Chemokines play an important role in the migration of leukocytes at sites of inflammation, and some constitutively expressed chemokines may direct lymphocyte trafficking within lymphoid organs and peripheral tissues. Thymus-expressed chemokine (TECK or Ckβ-15/CCL25), which signals through the chemokine receptor CCR9, is constitutively expressed in the thymus and small intestine but not colon, and chemoattracts a small fraction of PBLs that coexpress the integrin α4β7. Here we show that TECK is expressed in the human small bowel but not colon by endothelial cells and a subset of cells in intestinal crypts and lamina propria. CCR9 is expressed in the majority of freshly isolated small bowel lamina propria mononuclear cells (LPMC) and at significantly higher levels compared with colonic LPMC or PBL. TECK was selectively chemotactic for small bowel but not colonic LPMC in vitro. The TECK-induced chemotaxis was sensitive to pertussis toxin and partially inhibited by Abs to CCR9. TECK attracts predominantly the T cell fraction of small bowel LPMC, whereas sorted CD3+CCR9+ and CD3+CCR9− lymphocytes produce similar Th1 or Th2 cytokines at the single cell level. Collectively, our data suggest that the selective expression of TECK in the small bowel underlie the homing of CCR9+ intestinal memory T cells to the small bowel rather than to the colon. This regional specialization implies a segregation of small intestinal from colonic immune responses.
Chemokines (chemotactic cytokines) are small, 6- to 14-kDa heparin-binding proteins, which play a role in a variety of biological processes, most notably leukocyte chemotaxis (1, 2, 3). They are classified as C, CC, CXC, and CX3C based on the positioning of cysteine residues that form two disulfide bonds (3). Chemokines mediate their effects through G protein-coupled seven transmembrane domain receptors, which are currently divided into four families based on the type of chemokine that they bind; they are CXCR1 to CXCR5, and CCR1 to CCR9, XCR1, or CX3CR1 (3). Most chemokines recognize several receptors, and a single receptor can bind more than one chemokine (4). However, some receptors bind a single chemokine such as CCR6, which binds macrophage inflammatory protein (MIP)3-3α, CXCR5, which binds B cell-attracting chemokine 1, and CCR9, which binds thymus-expressed chemokine (TECK) only, among others (5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15). Chemokines were identified by their ability to direct extravasation of inflammatory cells during infection (1, 2). However, recent data identified several chemokines that are expressed constitutively in lymphoid and extra-lymphoid tissues, indicating that these chemokines might have homeostatic function by regulating lymphocyte trafficking to or within lymphoid organs and in peripheral tissues (3, 16, 17, 18, 19, 20, 21). Certain chemokines, such as stromal cell-derived factor 1 (SDF-1), 6-C-kine, and MIP-3β can also stimulate leukocyte adhesion and arrest on endothelium by triggering integrin activation (22, 23). It is thought that the combined expression of adhesion molecules and chemokine receptors on the cell surface provide an “address code” for leukocyte migration to different sites (24).
In contrast to naive T cells, memory/effector cells migrate mostly through peripheral tissues, and this process is controlled by the expression of different sets of integrins and chemokine receptors (25, 26, 27, 28). Thus, naive T cells express CXCR4, the receptor for SDF-1, and CCR7, the receptor for EBV-induced molecule 1 ligand chemokine and secondary lymphoid tissue chemokine (also called 6-C-kine). Mice deficient in CCR7 or secondary lymphoid tissue chemokine have defective homing of naive T cells to secondary lymphoid organs (16, 29). Gene knockout studies have established that CXCR5 is required for B cell migration to B cell follicles of spleen and Peyer’s patches (PP; Ref. 30). In addition, the defects in lymph node (LN) development observed in lymphotoxin-α and TNF knockout mice have been attributed, at least in part, to decreased production of chemokines by stromal LN cells (31). Collectively, these data suggest an important role of certain chemokines in regulating the homing of specific T cell subsets and other immune cells into microanatomic compartments of secondary lymphoid organs. Certain chemokine receptors are also preferentially expressed on naive T cells under Th1 or Th2 polarizing conditions in vitro. Th1 cells predominantly express CXCR3 and CCR5 (32, 33, 34, 35, 36). In contrast, Th2 cells express CCR3, CCR4, and CCR8 (3, 24, 32, 33, 35, 36). Recently, the orphan chemokine receptor GPR-9-6, now designated CCR9, was found to be expressed on a small percentage (2–4%) of circulating memory T cells, all of which express the mucosal homing ligand α4β7 (9). The ligand for CCR9, TECK, also CCL25/Ckβ-15 according to the recent chemokine/chemokine receptor nomenclature (3), is selectively expressed in the thymus and small intestine (9, 13, 37, 38). In addition, CCR9 mRNA is expressed in the thymus, small intestine, and at lower levels in the spleen, suggesting that the CC-chemokine TECK and its receptor, CCR9, may play an important role in T cell maturation and in the intestinal immune response (9).
The potential importance of CCR9 and TECK in mucosal immunity prompted us to study the expression of this chemokine/chemokine receptor pair in the normal small and large bowel. Our data suggest that the TECK/CCR9 chemokine/chemokine receptor pair is important for the regional specialization of intestinal immunity and the combined expression of CCR9/α4β7 on the cell surface may provide a small intestinal “address code” for circulating intestinal memory T cells.
Materials and Methods
Purification of lamina propria mononuclear cells (LPMC), PBMC, mesenteric lymph node (MLN) lymphocytes, and intestinal epithelial cells (IECs)
Intestinal specimens were obtained from patients undergoing surgical resection of the colon (with colon carcinoma) or small intestine usually during the second stage of a prior ileal anal-pouch anastomosis at Cedars-Sinai Medical Center (Los Angeles, CA). Approval for the use of human subjects was obtained from the Institutional Review Board at Cedars-Sinai Medical Center. In this study, all tissue specimens were taken from an uninvolved area of resected colon or small bowel. LPMC were isolated using a technique modified from that described previously (39). PBMC were isolated from normal healthy volunteers by separation on Ficoll-Hypaque gradients. The cells were subsequently washed three times with HBSS and resuspended in culture media (RPMI 1640 with 10% FCS) at a concentration of 2 × 106/ml. MLN lymphocytes were isolated following mechanical disruption of LNs. IECs from small bowel or colon were isolated as previously described (40).
Abs and reagents
Anti-CD3, -CD4, -CD8, -CD19, -CD56, -HLA-DR, -CD45RO dye-linked mAbs for immunofluorescence studies were obtained from Caltag (South San Francisco, CA). Anti-CD25, -CD69, and -CD95 were obtained from PharMingen (San Diego, CA). TECK and SDF-1α were purchased from PeproTech (Rocky Hill, NJ).
Generation of anti-TECK and anti-CCR9 mAbs
The mAb to human TECK (hTECK; clone LS202 5A9, IgG1) was generated by i.p immunization of BALB/c mice at 3-wk intervals with 10 μg of TECK (Peprotech) in CFA, IFA, and finally PBS. Fusions were performed after at least four immunizations 3 days after the last boost by fusion with SP2/0 myeloma cells (American Type Culture Collection, Manassas, VA). Fusions were screened by ELISA with plates coated with hTECK, and the positive hybridomas were subcloned. Of 20 anti-TECK mAbs tested only two, designated 4G1 and 5A9, were found to stain frozen sections of thymus. The mAb to human CCR9 has been previously described (9).
Freshly isolated or cultured LPMC or PBMC (2.5–5.0 × 105) were washed twice with 1 ml of PBS supplemented with 0.1% BSA and 0.1% azide. The cells were resuspended in 100 μl of 10% human Ab serum to block nonspecific Fc binding for 15 min. For the staining of surface Ags, cells were incubated with the mAb 3C3 for 30 min on ice, washed with PBS/BSA/azide, and incubated with a secondary goat anti-mouse anti-Fab (H+L) tricolor or goat anti-mouse IgG2b-PE for 30 min on ice. The cells were washed again with PBS/BSA/azide and incubated with mouse IgG for 15 min. FITC- and PE-conjugated mAb for surface Ag were used for 30 min. After washing twice, cells were resuspended in 400 μl of 1% paraformaldehyde in PBS and analyzed by FACS (Becton Dickinson, Mountain View, CA). Events (104) were routinely collected and analyzed using Lysis II software (Becton Dickinson Immunocytometry Systems, San Jose, CA). Both the percentage of positive cells and the mean fluorescence intensity (MFI) of the cells were determined.
Total RNA was extracted from small bowel or colonic intestinal mucosa and freshly isolated small bowel or colonic IECs and LPMC using the RNeasy Kit as recommended by the manufacturer (Qiagen, Valencia, CA). One microgram of RNA was reverse-transcribed into cDNA with oligo(dT) in a 20-μl volume using a Thermoscript RT-PCR System (Life Technologies, Grand Island, NY) according to a standard protocol. Primers were designed as described elsewhere (9). Primers for TECK were: sense 5′-TCGAAGAAGCTTATGAACCTGTGGCTCCTG-3′ antisense 5′-AAGAAGTCTAGATCACAGTCCTGAATTAGC-3′ (product 453 bp). Two microliters of the reaction cDNA was mixed with 10 mM dNTP, 10 μM primers, 50 mM MgCl2, and 5 U/μl of Platinum Taq DNA polymerase in a 50-μl volume as recommended by the manufacturer (Life Technologies). The cycle parameters were an initial melt at 95°C for 2 min, then 35 cycles: 95°C, 30 s; 55°C, 30 s; and 72°C, 1 min, followed by a final extension of 72°C, 7 min. Amplification with G3PDH primers (sense 5′-TGAAGGTCGGAGTCAACGGATTTGGT-3′, antisense 5′-CATGTGGGCCATGAGGTCCACCAC-3′) (product 983 bp) (Clontech, Palo Alto, CA) was examined in identical conditions as an internal control to demonstrate equivalence of template. The PCR products were visualized with ethidium bromide after 1.5% agarose gel electrophoresis.
Cytokine detection at the single cell level
Freshly isolated LPMC or sorted mucosal T cells were stimulated with 10−7 M PMA and 1 μg/ml ionomycin for 4 h. Brefeldin A (10 μg/ml) was added to the culture after 2 h of stimulation to block cytokine secretion. Unseparated cells (106 cells) were surface stained and washed twice as mentioned above and subsequently fixed with 4% paraformaldehyde and permeabilized with saponin. Fixed and permeabilized cells were stained with FITC anti-IFN-γ and PE anti-IL-4 mAbs (PharMingen) and analyzed by FACS (Becton Dickinson).
Cell migration was evaluated using a 48-well chemotaxis chamber (Neuroprobe, Cabin John, MD). TECK, diluted in HEPES-buffered RPMI 1640 supplemented with 1% BSA, was added to the lower wells, and 105 cells in the same buffer to the upper wells. Polyvinylpyrrolidone-free polycarbonate membranes (Neuroprobe) with 3-μm pores were used. After incubation for 120 min at 37°C, the cells that had migrated through the pores to the lower wells were counted by FACS. A known number of 3.2-μm fluorescent microsphere beads (PharMingen) was added to each sample before analysis to determine the absolute number of migrating cells. The assay was performed in triplicate.
Sections were fixed and stained as previously described (41). Briefly, fixed sections (6 μm) from normal small bowel and colonic mucosa, or thymus, were deparaffinized and treated with citrate buffer (pH 6). The sections were then incubated with anti-hTECK Ab (clone LS202 5A9, IgG1) followed by a goat anti-mouse Ab (Dako, Carpinteria, CA) at a 1:20 dilution. The sections were subsequently incubated with a mouse peroxidase anti-peroxidase (Dako) at 1:100 dilution. Following wash, the sections were developed with DAB substrate-chromogen (Dako). Sections were counterstained with hematoxylin (Fisher Scientific, Pittsburgh, PA).
Where indicated Student’s t test was used to calculate statistical significance for difference in a particular measurement between different groups. Values of p < 0.05 were considered statistically significant.
CCR9 is highly expressed on small bowel compared with colonic or peripheral blood T lymphocytes
We measured the percentage CCR9+ LPMC in the small bowel and colon by flow cytometry using a CCR9-specific mAb, designated 3C3. In cross-reactivity studies, 3C3 did not cross-react with CCR1-7 or CXCR1-4 transfectants (9). Three color immunofluorescence analysis of LPMC from small bowel or colon and PBMC showed that CCR9 is expressed predominantly in small bowel compared with colonic or peripheral blood CD3+ T lymphocytes. CCR9 was expressed at high levels in both the CD8 and CD4 subsets of small bowel CD3+ cells. As shown in Fig. 1⇓, CCR9 was expressed in 67% (range, 57–76%) of CD3+CD4+ small bowel compared with 20% (range, 15–25%) of colonic (p < 0.0001) and 4% (range, 3–5%) of peripheral blood CD3+CD4+ lymphocytes. The differences between groups were statistically significant (p < 0.0001). Similar differences in CCR9 expression were observed between small bowel, colonic, or peripheral blood CD3+CD8+ lymphocytes. CCR9 was expressed in 58% (range, 47–70%) compared with 10% (range, 6–14%) and 2% (range, 1.4–2.6%) of small bowel, colonic, or peripheral blood CD3+CD8+ cells, respectively (Fig. 1⇓). In addition, the density of CCR9 expression, as assessed by the MFI in each lymphocyte subset showed that CCR9 is expressed at higher levels in small bowel compared with colonic or peripheral blood T cells in both the CD4 and CD8 compartment (Table I⇓). We further analyzed the expression of CCR9 by flow cytometry on freshly isolated MLN lymphocytes draining small bowel and colon. CD3+CCR9+ lymphocytes were significantly enriched in small bowel vs colonic MLN (mean, 69 vs 11%, n = 3, p < 0.05). A representative FACS analysis of CCR9 expression in MLN lymphocytes is shown in Fig. 2⇓. Collectively, our data show that the chemokine receptor CCR9 is expressed in a much larger percentage and at a higher density in small bowel compared with colonic lamina propria and draining MLN T lymphocytes.
We next analyzed the phenotype of CCR9-expressing small bowel LPMC. As shown in Fig. 3⇓, CCR9 was expressed predominantly on CD3+ small bowel LPMC, although a significant percentage of CD3− cells also express CCR9. Also CCR9 was expressed on a subset of B and NK cells. Almost all CCR9+ LPMC, whether from small bowel or colon, coexpressed CD45RO, CD69, CD95, and a significant percentage were also HLA-DR+ (Fig. 3⇓, and data not shown). This is consistent with the known phenotype of mucosal lymphocytes as highly activated memory cells (39).
TECK is selectively expressed in the small bowel but not colon
It has been shown that hTECK message was detected by RT-PCR and Northern blot in the thymus and small bowel, but not colon (9, 37). In mice, murine TECK (mTECK) was initially reported to be expressed by thymic dendritic cells (37) but two recent reports showed it to be expressed by thymic and small IECs (13, 38). Therefore, we studied the expression of TECK in human intestinal tissues. First we analyzed TECK expression by RT-PCR in whole small bowel and colonic mucosa, as well as in IECs and LPMC isolated from small bowel and colon. Fig. 4⇓ shows that TECK was detected by RT-PCR in whole small bowel mucosa, as well as in small bowel IECs and LPMC. TECK message was absent from whole colonic mucosa and colonic IECs or LPMC. To investigate more precisely the distribution of TECK expression in small bowel, we performed immunohistochemistry using a mAb, designated 5A9, against hTECK. Thymus tissue was used as positive control for hTECK staining. As shown in Fig. 5⇓, TECK staining was detected in the small bowel and thymus but not colon (compare a–c, and f with d). Strong TECK immunoreactivity was detected in a subset of cells with elongated processes in the small bowel intestinal crypts and the lamina propria predominantly in areas of lymphocyte aggregates (Fig. 5⇓b). TECK staining was not detected in mucosal lymphocytes or surface epithelial cells. TECK-immunoreactive cells were also detected in small bowel but not colonic endothelial cells (compare Fig. 5⇓, a inset and c with d). In thymic sections, TECK immunoreactivity was observed in the stromal component of the thymic medulla and cortex, as has been shown previously for TECK expression in murine thymus (Fig. 5⇓f).
TECK mediates chemotaxis of small bowel but not colonic or PBLs
TECK has been shown to induce Ca2+ mobilization and chemotaxis of CCR9 transfectants (9, 10, 14), suggesting that CCR9 is a specific chemotactic receptor for TECK. In addition, the small percentage of PBMC that respond by chemotaxis to TECK express high levels of the mucosal homing ligand α4β7 (9). In addition, the selective expression of the chemokine TECK in the small bowel and its receptor CCR9 in small bowel lymphocytes suggest that TECK may be involved in the selective trafficking of CCR9-expressing lymphocytes to the small bowel instead of colon. Therefore, we examined whether small bowel and colonic LPMC respond differentially to TECK in a chemotaxis assay. In bulk migration experiments, TECK chemoattracts freshly isolated small bowel but not colonic LPMC or PBMC (Fig. 6⇓a). The migrating small bowel LPMC when plotted vs increasing concentration of TECK revealed a bell-shaped curve typically observed with other chemokines. The optimal chemotactic concentration of TECK was 500 nM for small bowel lymphocytes, which is within the range (1–1000 nM) of chemotactic activity seen with other chemokines (9) (Fig. 6⇓a). Checkerboard analysis established that TECK induced chemotaxis and not chemokinesis of small bowel LPMC (data not shown). The inability of colonic LPMC to migrate in response to TECK was not due to a general migratory defect of those cells compared with small bowel LPMC because both migrated to SDF-1α, which signals through CXCR4 (23) (data not shown). Migration in response to TECK could be partially blocked by preincubation of lymphocytes with anti-CCR9, but not by anti-CCR3 mAb (Fig. 6⇓b). Preincubation of the cells with pertussis toxin completely inhibited the migration in response to TECK, consistent with Gαi protein-coupled signaling through CCR9 (9), as has been shown for other chemokine receptors (1, 2, 3).
Small bowel LPMC migrating to TECK is primarily T lymphocytes
To further characterize the mononuclear cell subsets that respond to TECK, we analyzed the phenotype of small bowel LPMC that migrated to optimal concentrations (500 nM) of TECK. We used flow cytometry to identify T cells (CD3+), B cells (CD19+), or non-T/non-B cells (CD3− CD19−) cells and CD4+ or CD8+ subsets of T cells. LPMC that migrated to the lower wells of the chemotaxis chamber were pooled and stained with mAb for CD3, CD4, CD8, CD45, and CD19. The percentage of cells of each phenotype that migrated to TECK was analyzed by FACS. As shown in Fig. 7⇓, TECK was primarily chemotactic for T cells, both the CD4+ and CD8+ subsets. The percentage of T cells was 85% (60% CD4+, 20% CD8+) in the migrating population compared with 56% (39% CD4+, 13% CD8+) in the starting population. TECK was not chemotactic for B lymphocytes and non-T/non-B mononuclear cells (Fig. 7⇓).
CCR9 expression in small bowel lymphocytes does not correlate with the production of Th1 or Th2 cytokines
Distinct profiles of chemokine receptors are acquired during in vitro differentiation of naive T lymphocytes into Th1 or Th2 subsets. For example, Th1 cells preferentially express CXCR3 and CCR5 (32, 33, 34, 35, 36), and Th2 cells express CCR3, CCR4, and CCR8 (24, 32, 33, 35, 36). Therefore, we examined whether CCR9 expression in mucosal T cells defines a Th1 or Th2 cytokine-producing phenotype. We examined cytokine production at the single cell level by intracellular cytokine staining. In initial experiments, we found no differences in the percentage of IFN-γ-producing (Th1) or IL-4-producing (Th2) cells between small bowel and colonic CD3+ lymphocytes despite a significant difference in the expression of CCR9 between these lymphocyte subsets (data not shown). To further confirm the cytokine profile of CCR9+ mucosal T cells, CD3+ cells from the small bowel were sorted into CD3+CCR9+ and CD3+CCR9− subsets using flow cytometry and analyzed for cytokine production. The percentage of Th1 or Th2 cells among CD3+CCR9+ and CD3+CCR9− small bowel lymphocytes was similar, although IFN-γ-producing cells were slightly higher in the CD3+CCR9+ subset and the percentage of cells producing neither IFN-γ nor IL-4 were higher in the CD3+CCR9− subset (Fig. 8⇓). These data show that small bowel mucosal T cells expressing CCR9 exhibit a diverse cytokine profile and that CCR9 expression is rather linked to a phenotype with selective homing potential to the small bowel mucosa.
Ag-reactive memory and effector cells induced in response to intestinal immunization traffic preferentially into the intestinal wall and/or into the intestine-associated lymphoid organs (28, 42). Mucosal addressin cell-adhesion molecule (MAdCAM-1) is expressed on high-endothelial venules of PP, MLN, and postcapillary venules in the intestinal lamina propria and interact with the leukocyte integrin α4β7 (43). MAdCAM-1/α4β7 interaction seems to play an important role in the preferential homing of α4β7high memory lymphocytes, which carry memory for intestinal Ags (44) to the gastrointestinal tract and mucosal immune system (28, 42). Indeed, β7-deficient mice have hypoplastic gut-associated lymphoid tissue (GALT) and a significant reduction in the number of CD4+, CD8+, and plasma cells in the lamina propria (45). Therefore, the α4β7 expression in peripheral memory T cells provides a mechanism for the segregation of intestinal from systemic immune responses (42).
The chemokine receptor CCR9 has been shown to be expressed on thymocytes and on a subset of memory α4β7+ intestinal trafficking CD4 and CD8 PBL. In addition, all small intestinal LPL and intraepithelial lymphocytes (IELs) express CCR9 (9). In this study, we directly compared the expression of CCR9 between small bowel and colonic LPL and MLN lymphocytes. Although transcript of CCR9 is absent in colon, we found that a small percentage of colonic lymphocytes express CCR9 by flow cytometry. We show the selective expression of CCR9 on the majority of small bowel lymphocytes as was also shown previously (9) and, in addition, CCR9 expression on a much lower percentage of colonic lymphocytes. Moreover, we demonstrate that CCR9+ lymphocytes are significantly enriched in MLN draining small bowel vs colon. In addition, phenotypic analysis of CCR9-expressing small bowel LPL reveals that CCR9 is expressed predominantly on CD3+ lymphocytes, although a subset of B and NK cells are also CCR9+. It was shown previously that CCR9 is expressed on a significant percentage of B cells but not NK cells on PBL (9). Based on three-color flow cytometry, we show in this study that CCR9 is expressed on CD4 and CD8 small bowel T lymphocytes and virtually all CD3+CCR9+ lymphocytes coexpress CD45RO, CD69, and CD95. A subset of CD3+CCR9+ small bowel lymphocytes also coexpress CD25 and HLA-DR.
The ligand for CCR9, TECK, has been shown to be expressed in the thymus and small intestine (9, 13, 37). The mTECK was reported to be expressed in the thymus by dendritic cells as well as endothelial cells (37). Another report showed mTECK expression in the thymus and small intestine by epithelial cells (13). In this study, we define for the first time the cells expressing TECK in human intestinal tissues. Human TECK is selectively expressed in the small bowel but not colonic mucosa by a subset of cells in intestinal crypts and lamina propria as well as endothelial cells. In a recent report mTECK mRNA expression was reported to be restricted to villus epithelial cells, the expression beginning at or just below the crypt-villus junction, increasing to maximum level approximately one-third of the way up the villus, and subsequently decreasing toward the villus tip (13). Interestingly, mTECK mRNA was detected in the follicle-associated epithelium of the murine PP by in situ hybridization. Although we have not directly examined hTECK protein expression in human PP, immunohistochemistry of small bowel mucosa shows distinct hTECK protein expression compared with mTECK mRNA expression profile. In human tissue, scattered crypt but not villus epithelial cells stain for TECK protein. The reason for the discrepancy in TECK mRNA and protein expression in the small bowel mucosa between mouse and human is unknown. Importantly, we show that hTECK is expressed on small bowel but not colonic endothelial cells. In addition to the expression profile of hTECK protein in intestinal tissues, we demonstrate that TECK chemoattracts small bowel but not colonic lymphocytes in vitro. Therefore, the selective expression of TECK in small bowel compared with colon may account for the selective recruitment and retention of CCR9+α4β7+ T cells in small bowel lamina propria. The small number of colonic lymphocytes expressing CCR9, which are predominantly CD3+CD4+, may be recruited to the colonic lamina propria in response to other chemokine(s) and may only coincidentally express CCR9. It has been hypothesized that CCR9 expression is induced on naive T lymphocytes in the mucosal environment (9). We further propose that CCR9 induction must be unique to the small bowel mucosal environment. This idea is supported by the finding that draining MLN from small bowel contain a higher percentage of CCR9+ lymphocytes than do MLN draining colon. However, in naive T lymphocytes CCR9 expression could not be induced during culture with several cytokines, including TGF-β or IL-10, which are highly expressed in the mucosal environment (46). Therefore, the mechanism by which CCR9 is induced or up-regulated by local microenvironmental factors in the small bowel inductive sites of the mucosal immune system is currently unknown. Nevertheless, activated CCR9+ cells in draining LNs may subsequently recirculate and home selectively to intestinal sites that express TECK, such as the small bowel mucosa.
The precise mechanism by which TECK is involved in the homing of CCR9+ lymphocytes to the small bowel mucosa is currently unknown. A likely explanation is that TECK presented on the small bowel endothelial cell surface may trigger firm adhesion of circulating CCR9+ lymphocytes to the vessel wall via α4β7/MAdCAM-1 interactions, and subsequent transmigration into the lamina propria as has been shown for other chemokines (20, 21, 22, 23). TECK expressed by resident small bowel mucosal stromal cells may help retain these CCR9+ T lymphocytes into the small bowel mucosa. It is also tempting to speculate that TECK, which is constitutively expressed in the small bowel intestinal crypts, may be involved in the homing and retention of small bowel IELs. Indeed, most of small bowel IELs express CCR9 (9), whereas the majority of colonic IELs are CCR9− (K. A. Papadakis, unpublished data). Our data show for the first time a phenotypic difference between small bowel and colonic LPMC in humans, based on the selective expression of the chemokine receptor CCR9, which strongly suggests a mechanism for regulated trafficking of intestinal memory T cells to the small bowel vs colon.
We finally demonstrate that CCR9+ small bowel T lymphocytes have a diverse cytokine profile, as similar percentages of sorted CD3+CCR9+ and CD3+CCR9− small bowel lymphocytes produce IFN-γ or IL-4. Distinct patterns of chemokine receptors are acquired by T cells under Th1 (CXCR3, CCR5)- or Th2 (CCR3, CCR8)-polarizing conditions in vitro, which translates into new migratory behavior toward their respective chemokines (24, 32, 33, 34, 35, 36). Therefore, the expression of certain chemokines will influence the type of T cell immune response in a tissue, for instance, Th1 or Th2. However, other chemokines may have a role in tissue-selective recruitment of lymphocytes from the blood (47). Thymus and activation-regulated chemokine and macrophage-derived chemokine, for example, the ligands for CCR4, strongly attract skin-homing but not intestinal memory T cells. The receptor CCR4 is highly expressed on cutaneous lymphocyte Ag memory CD4+ cells; therefore, CCR4 expression may define a phenotype of predominantly skin homing subset of peripheral memory T cells (47). Consistent with these observations, CCR4 expression was virtually absent in mucosal T cells (K. A. Papadakis, unpublished results). Another example, cutaneous T cell-attracting chemokine (CTACK or CCL27), a recently described chemokine, is expressed in skin and selectively chemoattracts cutaneous lymphocyte Ag memory T cells (19).
Our data suggest that the TECK/CCR9 ligand/receptor pair is important for the selective homing and retention of CCR9+ T lymphocytes to the small intestine instead of colon, and provides a mechanism for regional specialization of the mucosal immune system and the segregation of small bowel from colonic immune responses. The CCR9 phenotype is not linked to a specific cytokine profile because both small bowel and colonic lymphocytes as well as sorted CCR9+ and CCR9− small bowel T lymphocytes produce similar Th1 or Th2 cytokines at the single cell level. Consistent with these observations is the finding that CCR9 expression cannot be induced in naive T cells under Th1-, Th2, or T-regulatory 1 polarizing conditions (9).
In summary, the preferential expression of the chemokine receptor CCR9 in small bowel compared with colonic lymphocytes, the expression of its ligand, TECK, by small bowel but not colonic endothelial and stromal cells, and the in vitro migration of small bowel but not colonic lymphocytes to TECK, suggest that the TECK/CCR9 ligand/receptor pair is important for the regional specialization of the mucosal immune system. The selective expression of a chemokine, such as TECK, in the small intestine could provide the fine-tuning of mucosal T cell trafficking in combination with the expression of the intestinal mucosal homing ligand α4β7. Therefore, the combination of CCR9/α4β7 expression may provide a novel mechanism to segregate small intestinal from colonic immune responses. The recent identification of the mouse homologs of TECK and CCR9 will permit further examination of their role in T cell development and intestinal immunity by selectively inhibiting their expression. Further understanding of the selective trafficking of T cell to the small bowel based on CCR9 expression will also help us study immune mechanisms such as oral tolerance and immune-mediated diseases of the small intestine.
We thank Dr. Phillip Fleshner and Joanne Gainnie for providing specimens, Krystine Nguyen for isolating LPMC, Patricia Lin for flow cytometry, Nassim Kassam for help and advice in generating the anti-TECK mAb 5A9, and Richard Deem, Carol Landers, and Offer Cohavy for help with the figures and statistical analysis. We also thank Loren Karp for critical reading of the manuscript.
↵1 This work was supported by National Institutes of Health Grants DK-46763 and DK-56328.
↵2 Address correspondence and reprint requests to Dr. Konstantinos A. Papadakis, Inflammatory Bowel Disease Center, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, D-4062, Los Angeles, CA 90048. E-mail address: Papadakisk@cshs.org or Dr. Stephan R. Targan, Inflammatory Bowel Disease Center, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, D-4063, Los Angeles, CA 90048. E-mail address:
↵3 Abbreviations used in this paper: MIP, macrophage inflammatory protein; LN, lymph node; LPMC, lamina propria mononuclear cells; PP, Peyer’s patches; TECK, thymus-expressed chemokine; hTECK, human TECK; IECs, intestinal epithelial cells; MAdCAM-1, mucosal addressin cell-adhesion molecule; mTECK, murine TECK; IELs, intraepithelial lymphocytes; SDF-1, stromal cell-derived factor 1; MLN, mesenteric lymph node; MFI, mean fluorescence intensity.
- Received May 22, 2000.
- Accepted July 31, 2000.
- Copyright © 2000 by The American Association of Immunologists