of CCL11 in the intestinal lesion of SAMP1/YP mice is consistent with the up-regulated IL-5 and CCR3 and accumulation of eosinophils in the lesion.22 These homeostatic chemokines are believed to recruit necessary immune cells such as immunoglobulin A+ plasma B cells, intraepithelial CD8+ T cells, α4β7+ T cells, and/or eosinophils to the gut environment. We found that the intestinal lesion of SAMP1/YP mice has an inflammatory chemokine network mainly composed of CCL5 and its receptor CCR5. T cells, particularly CD8+ memory/effector T cells, appear to be responsible for the overexpression of CCL5 in the intestinal lesion. The CCL5 production by intestinal CD8+ T cells of SAMP1/YP mice and control mice was dependent on T-cell activation. Therefore, the activated T cells in the chronically inflamed intestine of SAMP1/YP mice appear to be a major source of the overproduced CCL5 (Figure 3, Figure 4). Although CCL5 has several known chemokine receptors, we found that CCR5 is the major receptor for CCL5. Expression of CCR5 is also elevated in the ileum and terminal ileum segments of SAMP1/YP mice. SAMP1/YP T cells, compared with AKR/J T cells, were more responsive to CCL5 in chemotaxis. Taken together, these results suggest the presence of the CCL5-CCR5 chemokine axis in the inflamed intestinal lesion of SAMP1/YP mice. Induction of this chemokine network in human IBD has been well established; Mazzucchelli et al reported that CCL5-expressing cells are increased in number,34 and Oki et al reported that CCR5+ T cells are accumulated around the CCL5-producing cells in intestinal lesions of human IBD, including CD.35 Moreover, the CCL5-CCR5 is increased in other mouse models of IBD,14 suggesting that it is a universal chemokine axis induced in IBD.

For in vivo homing experiments, we used MLN T cells gene transferred with CCR5 and cultured in retinoic acid to generate highly enriched CCR5+ gut-homing T cells. These CCR5+ T cells were found to preferentially migrate to the inflamed terminal ileum but not to other organs. Lower levels of CCL5 expression were also detected in the surrounding histologically normal segments such as jejunum (Figure 1B), possibly accounting for the presence of some CCR5+ T cells in this area in addition to the severely inflamed ileum area. Furthermore, some CCR5+ T cells migrated to the apparently normal segments in addition to the inflamed segment. Overall, the kinetics of CCL5 expression in the intestine over a 30-week period closely resemble that of mononuclear cell infiltration in the inflamed intestine of SAMP1/YP mice. We used TAK-779, a specific antagonist for the CCR5-CCL5 chemokine axis, to disrupt the inflammatory chemokine network in the SAMP1/YP mice. This antagonist was effective in blocking the CCR5+ T-cell migration to the inflamed ileum (Figure 7). These data are consistent with the model that the CCL5-CCR5 chemokine axis is involved in T-cell migration to the intestinal lesion. Recently, it has been reported that anti–mucosal addressin cell adhesion molecule 1 (MAdCAM-1) or anti–vascular cell adhesion molecule 1 (VCAM-1) suppressed T-cell migration and ileitis in SAMP1/Yit mice, suggesting the involvement of MAdCAM-1 (a ligand of α4β7) or VCAM-1 (a ligand of α4β1) in inflammatory T-cell migration to the intestine of SAMP1/Yit mice.36 Another group reported that individual blocking of MAdCAM-1 or α4 integrins did not attenuate inflammation in a SAMP1/YitFc T-cell transfer model, but together with L-selectin it was effective.37 Despite the differences, these reports suggest the involvement of α4β7/MAdCAM-1 and α4β1/VCAM-1 in T-cell migration in SAMP1/YP mice. Because integrins are activated by chemoattractants, the CCR5-CCL5 chemokine axis could work together with these adhesion molecules.

CCR5 is preferentially expressed by inflammatory Th1 cells,38, 39, 40, 41, 42 and it is known that many Th1 cells infiltrate the intestinal lesion of SAMP1/Yit mice.20 In addition to Th1 cells,43 other inflammatory cells, such as subsets of CD8 T cells, eosinophils,44 monocytes/macrophages,45 and NK cells,46 also express CCR5. CCL5 is implicated also in T-cell polarization toward Th1 response.47 Based on the information, it might be expected that blocking of the CCL5-CCR5 axis in SAMP1/YP mice would inhibit the migration of inflammatory cells and intestinal inflammation. We tested this hypothesis utilizing 2 different mouse models (SAMP1/YP and SCID). The outcome, however, was opposite to the expectation, and CCR5 blockade with TAK-779 actually exacerbated the intestinal inflammation. Moreover, the TAK-779 treatment in SAMP1/YP mice exacerbated the inflammation in the ileum but did not induce inflammation in other segments of the intestine, suggesting that insufficient migration of FoxP3+ T cells would exacerbate the existing inflammation but does not cause de novo inflammation in normal intestinal segments. These results led us to examine CCR5 expression by FoxP3+ regulatory T cells that are known to suppress a number of inflammatory diseases,48 including IBD.49, 50 We found that the majority of FoxP3+ T cells, in the inflamed intestine but not other organs of SAMP1/YP mice, express CCR5. The enrichment of CCR5+ FoxP3+ T cells in the inflamed tissue of SAMP1/YP mice peaked at 7 weeks of age. We also found that FoxP3+ suppressor T cells migrate better than FoxP3− effector T cells to the inflamed intestine of SAMP1/YP mice. Furthermore, this migration was dependent on CCR5, because CCR5 blockade significantly decreased the migration. Similar to the CCR5+FoxP3+ T cells of SAMP1/YP mice, many more FOXP3+ (or CD4+CD25+) T cells than FOXP3− memory/effector T cells in humans express CCR5.51, 52 As mentioned previously, CCL5 is highly induced in human IBD34, 35 and therefore a similar operation of the CCL5-CCR5 axis in recruitment of FoxP3+ suppressor T cells in human IBD is expected. In addition, we found that the relative ratio of FoxP3+ to FoxP3− CD4+ T cells was significantly decreased in the inflamed intestinal ileum of SAMP1/YP mice. Unexpectedly, the FoxP3+ T-cell frequency was decreased in the normal intestinal segment at 7 weeks of age. This decrease, however, was transient because it rebounded to the normal level in older mice. In contrast, the FoxP3+ cell deficiency in the inflamed segment was not corrected in older mice. Thus, a sustained, but not transient, decrease is associated with the chronic spontaneous inflammation in SAMP1/YP mice. The exact cause and mechanism of the decrease warrant further investigation.

We conclude that (1) chronic active inflammation in the gut, as seen in the SAMP1/YP mice, is characterized by induction of an additional chemokine network (CCL5-CCR5) and (2) this chemokine network is involved in preferential recruitment of FoxP3+ regulatory T cells over FoxP3− inflammatory T cells into the chronically inflamed intestine. Potential therapies that compromise this chemokine network in IBD are likely to exacerbate, rather than ameliorate, the intestinal inflammation.

Back to Article Outline

 

The authors thank Dr D. Hollander (the Broad Medical Research Program) for his input in initiation of the study; J. H. Lee, C. W. Wang, and J. Kim (Kim Lab) for technical assistance; and the National Institutes of Health AIDS Research & Reference Reagent Program, Dr K. Hieshima (Kinki University), Dr K. Murphy (Washington University), and Dr G. Nolan (Stanford University) for valuable reagents.

Back to Article Outline

References 

  1. Rot A, von Andrian UH. Chemokines in innate and adaptive host defense: basic chemokinese grammar for immune cells. Annu Rev Immunol. 2004;22:891–928
  2. Kim CH. The greater chemotactic network for lymphocyte trafficking: chemokines and beyond. Curr Opin Hematol. 2005;12:298–304
  3. Kunkel EJ, Campbell DJ, Butcher EC. Chemokines in lymphocyte trafficking and intestinal immunity. Microcirculation. 2003;10:313–323
  4. Kunkel EJ, Campbell JJ, Haraldsen G, Pan J, Boisvert J, Roberts AI, et al. Lymphocyte CC chemokine receptor 9 and epithelial thymus-expressed chemokine (TECK) expression distinguish the small intestinal immune compartment: epithelial expression of tissue-specific chemokines as an organizing principle in regional immunity. J Exp Med. 2000;192:761–768
  5. Papadakis KA, Prehn J, Nelson V, Cheng L, Binder SW, Ponath PD, et al. The role of thymus-expressed chemokine and its receptor CCR9 on lymphocytes in the regional specialization of the mucosal immune system. J Immunol. 2000;165:5069–5076
  6. Bowman EP, Kuklin NA, Youngman KR, Lazarus NH, Kunkel EJ, Pan J, et al. The intestinal chemokine thymus-expressed chemokine (CCL25) attracts IgA antibody-secreting cells. J Exp Med. 2002;195:269–275
  7. Mora JR, Bono MR, Manjunath N, Weninger W, Cavanagh LL, Rosemblatt M, et al. Selective imprinting of gut-homing T cells by Peyer’s patch dendritic cells. Nature. 2003;424:88–93
  8. Pan J, Kunkel EJ, Gosslar U, Lazarus N, Langdon P, Broadwell K, et al. A novel chemokine ligand for CCR10 and CCR3 expressed by epithelial cells in mucosal tissues. J Immunol. 2000;165:2943–2949
  9. Wang W, Soto H, Oldham ER, Buchanan ME, Homey B, Catron D, et al. Identification of a novel chemokine (CCL28), which binds CCR10 (GPR2). J Biol Chem. 2000;275:22313–22323
  10. Kunkel EJ, Kim CH, Lazarus NH, Vierra MA, Soler D, Bowman EP, et al. CCR10 expression is a common feature of circulating and mucosal epithelial tissue IgA Ab-secreting cells. J Clin Invest. 2003;111:1001–1010
  11. Lazarus NH, Kunkel EJ, Johnston B, Wilson E, Youngman KR, Butcher EC. A common mucosal chemokine (mucosae-associated epithelial chemokine/CCL28) selectively attracts IgA plasmablasts. J Immunol. 2003;170:3799–3805
  12. Hieshima K, Ohtani H, Shibano M, Izawa D, Nakayama T, Kawasaki Y, et al. CCL28 has dual roles in mucosal immunity as a chemokine with broad-spectrum antimicrobial activity. J Immunol. 2003;170:1452–1461
  13. Rothenberg ME. Eotaxin (An essential mediator of eosinophil trafficking into mucosal tissues). Am J Respir Cell Mol Biol. 1999;21:291–295
  14. Scheerens H, Hessel E, de Waal-Malefyt R, Leach MW, Rennick D. Characterization of chemokines and chemokine receptors in two murine models of inflammatory bowel disease: IL-10−/− mice and Rag-2−/− mice reconstituted with CD4+CD45RBhigh T cells. Eur J Immunol. 2001;31:1465–1474
  15. Jugde F, Alizadeh M, Boissier C, Chantry D, Siproudhis L, Corbinais S, et al. Quantitation of chemokines (MDC, TARC) expression in mucosa from Crohn’s disease and ulcerative colitis. Eur Cytokine Netw. 2001;12:468–477
  16. Matsumoto S, Okabe Y, Setoyama H, Takayama K, Ohtsuka J, Funahashi H, et al. Inflammatory bowel disease-like enteritis and caecitis in a senescence accelerated mouse P1/Yit strain. Gut. 1998;43:71–78
  17. Strober W, Nakamura K, Kitani A. The SAMP1/Yit mouse: another step closer to modeling human inflammatory bowel disease. J Clin Invest. 2001;107:667–670
  18. Pizarro TT, Arseneau KO, Bamias G, Cominelli F. Mouse models for the study of Crohn’s disease. Trends Mol Med. 2003;9:218–222
  19. Mizoguchi A, Mizoguchi E, Bhan AK. Immune networks in animal models of inflammatory bowel disease. Inflamm Bowel Dis. 2003;9:246–259
  20. Kosiewicz MM, Nast CC, Krishnan A, Rivera-Nieves J, Moskaluk CA, Matsumoto S, et al. Th1-type responses mediate spontaneous ileitis in a novel murine model of Crohn’s disease. J Clin Invest. 2001;107:695–702
  21. Makita S, Kanai T, Matsumoto S, Iiyama R, Uraushihara K, Totsuka T, et al. The role of cryptopatch-derived intraepithelial lymphocytes in the development of chronic ileocecitis. Scand J Immunol. 2003;58:428–435
  22. Takedatsu H, Mitsuyama K, Matsumoto S, Handa K, Suzuki A, Funabashi H, et al. Interleukin-5 participates in the pathogenesis of ileitis in SAMP1/Yit mice. Eur J Immunol. 2004;34:1561–1569
  23. Rivera-Nieves J, Bamias G, Vidrich A, Marini M, Pizarro TT, McDuffie MJ, et al. Emergence of perianal fistulizing disease in the SAMP1/YitFc mouse, a spontaneous model of chronic ileitis. Gastroenterology. 2003;124:972–982
  24. Sugawara K, Olson TS, Moskaluk CA, Stevens BK, Hoang S, Kozaiwa K, et al. Linkage to peroxisome proliferator-activated receptor-gamma in SAMP1/YitFc mice and in human Crohn’s disease. Gastroenterology. 2005;128:351–360
  25. Kim CH, Kunkel EJ, Boisvert J, Johnston B, Campbell JJ, Genovese MC, et al. Bonzo/CXCR6 expression defines type 1-polarized T-cell subsets with extralymphoid tissue homing potential. J Clin Invest. 2001;107:595–601
  26. Kim CH, Qu CK, Hangoc G, Cooper S, Anzai N, Feng GS, et al. Abnormal chemokine-induced responses of immature and mature hematopoietic cells from motheaten mice implicate the protein tyrosine phosphatase SHP-1 in chemokine responses. J Exp Med. 1999;190:681–690
  27. Kim CH, Rott L, Kunkel EJ, Genovese MC, Andrew DP, Wu L, et al. Rules of chemokine receptor association with T cell polarization in vivo. J Clin Invest. 2001;108:1331–1339
  28. Ouyang W, Ranganath SH, Weindel K, Bhattacharya D, Murphy TL, Sha WC, et al. Inhibition of Th1 development mediated by GATA-3 through an IL-4-independent mechanism. Immunity. 1998;9:745–755
  29. Iwata M, Hirakiyama A, Eshima Y, Kagechika H, Kato C, Song SY. Retinoic acid imprints gut-homing specificity on T cells. Immunity. 2004;21:527–538
  30. Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the transcription factor Foxp3. Science. 2003;299:1057–1061
  31. Fontenot JD, Gavin MA, Rudensky AY. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol. 2003;4:330–336
  32. Baba M, Nishimura O, Kanzaki N, Okamoto M, Sawada H, Iizawa Y, et al. A small-molecule, nonpeptide CCR5 antagonist with highly potent and selective anti-HIV-1 activity. Proc Natl Acad Sci U S A. 1999;96:5698–5703
  33. Dragic T, Trkola A, Thompson DA, Cormier EG, Kajumo FA, Maxwell E, et al. A binding pocket for a small molecule inhibitor of HIV-1 entry within the transmembrane helices of CCR5. Proc Natl Acad Sci U S A. 2000;97:5639–5644
  34. Mazzucchelli L, Hauser C, Zgraggen K, Wagner HE, Hess MW, Laissue JA, et al. Differential in situ expression of the genes encoding the chemokines MCP-1 and RANTES in human inflammatory bowel disease. J Pathol. 1996;178:201–206
  35. Oki M, Ohtani H, Kinouchi Y, Sato E, Nakamura S, Matsumoto T, et al. Accumulation of CCR5+ T cells around RANTES+ granulomas in Crohn’s disease: a pivotal site of Th1-shifted immune response?. Lab Invest. 2005;85:137–145
  36. Matsuzaki K, Tsuzuki Y, Matsunaga H, Inoue T, Miyazaki J, Hokari R, et al. In vivo demonstration of T lymphocyte migration and amelioration of ileitis in intestinal mucosa of SAMP1/Yit mice by the inhibition of MAdCAM-1. Clin Exp Immunol. 2005;140:22–31
  37. Rivera-Nieves J, Olson T, Bamias G, Bruce A, Solga M, Knight RF, et al. L-selectin, alpha 4 beta 1, and alpha 4 beta 7 integrins participate in CD4+ T cell recruitment to chronically inflamed small intestine. J Immunol. 2005;174:2343–2352
  38. Bonecchi R, Bianchi G, Bordignon PP, D’Ambrosio D, Lang R, Borsatti A, et al. Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J Exp Med. 1998;187:129–134
  39. Loetscher P, Uguccioni M, Bordoli L, Baggiolini M, Moser B, Chizzolini C, et al. CCR5 is characteristic of Th1 lymphocytes. Nature. 1998;391:344–345
  40. Sallusto F, Lenig D, Mackay CR, Lanzavecchia A. Flexible programs of chemokine receptor expression on human polarized T helper 1 and 2 lymphocytes. J Exp Med. 1998;187:875–883
  41. Annunziato F, Cosmi L, Galli G, Beltrame C, Romagnani P, Manetti R, et al. Assessment of chemokine receptor expression by human Th1 and Th2 cells in vitro and in vivo. J Leukoc Biol. 1999;65:691–699
  42. Odum N, Bregenholt S, Eriksen KW, Skov S, Ryder LP, Bendtzen K, et al. The CC-chemokine receptor 5 (CCR5) is a marker of, but not essential for the development of human Th1 cells. Tissue Antigens. 1999;54:572–577
  43. Siveke JT, Hamann A. T helper 1 and T helper 2 cells respond differentially to chemokines. J Immunol. 1998;160:550–554
  44. Rot A, Krieger M, Brunner T, Bischoff SC, Schall TJ, Dahinden CA. RANTES and macrophage inflammatory protein 1 alpha induce the migration and activation of normal human eosinophil granulocytes. J Exp Med. 1992;176:1489–1495
  45. Campbell EM, Proudfoot AE, Yoshimura T, Allet B, Wells TN, White AM, et al. Recombinant guinea pig and human RANTES activate macrophages but not eosinophils in the guinea pig. J Immunol. 1997;159:1482–1489
  46. Maghazachi AA, al-Aoukaty A, Schall TJ. C-C chemokines induce the chemotaxis of NK and IL-2-activated NK cells (Role for G proteins). J Immunol. 1994;153:4969–4977
  47. Appay V, Rowland-Jones SL. RANTES: a versatile and controversial chemokine. Trends Immunol. 2001;22:83–87
  48. Kim CH. Migration and function of FoxP3+ regulatory T cells in the hematolymphoid system. Exp Hematol. 2006;34:1033–1040
  49. Mottet C, Uhlig HH, Powrie F. Cutting edge: cure of colitis by CD4+CD25+ regulatory T cells. J Immunol. 2003;170:3939–3943
  50. Martin B, Banz A, Bienvenu B, Cordier C, Dautigny N, Becourt C, et al. Suppression of CD4+ T lymphocyte effector functions by CD4+CD25+ cells in vivo. J Immunol. 2004;172:3391–3398
  51. Lim HW, Hillsamer P, Kim CH. Regulatory T cells can migrate to follicles upon T cell activation and suppress GC-Th cells and GC-Th cell-driven B cell responses. J Clin Invest. 2004;114:1640–1649
  52. Lim HW, Broxmeyer HE, Kim CH. Regulation of trafficking receptor expression in human forkhead box P3+ regulatory T cells. J Immunol. 2006;177:840–851

 Supported in part by grants from the National Institutes of Health/National Institute of Allergy and Infectious Diseases (AI063064), the Eli and Edythe L. Broad Medical Foundation, the Sidney Kimmel Foundation, and the American Heart Association (to C.H.K.) and by research fellowships from Purdue Research Foundation (S.G.K.), ARP (H.W.L.), and Indiana University School of Medicine, Lafayette (R.J.P.).

PII: S0016-5085(07)00010-8

doi:10.1053/j.gastro.2007.01.008

Gastroenterology
Volume 132, Issue 3 , Pages 966-981, March 2007