HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Strauch, U. G. Articles by Parker, C. M. Search for Related Content PubMed PubMed Citation Articles by Strauch, U. G. Articles by Parker, C. M.

Integrin _E(CD103)₇ Mediates Adhesion to Intestinal Microvascular Endothelial Cell Lines Via an E-Cadherin-Independent Interaction¹

Ulrike G. Strauch^*, Ruth C. Mueller^*, Xiao Y. Li^*, Manuela Cernadas^*,, Jonathan M. G. Higgins^*, David G. Binion and Christina M. Parker^2,*

^* The Lymphocyte Biology Section, Division of Rheumatology, Immunology, and Allergy, and Pulmonary and Critical Care Division, Brigham and Women’s Hospital, Boston, MA 02115; and Medical College of Wisconsin, Milwaukee, WI 53226

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Integrins are important for T cell interactions with endothelial cells. Because the integrin _E7 is expressed on some circulating gut-homing T cells and as T cell numbers are reduced in the intestinal lamina propria of _E-deficient mice, we evaluated whether _E7 mediates binding to intestinal endothelial cells. We found that anti-_E7 mAbs partially blocked the binding of cultured intraepithelial T cells to human intestinal microvascular endothelial cells (HIMEC). Furthermore, _E7-transfected K562 cells bound more efficiently than vector-transfected K562 cells to HIMEC. Finally, HIMEC bound directly to an _E7-Fc fusion protein. These interactions were partially blocked by anti-_E7 mAbs, and endothelial cell binding to the _E7-Fc was dependent upon the metal ion-dependent adhesion site within the _E A domain. Of note, the HIMEC lacked expression of E-cadherin, the only known _E7 counterreceptor as assessed by functional studies, flow cytometry, and RT-PCR. Thus, HIMEC/_E7 binding was independent of E-cadherin. In addition, this interaction appeared to be tissue selective, as HIMEC bound to the _E7-Fc, whereas microvascular endothelial cells from the skin did not. Finally, there was evidence for an _E7 ligand on intestinal endothelial cells in vivo, as _E7 expression enhanced lymphocyte binding around vessels in the lamina propria in tissue sections. Thus, we have defined a novel interaction for _E7 at a nonepithelial location. These studies suggest a role for _E7 in interactions with the intestinal endothelium that may have implications for intestinal T cell homing or functional responses.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Lymphocyte trafficking into lymph nodes has been extensively studied (1). In these sites, naive and memory lymphocytes flow through high endothelial venules where they interact with the endothelium through a weak, rolling interaction. This initial binding is, in many cases, mediated by the interaction of selectins with glycosylated ligands. For some adhesive interactions, the rolling interaction is enhanced by the presence of chemokines (2). As the lymphocyte rolls along the endothelium, chemokine receptors expressed on the T cell come into further contact with chemokines displayed within the endothelial cell glycocalyx (3). If the chemokine receptor encounters an appropriate chemokine, it triggers integrins to bind their Ig superfamily member ligands with greater avidity (4). As a result of this interaction, the lymphocyte binds more firmly, stops rolling, and then extravasates through the endothelium and into the tissue. Lymphocytes are maintained within the tissue by additional chemokine signals and adhesive interactions, which might include the binding of integrins to extracellular matrix or cellular counterreceptors expressed on cells within the tissue.

The mechanisms of T lymphocyte extravasation and localization within the intestinal lamina propria appear to involve processes similar to those described above, although the specific molecules that are critical for each step differ (1). In particular, the ₇ integrin subfamily, including ₄₇ and _E7, are important in T cell homing to and localization in the intestine (1, 5). Integrin ₄₇ is expressed at high levels on a small subset of circulating memory T cells (6, 7) that selectively localize to the Peyer’s patches and to the intestinal lamina propria (1). In addition, it is found on 70% of lamina propria T lymphocytes but on only a minority of intestinal intraepithelial lymphocytes (iIEL)³ (8). Integrin ₄₇ binds to mucosal addressin cell adhesion molecule-1 (MAdCAM-1), a molecule expressed selectively on intestinal endothelial cells that mediates T cell interactions with intestinal vessels (9, 10, 11, 12). This interaction appears to be important for T cell homing to the lamina propria as intestinal T cell numbers are reduced in ₇ integrin-deficient mice (5). In addition, mAbs that recognize ₄₇ or MAdCAM-1 inhibit lymphocyte migration into Peyer’s patches and the lamina propria (13), and inhibit the generation of intestinal inflammation in murine models (14, 15).

Overall, there appears to be a correlation between the distribution of an adhesion molecule on the cell surface and its capacity to mediate rolling on endothelia. Molecules that have the capacity to mediate rolling on endothelia, such as L-selectin, are displayed on the microvillus tips protruding from the surface of the cells (16, 17). Furthermore, molecules expressed on microvillus tips are especially able to mediate rolling under conditions of flow (18, 19). Consistent with the finding that ₄₇ is expressed on the microvillus tips of ₄₇-transfected K562 cells (17), it mediates rolling of lymphocytes on endothelia (9, 20). In addition, ₄₇ can mediate the stronger interaction that is typical of an integrin after chemokine-induced activation (20). Thus, it is apparent that integrin ₄₇ is important for lymphocyte trafficking to the intestinal mucosa. Its role in T cell retention with the lamina propria or in the intestinal epithelium is not known.

The only other known ₇ integrin is _E7. It is expressed on a subset of the circulating ₄₇^high memory CD8⁺ T cells, approximately two-thirds of which coexpress the C-C chemokine receptor 9, which is selectively expressed on gut-homing T cells (21). Integrin _E7 also is expressed on >90% of CD8⁺ and 40% of CD4⁺ lamina propria T cells and iIEL (22, 23, 24). The _E7 integrin mediates T cell adhesion to epithelial cells via its interaction with epithelial cadherin (E-cadherin; Refs. 25, 26, 27, 28), leading to the suggestion that _E7 is involved in iIEL retention. In support of a role of _E7 in T cell binding to epithelial cells, _E7 is expressed on epidermotropic lymphomas (29, 30, 31). In addition, a recent report has suggested that _E7 mediates lymphocyte binding to a skin-derived epithelial cell line through an E-cadherin-independent interaction (32).

Whether the _E7 integrin is also involved in T cell extravasation into the lamina propria has not been resolved. Of note, splenocytes induced to express _E7 after culture with TGF-₁ do not localize preferentially to the intestine (33). Furthermore, _E deficiency does not modulate the localization of OVA-specific CD8⁺-transgenic T cells to the intestinal epithelium after adoptive transfer and systemic T cell activation (34). However, like ₄₇ and other molecules that can mediate leukocyte rolling on endothelial cells, _E7 is expressed on the microvillus tips of _E7-transfected K562 cells (17). In addition, lamina propria T cell numbers are reduced in _E^-/- mice to only 50% of that observed in _E^+/+ mice (35), whereas lamina propria T cell numbers are normal in mice whose T cells lack ₄ expression (36). Thus, although ₄₇ is critical in T cell localization to Peyer’s patches, another ₇ integrin appears to mediate T cell extravasation within the intestinal mucosa. These findings would be consistent with a possible role of _E7 in gut-homing T cell interactions with the intestinal endothelium, and led us to evaluate whether such endothelial cells express an _E7 ligand.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
mAbs and cell lines

mAbs used included anti E-cadherin (E4.6, Ref. 26 ; and HECD-1; Zymed, San Francisco, CA), anti-E (E7-1 (28C12), E7-2 (26F1), E7-3 (2G9; Ref. 24)), and BerACT-8 (37), anti-1 (4B4; Ref. 38), anti-₇ (Fib504; American Type Culture Collection (ATCC), Manassas, VA), anti-₄₇ (ACT-1; Ref. 39), anti-LFA-1 (D6.21; Ref. 25), anti-MHC class I (W6/32; ATCC), anti-vascular endothelial cadherin (anti-VE-cadherin; PharMingen, San Diego, CA), noncell-binding control IgG1 (P3; ATCC), and anti-human Fc (Zymed). Purified human IgG1 was obtained from Calbiochem (San Diego, CA). _E- and ₇-transfected K562 cells, the E-cadherin-Fc-producing 293 cell line, human IEL lines, the _E-transfected human B lymphoblastoid cell line JY'-E, and the human breast epithelial cell line MCF-7 were grown as described (17, 28, 40). Human intestinal microvascular endothelial cell (HIMEC) lines were isolated and grown as characterized previously (41). Human microvascular endothelial cells (HMVEC) were obtained from Clonetics (Walkersville, MD) and cultured according to the supplier’s instructions. Endothelial cells were not used beyond passage 7.

Production of soluble recombinant _E7-Fc fusion protein

To generate an _E-Fc construct, a fragment of the _E-cDNA encoding for the entire extracellular domain (nucleotides 126-3496, GenBank Accession no. NM_002208) was ligated to the hinge and Fc region of human IgG1 as described for human E-cadherin-Fc (28), incorporating a linker encoding for the peptide sequence Ala-Ser-Gly-Gly-Gly-Leu-Glu between the integrin and Fc domain to confer flexibility. The construct was ligated into the pCEP4 expression vector (Invitrogen, Carlsbad, CA). The ₇-Fc fusion protein, incorporating nucleotides 151-2323 (GenBank Accession no. NM_000889), was generated similarly. The constructs were cotransfected into human embryonic kidney (HEK)-293 cells by calcium phosphate precipitation (42), and the transfected cell line was cloned by limiting dilution. Clones producing the greatest quantity of soluble heterodimer were identified based on sandwich ELISA of the supernatant (43) using the anti-₇ mAb Fib504 for capture and the biotinylated anti-_E mAb E7-1 for detection. For biochemical analysis, 20 ml of supernatant was passed over a 300-µl protein G column, and the adsorbed protein was analyzed by SDS-PAGE followed by Coomassie blue staining as described previously (predicted M_r: -Fc = 280 kDa, -Fc = 160 kDa, -Fc = 120 kDa, -Fc = 320 kDa, -Fc = 240 kDa) (44). A soluble _E construct containing a mutation in the metal ion-dependent adhesion site (MIDAS) was generated by exchanging fragment 236-1256 of the _E-Fc with a similar fragment from the full-length _E(D190A) subunit (44) using BpuII02I sites. Cotransfection and subcloning of _E(D190A)₇-Fc was performed as described for _E7-Fc.

Adhesion assays

Static cell-cell adhesion assays were conducted in 96-well tissue culture plates as previously described (25), with the modification that labeled cells were incubated on confluent HIMEC monolayers for 30 min at 37°C. Nonadherent cells were removed by inverting the plate in HEPES-buffered saline/50 mM dextrose/1 mM Ca²⁺/Mg²⁺/Mn²⁺ for 30 min and subsequently flicking the inverted plate. For cell-fusion protein adhesion assays, ELISA plates were coated with goat anti-human Fc mAb as described (28) and incubated for 14–18 h at 4°C with 100 µl/well of undiluted supernatant from HEK-293 cells producing high levels of _E7-Fc (100 ng/well). In parallel, adherent cells were released from the tissue culture flask using 0.02% trypsin/HEPES-buffered saline/2 mM Ca²⁺. After fluorescence labeling, 3 x 10⁴ cells were added to each well, and the adhesion assay was performed in the presence of 1 mM Ca²⁺/Mg²⁺/Mn²⁺ as described (28), unless otherwise indicated. For mAb blocking experiments, cells were incubated with 20 µg/ml of purified Ab for 15 min on ice before performing the assay. The concentration of _E7-Fc and of _E(D190A)₇-Fc in culture supernatants were estimated to be 1 and 0.8 µg/ml, respectively, based upon anti-Fc-based ELISA. For adhesion assays comparing HIMEC binding to both fusion proteins, the _E7-Fc containing supernatant was diluted to 0.8 µg/ml before use.

PCR assay

For PCR, 3 x 10⁶ MCF-7 cells and a similar number of HIMEC were harvested using cells from a culture passage at which they were known to bind to the _E7-Fc in adhesion assays. Then, mRNA was prepared using the QuickPrep Micro mRNA purification Kit (Pharmacia Biotech, Uppsala, Sweden). After DNAaseI digestion, cDNA synthesis was initiated using a cDNA Synthesis Kit (Clontech, Palo Alto, CA). PCR was performed using G3PDH primers (Clontech) and the primer pair E-cad-5' (5'-aagagagactgggttattcctcccatc-3') and E-cad-3' (5'-gccatcgttgttcactggatttgt-3') encoding for a 848-bp piece of the N-terminal region of E-cadherin, spanning 6 introns, which include a total of 166 bp. The PCR conditions for E-cadherin detection were 94°C for 3 min followed by 40 cycles of 94°C for 1 min, 60°C for 1 min, and 72°C for 2 min.

Flow cytometry analysis

FACS analysis was performed as previously described (24) using the FACScan flow cytometer (Becton Dickinson, Mountain View, CA) and CellQuest software (Becton Dickinson).

Stamper-Woodruff assays

A modified Stamper-Woodruff assay was performed to assess cell binding to lamina propria (45). BCECF labeled JY'-E (which express _E7 due to the presence of an endogenous ₇ subunit) or JY'-v cells were incubated in RPMI 1640 medium on 10-µm cryosections of human proximal small intestine under static conditions for 30 min at 37°C. Then, nonadherent cells were removed by dipping the slides in PBS, and the tissue was fixed with 7% formalin/PBS. The number of bound cells per 0.5-mm villus length was determined via visual inspection by a observer, blinded to the cell source, using a fluorescence microscope (Nikon, model UFX-DX; Tokyo, Japan, objective). For colocalization experiments, IEL lines were used because of their high level of _E7 cell surface expression. Tissue sections were preincubated with 10 µg/ml of the endothelial cell-specific anti-VE-cadherin mAb for 45 min, followed by 1:150 Texas Red-conjugated goat anti-mouse (Jackson ImmunoResearch, West Grove, PA) and washed in HEPES-buffered saline. Thus, BCECF-labeled IEL, after incubation with 20 µg of the anti-_E mAb E7-1 or control-mAb (W6/32), were added to the tissue sections in HEPES-buffered saline containing 1 mM Ca²⁺, Mg²⁺, and Mn²⁺. The assay was subsequently performed as described above. Of note, the green labeling of IEL was visible under all powers of magnification. However, the red labeling of endothelial cells was only visible in the high power views. Cells were counted per 0.1-mm villus length by an unblinded observer.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Anti-_E7 mAbs partially block the binding of iIEL lines to HIMEC

To determine whether _E7 mediates binding of T lymphocytes to HIMEC, the 486 IEL-derived T cell line was used. Approximately 80% of the cells in this line constitutively expressed _E7 at moderate to high levels (mean fluorescence intensity (MFI) = 150). In two of three experiments performed, 35% of the IEL line bound to a HIMEC monolayer. When adhesion above background was considered, three different anti-_E7 mAbs blocked the IEL/HIMEC binding significantly, by 35–40%, and a fourth mAb blocked more modestly (Fig. 1). In contrast, the IEL/HIMEC interaction was blocked only marginally by an anti-₄₇ mAb and was not blocked by isotype-matched mAbs specific for MHC-I, E-cadherin, or _L2. Of note, the 486 IEL line expressed ₄₇ at only modest levels (MFI 30), that may not be sufficient for assessing adhesion in these assays. These studies suggested that _E7⁺ IEL bind to HIMEC in a static adhesion assay, in combination with other interactions that have not been defined.

View larger version (24K): [in this window] [in a new window]   FIGURE 1. Adhesion of IEL to HIMEC. Fluorescently labeled IEL line 486 cells were added to confluent monolayers of HIMEC in 96-well plates in the presence of 1 mM Mn2+/Mg2+/Ca2+. After a 30-min incubation at 37°C, input fluorescence was determined. Then, unbound cells were removed by washing, the bound fluorescence was determined, and the percentage of adherent cells was calculated. In two of three experiments, the HIMEC appeared to express E7 binding capacity, as indicated by the binding of E7-transfected K562 cells. A, Results are shown for two plates, washed separately, in experiments performed on the same day. B, Results of a second experiment performed on a separate day. Experiments are shown as the mean percentage of adherent cells bound ± SD (n = 6); *,p < 0.05; **, p < 0.01; Mann-Whitney U test comparing the adhesion in the presence of an mAb with adhesion in the absence of mAb (mAbs used: control-P3, MHC I-W6/32, E-cadherin-E4.6, 47-Act-1, LFA-1-D6.21, integrin 1-4B4, E (4 )-BerACT-8, E (3 )-E7–3, E (2 )-E7-2, E (1 )-E7-1).

K562-_E7 bind to HIMEC

To confirm that _E7 expression confers cell binding to HIMEC, static cell-cell adhesion assays were performed using _E7-transfected K562 cells (17). These cells constitutively expressed moderate levels of _E7 (MFI 150). In the experiments shown, 11.8 or 12.9% of the K562-_E7 cells bound to a confluent layer of HIMEC, as compared with only 4.2 or 5.1% of vector-transfected K562-v cells, respectively (Fig. 2). Three different anti-_E mAbs and one anti-₇ mAb blocked the adhesion of K562-_E7 cells to HIMEC, almost to the level of adhesion seen with K562-v cells (Fig. 2). In contrast, K562-_E7 adhesion to HIMEC was not blocked by isotype-matched cell-binding control mAbs (anti-MHC I, anti-_L2) or by an anti-E-cadherin mAb that is known to block the binding of E-cadherin to _E7 (26, 28) (Fig. 2B). The difference between vector- and _E7-tranfected K562 cell binding was highly reproducible in multiple independent assays using HIMEC isolated from five different donors and two independently derived K562-_E7 lines. In the ten assays performed, on average 15.6% ±3.04 of K562-_E7 cells adhered to HIMEC as compared with 4.41% ± 1.68 using vector-transfected cells (p = 0.0001, Mann-Whitney U test).

View larger version (25K): [in this window] [in a new window]   FIGURE 2. Adhesion of E7-transfected K562 cells to HIMEC. Fluorescently labeled K562 cells were added to confluent monolayers of HIMEC in 96-well plates in the presence of 1 mM Mn2+/Mg2+/Ca2+. After a 30-min incubation at 37°C input fluorescence was determined, unbound cells were removed by washing, the bound fluorescence was determined, and the percentage of adherent cells was calculated. Results of two representative experiment of 10 performed on eight separate days are shown as the mean percentage of cells bound ± SD (n = 3); *, p < 0.01; Student’s t test. (mAbs used: control-P3, MHC I-W6/32, 7-Fib504, E3-E7-3, E2-E7-2, and E1-E7-1).

Generation of a recombinant soluble _E7-Fc fusion protein

The analysis of cell-cell adhesion assays is complicated by possible interactions of multiple adhesion receptors. Also, anti-integrin mAbs can trigger intracellular signals that may modify the functions of other molecules. Thus, although suggestive, the cell-cell adhesion assays described above did not definitively demonstrate that K562-_E7 binding to HIMEC was _E7 mediated. To establish an _E7-specific adhesion assay system, a pair of _E- and ₇-Fc fusion protein encoding constructs was generated (Fig. 3A) and transfected stably into HEK-293 cells. Based upon SDS-PAGE analysis, the size of the predominant secreted protein was consistent with the predicted size of a disulfide-linked _E7 heterodimer, as it migrated with an M_r of 270 kDa under nonreducing conditions and 120 and 160 kDa under reducing conditions. In addition, weaker bands were observed at 300 and 230 kDa under nonreducing conditions, suggesting the generation of lesser amounts of and homodimers, respectively (Fig. 3B). The _E7-Fc fusion protein was not stable to purification using a protein G column and elution at low pH. However, due to the presence of human IgG1-Fc it was captured onto wells with an anti-Fc Ab for use in adhesion assays.

View larger version (23K): [in this window] [in a new window]   FIGURE 3. Structure and function of the soluble recombinant E7-Fc fusion protein. A, Structure of the E7-Fc fusion protein. In the E chain, the repeats are labeled I-VII, the extra "X" domain is shown in black and the A-domain in gray. In the -chain, the region highly conserved among subunits is shown in gray and cysteine-rich repeats are hatched. The sequence of the linker between the extracytoplasmic domain of E or 7 and the human Fc region is indicated. B, SDS-PAGE of recombinant E7-Fc fusion protein. Supernatant from E- and 7-transfected HEK-293 cells was immunoabsorbed with protein G-Sepharose, subjected to 6% SDS-PAGE under reducing (R) or nonreducing (NR) conditions, and the proteins were visualized by Coomassie Blue staining. The position of m.w. standards is indicated on the left, and the predicted positions of -, -, and - dimers are indicated by arrows on the right. C, E-cadherin-expressing MCF-7 cells adhere to plate-bound E7-Fc fusion protein. MCF-7 cells were fluorescently labeled and 30-min adhesion assays were performed on plate-bound E7-Fc or E-cadherin-Fc. Results are expressed as the mean percentage of adherent cells ± SD (n = 3); *, p < 0.01 (Student’s t test) of one representative experiment of seven performed (mAbs: MHC I-W6/32, E-cad2-HECD-1, E-cad1-E4.6, 7-Fib504, E1-E7-1, E3-E7-3).

HIMEC bind to the _E7-Fe

In additional adhesion assays, >30% of labeled HIMEC bound to _E7-Fc-coated plates (Fig. 4A). Optimal HIMEC binding was observed with 1 mM Mn²⁺/Mg²⁺/Ca²⁺ (Fig. 4A) or with 5 mM Mg²⁺ (data not shown), whereas the cells bound only weakly in the presence of 1 mM Ca²⁺ or 1 mM Mg²⁺ when Mn²⁺ was absent (data not shown). This pattern of cation dependence is similar to that observed for interactions of other integrins with their ligands (47), including _E7 binding to E-cadherin (28). Binding of HIMEC to _E7-Fc-coated plates appeared to involve the _E7 portion of the fusion protein, as HIMEC did not bind to the IgG1 or to an E-cadherin-Fc fusion protein in which the human IgG1-Fc was incorporated. Furthermore, the binding was completely blocked by two distinct anti-E mAbs and by the ₇ mAb Fib504 (Fig. 4A). Thus, HIMEC bound to an _E7-Ig Fc fusion protein through an _E7-dependent interaction.

View larger version (23K): [in this window] [in a new window]   FIGURE 4. HIMEC binding to plate-bound fusion proteins. HIMEC (A and B) and HMVEC (C) were fluorescently labeled and incubated on fusion protein-coated plates as indicated. After 30 min, the input fluorescence was determined, the unbound cells were removed by washing, the bound fluorescence was determined, and the percentage of cells bound was calculated. Results are expressed as the mean percent cells bound ± SD (n = 6); *, p < 0.01; Mann-Whitney U test, of one representative experiment of nine performed for HIMEC using five independent cell lines and four performed for HMVEC using two independent cell lines. (mAbs: MHC I-W6/32, E-cad-E4.6, 7-Fib504, E3-E7-3, E1-E7-1).

The _E MIDAS is important in _E/HIMEC interactions

To further evaluate the specificity of the interaction between the _E7-Fc and HIMEC, we generated an integrin fusion protein with a D190A mutation. This mutation results in loss of function of the MIDAS, a structural binding site crucial for integrin A domain-ligand interactions (48), based upon the observation that transfected cells expressing full-length _E(D190A)₇ do not bind to E-cadherin-Fc fusion protein-coated plates (45). Thus the _E A-domain MIDAS is critical for the _E7/E-cadherin interaction. HIMEC also failed to bind to the _E(D190A)₇-Fc (Fig. 4B), confirming that the MIDAS within the _E A domain was critical for this interaction. In addition, multiple anti-_E mAbs whose binding sites are mapped to a region close to the MIDAS motif (44) blocked both E-cadherin/_E7 and HIMEC/_E7 adhesion (data not shown). Thus, _E7 likely uses a similar region within its A domain both for binding to E-cadherin and to the ligand on HIMEC.

HIMEC do not bind to _E7 through an E-cadherin-based interaction

E-cadherin is the only known ligand for _E7. Thus, studies were performed to evaluate whether HIMEC binding to _E7 was E-cadherin mediated. Although anti-_E7 mAb blocked adhesion, the anti-E-cadherin mAb E4.6 did not block the HIMEC/_E7 interaction (Figs. 2B and 4A). As this mAb is known to block E-cadherin⁺ MCF-7 binding to _E7-Fc-coated plates (Fig. 3C), this finding suggested that the HIMEC do not bind to _E7 via E-cadherin. In support of this view, HIMEC did not adhere to E-cadherin-Fc fusion protein-coated plates (Fig. 4A), demonstrating that there was not sufficient E-cadherin on the HIMEC cell surface to mediate E-cadherin/E-cadherin interactions.

To confirm that HIMEC do not express E-cadherin, flow cytometry and PCR were used. Flow cytometry was performed using two mAbs, E4.6 and HECD-1, that recognize independent epitopes of E-cadherin. Both mAbs detected E-cadherin on the cell surface of the epithelial cell line MCF-7 but did not stain HIMEC, indicating that there was no detectable cell surface expression level of E-cadherin on the endothelial cells. In contrast, HIMEC were readily stained with a mAb that recognizes the endothelial cell-specific molecule VE-cadherin (Fig. 5A). Finally, mRNA was prepared from MCF-7 cells and from HIMEC in tandem, and RT-PCR was performed using E-cadherin-specific primers. In this analysis, E-cadherin mRNA was readily detected in MCF-7 cells, but was not observed or was seen only weakly using cDNA derived from HIMEC that were known to bind to _E7-Fc in the adhesion assay (Fig. 5B). Taken together, these findings indicate that the binding of HIMEC to _E7-Fc is mediated via an E-cadherin-independent interaction.

View larger version (28K): [in this window] [in a new window]   FIGURE 5. Analysis of E-cadherin expression on HIMEC. A, Single cell suspensions of endothelial cells and MCF-7 cells were prepared using 0.02% trypsin/2 mM Ca2+ in HBS, and FACS analysis was performed as indicated. Dashed lines represent nonbinding controls. B, cDNA was prepared in parallel from HIMEC and MCF-7 cells that were known to function in adhesion assays. Then, semiquantitative RT-PCR was performed with E-cadherin and G3PDH-specific primers as indicated.

Dermal derived HMVEC do not bind to the _E7-Fc fusion protein

To assess whether _E7-Fc binding was specific to endothelial cells derived from the intestine, the binding of HIMEC to the _E7-Fc was compared with the binding of a microvascular endothelial cell line derived from human skin (HMVEC). Although HMVEC bound efficiently to fibronectin, in contrast to HIMEC they did not bind to the _E7-Fc fusion protein (Fig. 4C). To evaluate whether there were known cell adhesion molecules expressed differently on HIMEC and HMVEC that might account for this differential adhesion, flow cytometry was performed. Both HIMEC and HMVEC lacked detectable expression of E-cadherin, MAdCAM-1, VCAM-1, or E-selectin, and expressed similar levels of PECAM-1, VE-cadherin, CD34, ₁-integrins, and MHC class I (data not shown). The only difference in cell surface adhesion molecule expression between these lines was the expression of low levels of ICAM-1 on HMVEC, which was not observed on HIMEC. These findings are consistent with previous data demonstrating that _E7 does not bind to VCAM-1, ICAM-1, or MAdCAM-1 (10) and suggest that _E7 binds to intestinal endothelial cells via a ligand that is distinct from the previously characterized integrin counterreceptors on endothelial cells.

_E7 expression enhances cell binding to lamina propria tissue

To further evaluate whether _E7 may mediate lymphocyte adhesion within the lamina propria, modified Stamper-Woodruff assays were performed (45). For these assays, JY'-E cells were used because they are smaller than K562 cells, which made it possible to evaluate their site of binding within tissue sections. JY'-_E cells bound more efficiently to the lamina propria in human intestinal tissue sections than mock-transfected JY'-v cells, suggesting that _E7 expression on JY' cells conferred their binding to the lamina propria (Fig. 6).

View larger version (9K): [in this window] [in a new window]   FIGURE 6. E7+-Jy cell binding the lamina propria in tissue sections. Integrin E or vector-transfected JY-' cells were fluorescently labeled and allowed to adhere to cryosections of human jejunal tissue for 30 min at 37°C. Unbound cells were removed by dipping the slides in buffer. After fixation, the cells bound to the lamina propria were counted per 0.5-mm villus length using a fluorescence microscope. The graph shows the median number of cells bound to the lamina propria/0.5 mm villus length ± SD (n = 7 sections), p < 0.01, Mann-Whitney U test, comparing the cells bound/0.5 mm villus length of E-transfected JY' vs vector-JY' cells. The assay shown is representative of seven performed.

View larger version (72K): [in this window] [in a new window]   FIGURE 7. IEL binding to lamina propria in cryosections of jejunum. Jejunal tissue sections were preincubated with VE-cadherin mAb. In parallel, IEL were fluorescently labeled and incubated with the anti-E mAb E7-1 or with the anti-MHC class I mAb W6/32. Subsequently, the labeled IEL were incubated with the tissue for 30 min, and the slides were dipped in buffer to remove nonadherent cells. Then, the tissue sections were fixed, coverslipped, and evaluated by fluorescence microscopy. A, Photographs show representative villi with IEL labeled in green and endothelial cells stained in red. The lower figure represents the higher magnification of the area indicated in the upper figure. The yellow color observed in some cells under higher magnification represents colocalization of red stained vessels underlying green stained cells. The dashed line shows the approximate location of the basement membrane that divides the intestinal epithelium from the lamina propria. The endothelial cell staining was not visible under low power and, therefore, seen only in the high power view. In this analysis three tissue sections were evaluated for each mAb. (scale bars: upper figures = 0.1 mm, lower figure = 0.02 mm). B, The number of IEL bound to the lamina propria endothelial cells was determined and graphed ± SD (n = 11 high power fields evaluated). Experiment is representative of one of two performed.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In considering the possible role of _E7 in intestinal T cell localization, several issues are relevant. First, _E7 is expressed on a small population of circulating, memory CD8⁺ T lymphocytes. These cells coexpress CCR9, a chemokine receptor that has been found to be selectively expressed on gut-homing T cells in the circulation (21, 49). In addition, _E7 is expressed on the microvillus tips of _E- and ₇-transfected K562 cells (17). These findings suggest a role of _E7 in a rolling interaction within the endothelium, which would be consistent with the observation that the _E7/HIMEC interaction was relatively weak. It also is possible that _E7 could mediate firm adhesion of T cells to the endothelium under appropriate conditions of stimulation, as the function of _E7 is known to be regulated by inside-out signals (28).

A possible function of _E7 in T cell interactions with the lamina propria endothelium would be consistent with the previous observation that _E-deficient mice have diminished numbers of T cells in the lamina propria (35). However, it is notable that splenic T cells induced to express _E7 do not home to the intestinal mucosa (33). In addition, _E7 did not play a role in the localization of transgenic, class I-restricted T cells to the epithelium following systemic activation (34). Thus, it is clear that _E7 expression is not essential for T cell homing to the intestinal mucosa, nor is it sufficient for this process. However, T cell homing might involve _E7/intestinal endothelial cell interactions in the context of other factors or conditions, which might include the selective expression of chemokine receptors on certain lymphocytes subpopulations, and/or the variable expression of the _E7 ligand or a chemokine by lamina propria endothelial cells.

The adhesion of _E7 transfectants to HIMEC was weak, raising question as to the significance of the interaction. However, the interaction was highly reproducible using five different HIMEC lines, two independent _E7-transfected cells lines, and the binding of IEL to HIMEC was partially blocked by anti-_E7 mAbs. Finally, the Stamper-Woodruff assays support the view that _E7 mediates binding of T cells to the intestinal lamina propria endothelial cells in tissues, possibly indicating that these findings are physiologically relevant. Many weak molecular interactions, as assessed in an adhesion assay, can be highly significant in terms of cellular function, such as selectins binding to their glycosylated ligands (50). In addition, the _E7/HIMEC binding could be important for signal transduction that might alter endothelial cell and/or T cell phenotype, adhesion, or function. Indeed, some endothelial cell surface molecules that are important in adhesion also transduce signals to the endothelial cells, including ICAM-1, PECAM-1, and L-selectin (51, 52, 53, 54), and integrins are known to transduce signals within leukocytes (55). Finally, it is possible that the weak adhesion observed in these assays indicate that there are only low levels of the _E7 ligand expressed on the primary HIMEC in culture, but that but these levels might be up-regulated by inflammatory or other cytokines. Regulated expression is a characteristic of other integrin ligands expressed on endothelial cells, including ICAM-1, an _L2 ligand; VCAM-1, an ₄₁ ligand (54); and MAd-CAM-1, an ₄₇ ligand (56, 57). Studies are underway to further characterize the integrin _E7 counterreceptor expressed on intestinal endothelial cells. Once the identity of this molecule is defined at the molecular level, it will be possible to determine the specific functional impact of the _E7/HIMEC counterreceptor interaction.

The data presented herein demonstrate, for the first time, the presence of another counterreceptor for the mucosal integrin _E7 on nonepithelial cells. This second _E7 ligand may play a role in the selective recruitment of _E7⁺ T cell subsets into the intestinal mucosa in combination with other factors, or could transduce signals in either T cells or intestinal endothelial cells. As an anti-_E7 mAb ameliorates or abrogates intestinal inflammation in two murine models (Ref. 58 and M.P. Schoen, J. Donohue, and C.M.P., unpublished data), agents that target the _E7 interaction with its ligands in the intestine may have efficacy in the treatment of intestinal inflammation. Further studies will be required to define the basic impact of the _E7/HIMEC counterreceptor interaction upon intestinal T cells homing and in the functions of intestinal T cells and endothelial cells.

	Acknowledgments

 
We thank Dr. Michael Brenner (Brigham and Women’s Hospital, Boston, MA) for generously providing the human E-cadherin-Fc-producing cell line, _E7 expressing JY'-E cells, the anti-E-cadherin mAb E4.6, the _E(D190A)₇-Fc, and for his ongoing support; Marie Stockhausen for help with the culture of IEL lines; Dr. David Erle (University of California, San Francisco, CA) for _E7-transfected K562 cells; Drs. Michael Briskin, Chafen Lu, and Dominic Picarella (Leukosite, Cambridge, MA) for providing important reagents and for useful discussion; and Dr. Jay Ponder (Washington University Medical School, St. Louis, MO) for advice on linker construction within the _E7-Fc fusion protein encoding constructs. In addition, we want to thank members of the Parker and Brenner Laboratories for helpful discussions.

	Footnotes

 
¹ This work was supported by grants from the Deutsche Forschungsgesellschaft and the Crohn’s and Colitis Foundation of America (to U.G.S.), a grant from the Crohn’s and Colitis Foundation of America (to J.M.G.H.), and by National Institutes of Health Grants T32HLO7633 (to M.C.), AI43992 and DK52978 (to C.M.P.), and DK02407 (to D.G.B.).

² Address correspondence and reprint requests to Dr. Christina M. Parker, Division of Rheumatology, Immunology, and Allergy, Brigham and Women’s Hospital, Smith Building Room 552B, 1 Jimmy Fund Way, Boston, MA 02115.

³ Abbreviations used in this paper: iIEL, intestinal intraepithelial lymphocytes; MAdCAM-1, mucosal addressin cell adhesion molecule-1; HIMEC, human intestinal microvascular endothelial cells; HMVEC, human microvascular endothelial cells; MIDAS, metal ion-dependent adhesion site, VE-cadherin, vascular endothelial cadherin; E-cadherin, epithelial cadherin; HEK, human embryonic kidney; MFI, mean fluorescence intensity.

Received for publication June 21, 2000. Accepted for publication December 13, 2000.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Butcher, E. C., M. Williams, K. Youngman, L. Rott, M. Briskin. 1999. Lymphocyte trafficking and regional immunity. Adv. Immunol. 72:209.[Medline]
2. Grabovsky, V., S. Feigelson, C. Chen, D. A. Bleijs, A. Peled, G. Cinamon, F. Baleux, F. Arenzana-Seisdedos, T. Lapidot, Y. van Kooyk, et al 2000. Subsecond induction of ₄ integrin clustering by immobilized chemokines stimulates leukocyte tethering and rolling on endothelial vascular cell adhesion molecule 1 under flow conditions. J. Exp. Med. 192:495.[Abstract/Free Full Text]
3. Tanaka, Y., D. H. Adams, S. Hubscher, H. Hirano, U. Siebenlist, S. Shaw. 1993. T-cell adhesion induced by proteoglycan-immobilized cytokine MIP-1. Nature 361:79.[Medline]
4. Campbell, J. J., J. Hedrick, A. Zlotnik, M. A. Siani, D. A. Thompson, E. C. Butcher. 1998. Chemokines and the arrest of lymphocytes rolling under flow conditions. Science 279:381.[Abstract/Free Full Text]
5. Wagner, N., J. Lohler, E. J. Kunkel, K. Ley, E. Leung, G. Krissansen, K. Rajewsky, W. Muller. 1996. Critical role for ₇ integrins in formation of the gut-associated lymphoid tissue. Nature 382:366.[Medline]
6. Schweighoffer, T., Y. Tanaka, M. Tidswell, D. Erle, K. Horgan, G. Ginther Luce, A. Lazarovits, D. Buck, S. Shaw. 1993. Selective expression of integrin ₄₇ on a subset of human CD4⁺ memory T cells with hallmarks of gut-trophism. J. Immunol. 151:717.[Abstract]
7. Rott, L. S., M. J. Briskin, D. P. Andrew, E. L. Berg, E. C. Butcher. 1996. A fundamental subdivision of circulating lymphocytes defined by adhesion to mucosal addressin cell adhesion molecule-1: comparison with vascular cell adhesion molecule-1 and correlation with ₇ integrins and memory differentiation. J. Immunol. 156:3727.[Abstract]
8. Farstad, I. N., T. S. Halstensen, B. Lien, P. J. Kilshaw, A. I. Lazarovits, P. Brandtzaeg, A. I. Lazarovits. 1996. Distribution of ₇ integrins in human intestinal mucosa and organized gut-associated lymphoid tissue: [Published erratum appears in 1997 Immunology 91:322.]. Immunology 89:227.[Medline]
9. Berlin, C., E. L. Berg, M. J. Briskin, D. P. Andrew, P. J. Kilshaw, B. Holzmann, I. L. Weissman, A. Hamann, E. C. Butcher. 1993. ₄₇ integrin mediates lymphocyte binding to the mucosal vascular addressin MAdCAM-1. Cell 74:185.[Medline]
10. Strauch, U. G., A. Lifka, U. Gosslar, P. J. Kilshaw, J. Clements, B. Holzmann. 1994. Distinct binding specificities of integrins ₄₇ (LPAM-1), ₄₁ (VLA-4), and _IEL7. Int. Immunol. 6:263.[Abstract/Free Full Text]
11. Briskin, M., D. Winsor-Hines, A. Shyjan, N. Cochran, S. Bloom, J. Wilson, L. M. McEvoy, E. C. Butcher, N. Kassam, C. R. Mackay, et al 1997. Human mucosal addressin cell adhesion molecule-1 is preferentially expressed in intestinal tract and associated lymphoid tissue. Am. J. Pathol. 151:97.[Abstract]
12. Williams, M. B., E. C. Butcher. 1997. Homing of naive and memory T lymphocyte subsets to Peyer’s patches, lymph nodes, and spleen. J. Immunol. 159:1746.[Abstract]
13. Hamann, A., D. P. Andrew, D. Jablonski-Westrich, B. Holzmann, E. C. Butcher. 1994. Role of ₄-integrins in lymphocyte homing to mucosal tissues in vivo. J. Immunol. 152:3282.[Abstract]
14. Picarella, D., P. Hurlbut, J. Rottman, X. Shi, E. Butcher, D. J. Ringler. 1997. Monoclonal antibodies specific for ₇ integrin and mucosal addressin cell adhesion molecule-1 (MAdCAM-1) reduce inflammation in the colon of scid mice reconstituted with CD45RB^high CD4⁺ T cells. J. Immunol. 158:2099.[Abstract]
15. Kato, S., R. Hokari, K. Matsuzaki, A. Iwai, A. Kawaguchi, S. Nagao, T. Miyahara, K. Ihoh, H. Ishii, S. Miura. 2000. Amelioration of murine experimental colitis by inhibition of mucosal addressin cell adhesion molecule-1. J. Pharmacol. Exp. Ther. 295:183.[Abstract/Free Full Text]
16. Erlandsen, S. L., S. R. Hasslen, R. D. Nelson. 1993. Detection and spatial distribution of the ₂ integrin (Mac-1) and L-selectin (LECAM-1) adherence receptors on human neutrophils by high-resolution field emission SEM. J. Histochem. Cytochem. 41:327.[Abstract]
17. Abitorabi, M. A., R. K. Pachynski, R. E. Ferrando, M. Tidswell, D. J. Erle. 1997. Presentation of integrins on leukocyte microvilli: a role for the extracellular domain in determining membrane localization. J. Cell Biol. 139:563.[Abstract/Free Full Text]
18. von Andrian, U. H., S. R. Hasslen, R. D. Nelson, S. L. Erlandsen, E. C. Butcher. 1995. A central role for microvillous receptor presentation in leukocyte adhesion under flow. Cell 82:989.[Medline]
19. Stein, J. V., G. Cheng, B. M. Stockton, B. P. Fors, E. C. Butcher, U. H. von Andrian. 1999. L-selectin-mediated leukocyte adhesion in vivo: microvillus distribution determines tethering efficiency, but not rolling velocity. J. Exp. Med. 189:37.[Abstract/Free Full Text]
20. Berlin, C., R. F. Bargatze, J. J. Campbell, U. H. Vonandrian, M. C. Szabo, S. R. Hasslen, R. D. Nelson, E. L. Berg, S. L. Erlandsen, E. C. Butcher. 1995. ₄ integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 80:413.[Medline]
21. Zabel, B. A., W. W. Agace, J. J. Campbell, H. M. Heath, D. Parent, A. I. Roberts, E. C. Ebert, N. Kassam, S. Qin, M. Zovko, et al 1999. Human G protein-coupled receptor GPR-9–6/CC chemokine receptor 9 is selectively expressed on intestinal homing T lymphocytes, mucosal lymphocytes, and thymocytes and is required for thymus-expressed chemokine-mediated chemotaxis. J. Exp. Med. 190:1241.[Abstract/Free Full Text]
22. Cerf-Bensussan, N., A. Jarry, N. Brousse, B. Lisowska-Grospierre, D. Guy-Grand, C. Griscelli. 1987. A monoclonal antibody (HML-1) defining a novel membrane molecule present on human intestinal lymphocytes. Eur. J. Immunol. 17:1279.[Medline]
23. Kilshaw, P. J., S. J. Murant. 1990. A new surface antigen on intraepithelial lymphocytes in the intestine. Eur. J. Immunol. 20:2201.[Medline]
24. Russell, G., C. Parker, K. Cepek, D. Mandelbrot, A. Sood, E. Mizoguchi, E. Ebert, M. Brenner, A. Bhan. 1994. Distinct structural and functional epitopes of the _E7 integrin. Eur. J. Immunol. 24:2832.[Medline]
25. Cepek, K. L., C. M. Parker, J. L. Madara, M. B. Brenner. 1993. Integrin _E7 mediates adhesion of T-lymphocytes to epithelial cells. J. Immunol. 150:3459.[Abstract]
26. Cepek, K. L., S. K. Shaw, C. M. Parker, G. J. Russell, J. S. Morrow, D. L. Rimm, M. B. Brenner. 1994. Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the _E7 integrin. Nature 372:190.[Medline]
27. Karecla, P., S. Bowden, S. Green, P. Kilshaw. 1995. Recognition of E-cadherin on epithelial cells by the mucosal T cell integrin _M2907(_E7). Eur. J. Immunol. 25:852.[Medline]
28. Higgins, J. M. G., D. A. Mandelbrot, S. K. Shaw, G. J. Russell, E. A. Murphy, Y.-T. Chen, W. J. Nelson, C. M. Parker, M. B. Brenner. 1998. Direct and regulated interaction of integrin _E7 with E-cadherin. J. Cell Biol. 140:197.[Abstract/Free Full Text]
29. Sperling, M., P. Kaudewitz, F. O. Braun, H. Stein. 1989. Reactivity of T cells in mycosis fungoides exhibiting marked epidermotropism with the monoclonal antibody HML-1 that defines a membrane molecule on human mucosal lymphocytes. Am. J. Pathol. 134:955.[Abstract]
30. Simonitsch, I., B. Volc-Platzer, I. Mosberger, T. Radaszkiewicz. 1994. Expression of monoclonal antibody HML-1-defined _E7 integrin in cutaneous T cell lymphoma. Am. J. Pathol. 145:1148.[Abstract]
31. Dietz, S. B., D. Whitaker-Menezes, S. R. Lessin. 1996. The role of _E7 integrin (CD103) and E-cadherin in epidermotropism in cutaneous T-cell lymphoma. J. Cutan. Pathol. 23:312.[Medline]
32. Brown, D. W., J. Furness, P. M. Speight, G. J. Thomas, J. Li, M. H. Thornhill, P. M. Farthing. 1999. Mechanisms of binding of cutaneous lymphocyte-associated antigen-positive and _E7-positive lymphocytes to oral and skin keratinocytes. Immunology 98:9.[Medline]
33. Aust