Influence of 1 Integrin Intracytoplasmic Domains in the Regulation of VLA-4-Mediated Adhesion of Human T Cells to VCAM-1 under Flow Conditio
http://www.100md.com
免疫学杂志 2005年第14期
Abstract
The VLA-4 integrin supports static cell-cell, cell-matrix adhesion, and dynamic interactions with VCAM-1. Although functions for well-conserved 1 integrin cytoplasmic domains in regulating static cell adhesion has been established, the molecular basis for 1 integrin-dependent arrest on VCAM-1 under flow conditions remains poorly understood. We have transfected the 1 integrin-deficient A1 Jurkat T cell line with 1 cDNA constructs with deletions of the NPXY motifs and specific mutations of tyrosine residues. Deletion of either NPXY motif impaired static adhesion induced by CD2 or CD47 triggering or direct 1 integrin stimulation. In contrast, PMA-induced adhesion to VCAM-1 was unaffected by deletion of the NPIY motif and only slightly impaired by deletion of NPKY. Moreover, deletion of the NPIY motif resulted in enhanced rolling and reduced arrest on VCAM-1 under shear flow conditions. In contrast, deletion of the NPKY motif did not alter arrest under flow. Although tyrosine to phenylalanine substitutions within two NPXY motifs did not alter static adhesion to VCAM-1, these mutations enhanced arrest on VCAM-1 under flow conditions. Furthermore, although deletion of the C'-terminal 5 AA of the 1 cytoplasmic domain dramatically impaired activation-dependent static adhesion, it did not impair arrest on VCAM-1 under flow conditions. Thus, our results demonstrate distinct structural requirements for VLA-4 function under static and shear flow conditions. This may be relevant for VLA-4 activity regulation in different anatomic compartments, such as when circulating cells arrest on inflamed endothelium under shear flow and when resident cells in bone marrow interact with VCAM-1- positive stromal cells.
Introduction
Lymphocyte recruitment into sites of inflammation requires multiple adhesion and signaling interactions with vascular endothelium. These interactions involve functionally specialized macromolecules on the surface of leukocytes and endothelial cells, acting in a multistep pathway (1, 2, 3, 4). L-selectin (CD62L), constitutively expressed by most leukocytes, and P- and E-selectins, up-regulated on activated endothelial cells, mediate the initial leukocyte tethering and rolling along the endothelium (5, 6, 7). This labile adhesion to the vessel walls brings the cells into proximity with chemoattractants displayed on the endothelial surface, which, by coupling to G proteins (8), transduce signals that activate integrin-mediated firm adhesion (9, 10, 11). Several integrins, including VLA-4 (41) and LFA-1 (L2), are involved in firm adhesion. VLA-4 (12), the mucosal homing receptor 47 (13, 14) and, under special circumstances, L2 (15, 16) are also able to mediate rolling.
VLA-4 is a transmembrane 41 heterodimeric receptor that mediates cell-cell and cell-matrix interactions (17, 18). It is widely expressed on mononuclear leukocytes and plays a major role in several cellular functions, including cell spreading and migration and recruitment of leukocyte subsets into inflammatory sites. The VLA-4 ligand VCAM-1 is expressed on vascular endothelium at sites of inflammation and on other selected sites, such as bone marrow stroma, where interaction with VLA-4 mediates stem cell attachment and mobilization (19, 20, 21, 22, 23, 24). VLA-4 mediates static adhesion in the bone marrow hemopoietic compartment by interacting with VCAM-1, which is constitutively expressed on bone marrow stromal cells (25). Moreover, interaction between leukemic cell VLA-4 and stromal fibronectin is crucial for minimal residual disease and acute myeloid leukemia progression (26). Whereas most leukocyte integrins support stationary cell-cell or cell-matrix adhesion, VLA-4 can also mediate versatile dynamic interactions with VCAM-1 under shear flow (12, 14). The interaction of VLA-4 integrin with VCAM-1 can support all of the adhesive steps required for the arrest of lymphocytes on inflamed vascular endothelium (12, 14, 27). Integrins sharing the 4-chain, such as 41 and 47, retain basal recognition of ligand at adhesive contacts under shear stress and can facilitate the spontaneous firm arrest of leukocytes on VCAM-1-expressing endothelium, bypassing the requirement for chemokine stimulation (28, 29, 30). Thus, VLA-4 mediates cell adhesion in distinct environmental conditions.
The molecular basis for support of 1 integrin-dependent rolling and spontaneous arrest on VCAM-1 under flow conditions are poorly understood. In the current study, we used the 1-negative Jurkat T cell line A1 (31) to investigate the role of specific intracytoplasmic motifs of the 1 subunit of the VLA-4 molecule to identify the specific domains critical for both static adhesion and dynamic adhesion under conditions of shear flow. We show that specific intracytoplasmic 1 integrin sequences, including two highly conserved motifs (NPIY and NPKY) and the carboxyl-terminal end of the 1 tail are differentially involved in the regulation of 1 integrin adhesion under dynamic or static conditions. These differences may reflect different roles played by VLA-4, such as in inflammation or in hemopoietic progenitor maturation and mobilization.
Materials and Methods
Reagents and Abs
The CD29 (K20) and CD2 (D66 and X11) mAbs were produced in our laboratory and were described elsewhere (32, 33, 34, 35). 4 and 5 mAbs were purchased from Immunotech; CD47 (B6H12) was produced in our laboratory from the American Type Culture Collection cells (American Type Culture Collection). TS2/16, which stimulates direct 1 integrin activation, was a kind gift from Dr. Sanchez Madrid (Hospital de la Princesa, Madrid, Spain). PMA was purchased from Sigma-Aldrich; VCAM-1, fibronectin (FN),3 and TNF- were obtained from R&D Systems.
Cells
The Jurkat T cell line JE6.1 (American Type Culture Collection) and the 1 integrin-deficient Jurkat T cell line A1 (31) were both cultured in RPMI 1640 (Invitrogen Life Technologies) supplemented with 5% FCS (Invitrogen Life Technologies), 50 U/ml penicillin, 50 μg/ml streptomycin, 2 mM L-glutamine, and 1 mM pyruvate (Invitrogen Life Technologies). The A1 cell line lacks expression of the 1 integrin subunit and is unable to adhere to 1 integrin ligands, including VCAM-1 and FN (31). The transformed human umbilical endothelial cell line EA.hy926 was kindly provided by Dr. Cora Jean Edgell (University of North Carolina, Chapel Hill, NC) and cultured in DMEM (Invitrogen Life Technologies) supplemented with 20% FCS. They were seeded on Lab-Tek chamber slides (Nalge Nunc International), allowed to grow to confluence, and activated with 25 ng/ml TNF- for 24 h before use.
cDNA, transfections, and expression
The wild-type and mutant 1 integrin constructs used in this study (see Table I) have been previously described (31). Areas targeted included tyrosine residues Y783 and Y795, representing potential sites of phos-phorylation, the two NPXY internalization motifs, and the last five amino acids of the 1 cytoplasmic tail. Stable transfections were produced using Effectene reagent (Qiagen) according to the manufacturer’s protocol and cells were selected with 1 g/L geneticin (Invitrogen Life Technologies). Single-color flow cytometry analysis was performed by staining cells with saturating amounts of mouse anti-human CD29 mAb K20, followed by rabbit anti-mouse FITC-conjugated secondary Ab (DakoCytomation). Stained cells were sorted on a FACSVantage SE (BD Biosciences) flow cytometer.
Adhesion experiments
Adhesion assays were performed on 96-well microplates (Costar). Plates were incubated overnight at +4°C with 2.5 μg/ml VCAM-1 or 5 μg/ml FN. Unbound protein binding sites were subsequently blocked with PBS/0.1% BSA for 1 h at room temperature. Cells were labeled with green fluorescent dye CellTracker Green (CMFDA) at a final concentration of 1 μg/ml for 5 x 106 cells for 30 min at 37°C. Cells were then washed and resuspended in HBSS medium complemented with 1 mM CaCl2 and MgCl2 for stimulation with various effectors at 37°C for 20 min. PMA was used at 10 ng/ml and TS2/16, 4B4, CD2 mAb pair D66 and X11, and B6H12 mAbs were used at 10 μg/ml. Cells were allowed to settle in the microplates for 30 min at 37°C. Cell fluorescence was measured by a fluorescence plate reader (Cytofluor; Millipore) before and after washing in warm PBS with 1 mM CaCl2 and MgCl2. Control wells were coated with 0.1% BSA. Specific adhesion was calculated for each well as: (fluorescence before wash – control well fluorescence)/(fluorescence after wash – control well fluorescence) x 100. Data are expressed as mean ± SD of specific adhesion per three replicate wells.
Flow chamber and laminar flow adhesion assay
The flow chamber (Immunetics) used has been previously described elsewhere (36, 37, 38). Briefly, the chamber allows stabilized laminar flow between 0.1 and 2 dyn/cm2. It was mounted on an Axiovert 25 B/W epifluorescence inverted microscope (Zeiss) for direct visualization in real time of dynamic cell adhesion process using a x10 objective. The microscope was coupled to an Axiocam high-resolution numeric camera (Zeiss) to record 10 random fields/coverslip of 1 mm2. The system was directly linked to a Power Macintosh G3 (Apple Computer) with a video card (Micromotion DC30; Pinnacle Systems). Commercial software (Adobe Premiere and Adobe Photoshop; Adobe Systems) was used to count the number of arrested cells and mean values ± SD were calculated per square millimeter.
Cells were recovered in HBSS medium supplemented with 1 mM CaCl2 and MgCl2 and injected through the flow chamber using a withdrawal syringe pump (Harvard Apparatus), at a concentration of 1 x 106/ml, on VCAM-1-coated Lab-Tek chamber slides (Nalge Nunc International). Chamber slides were coated overnight at +4°C with VCAM-1 at 2.5 μg/ml and washed twice in PBS and unbound protein binding sites were blocked with BSA for 1 h at room temperature before their use. For some experiments, cells were injected through the flow chamber on TNF--activated EA cells adherent to Lab-Tek chamber slides.
All flow experiments were performed at 37°C. Temperature was maintained by warming the microscope plate (Warming Plate; Minitub). Each flow experiment has been done at least in triplicate. In rolling experiments, lymphocyte interactions with coated sVCAM-1 in different fields were video recorded for 1 min using the Axiocam video camera. Rolling velocity was measured by determining the number of frames (each 1/24th s length of time) it took 30 cells to cross a 350-μm length field.
Detachment assay
A1-1 control cells and A1-1NPIY or A1-1NPKY cells were stained with fluorescence dyes. Green fluorescent fluorescein diacetate and orange fluorescence tetramethylrodamine were used for control A1-1 and A1-1NPXY, respectively. Cell adhesion was evaluated without stimulation and after a 20-min incubation at 37°C with various stimuli at optimal concentration, namely, PMA (10 ng/ml) and TS2/16 stimulating CD29 mAb at 10 μg/ml as indicated for static adhesion. Cells were mixed in equal amounts at 1 x 106 cells/ml final concentration and allowed to adhere on Lab-Tek chamber slides coated with 2.5 μg/ml VCAM-1 and saturated at 37°C with 0.1% BSA. The strength of adhesion was evaluated by counting residual cells adhering on Lab-Tek slides after washing for 1 min with each increasing shear stress of warm medium (0.5–30 dyn/cm2) on chamber slides. The amount of cells resting adherent on 10 different fields at each shear stress conditions was counted. Comparisons between the A1-1 control cells and the various transfectants were performed using a variance analysis test, with p 0.05 as the statistically significant value.
Fluorescent dyes
For flow experiments, cells were stained with fluorescent dyes purchased from Molecular Probes. Green fluorescent fluorescein diacetate (CellTracker Green CMFDA) is a green tracer with excitation at 522 nm and emission at 529 nm; orange fluorescent tetramethylrhodamine (CellTracker Orange CMTMR) is a red tracer with excitation at 541 nm and emission at 547 nm. Transfected cells were labeled in the culture medium at a concentration of 5 x 106 cells/ml with the different probes (0.5–2 μg/ml) for 30 min at 37°C in the dark, then washed twice before use.
Statistical analysis
Results from static adhesion assays of at least three independent experiments in triplicate wells were pooled and expressed as mean ± SD. Whole data were analyzed by the EpiInfo 6.0 software (Centers for Disease Control and Prevention) using the ANOVA parametric test for variance comparison and by the Kruskal-Wallis test, with p 0.05 as the statistically significant value. For each condition, results of transfectants were compared with control A1-1 cells. For shear flow assays, the number of firm arrested cells, counted in 10 different fields, were recovered from at least three different experiments. Whole results of the firm arrest of A1-1 control cells and the various transfectants were compared using the Kruskal-Wallis test and the ANOVA test for variance comparison, with p 0.05 as the statistically significant value.
Results
Reconstitution of 1 integrin-deficient cell line
The A1 Jurkat T cell derivative lacks expression of the common 1 integrin subunit (31). Therefore, A1 cells do not express either VLA-4 or VLA-5 molecules and exhibit complete loss of adhesion to FN (31). A1 cells were still able to secrete IL-2 in response to classical T cell activation stimuli such as PMA plus CD3 mAb and suboptimal mitogenic concentrations of CD3 mAb did induce Ca2+ release from intracellular stores (data not shown).
We stably transfected A1 cells with the human 1A cDNA, which resulted in re-expression of VLA-4 and VLA-5 on the cell surface (Fig. 1). There was no change in expression of surface molecules capable of activating integrins, such as CD2, CD3, CD28, CD47, or CD99 (data not shown).
Deletion of these motifs also significantly impaired adhesion to both VCAM-1 (Figs. 3E and 4E) and FN (Figs. 3F and 4F) induced by integrin-activating pairs of CD2-specific mAb pairs or a CD47-specific mAb. Thus, our results demonstrate a role for the NPXY motifs in the 1 integrin tail in adhesion induced by the activating 1-specific mAb and inside-out signals mediated by CD2 and CD47. In addition, there is a role for the NPKY motif, but not the NPIY motif, in PMA-induced adhesion to VCAM-1 specifically.
The results obtained in the static adhesion assay on VCAM-1 were confirmed by using the detachment assay (Fig. 5). A1-1NPIY cells (Fig. 5, A and B) and A1-1NPKY cells (Fig. 5, C and D) were stimulated with PMA (Fig. 5, A and C, respectively) and TS2/16 (Fig. 5, B and D, respectively) and allowed to adhere in the Lab-Tek chamber slide. When increasing shear stresses of washing medium were administrated, significant differences (see Fig 5 legend) appeared between the NPXY-deleted cells and their control. The strength of adhesion of A1-1NPIY cells was lower than that of control cells after stimulation with the TS2/16 mAb (Fig. 5B), but not with PMA (Fig. 5A). For the A1-1NPKY cells, the strength of adhesion was lower than that of control cells after stimulation with both PMA (Fig. 5C) and TS2/16 (Fig. 5D).
Deletion of the 1 NPIY intracytoplasmic motif, but not NPKY, affects the firm arrest of T cells on VCAM-1 under flow conditions
Since static adhesion experiments showed that VLA-4 adhesion regulation was dependent on NPXY motifs, we next performed experiments to determine whether firm arrest mediated by VLA-4 was also related to these motifs. Under flow conditions at a shear stress of 1 dyn/cm2 as described in Materials and Methods, the number of A1-1NPIY cells that arrested on coated VCAM-1 was significantly lower then A1-1 control cells. However, no differences were observed at a higher shear stress (2 dyn/cm2) (Fig. 6A). The same significant impairment of A1-1NPIY firm arrest was observed at a very low shear stress of 0.5 dyn/cm2 (data not shown). When cells were stimulated with TS2/16 or CD47 mAbs, an increase of A1-1NPIY was obtained. However, adhesion of A1-1NPIY cells was significantly lower than A1-1 control cells both in basal conditions and after stimulation with TS2/16 and CD47 mAbs. At 2 dyn/cm2, a very moderate increase of firm adhesion was obtained after stimulation with TS2/16 and CD47 mAbs, but not in A1-1NPIY cells (Fig. 6A). A1-1NPIY-deleted cells rolled faster than A1-1 control cells (Fig. 6B). When cells were allowed to roll on TNF--activated human endothelial cells, firm adhesion of A1-1NPIY cells was significantly lower than that of A1-1 cells, both in basal conditions and after stimulation with TS2/16 mAb (Fig. 6C).
In marked contrast to A1-1NPIY-deleted cells, no difference was observed in the firm arrest of A1-1NPKY transfectants on VCAM-1 when compared with A1-1 cells at either shear stress tested (Fig. 6D). These results were confirmed when A1-1NPKY-deleted T cells were injected into the flow chamber on TNF--activated EA cells (data not shown). A significant increase of firm adhesion was obtained after TS2/16 stimulation at both 1 and 2 dyn/cm2 (Fig. 6E). Thus, our results demonstrate a role for the NPIY motif, but not the NPKY motif, in the regulation of T cell adhesion to VCAM-1 under shear flow conditions.
Deletion of the carboxyl-terminal five amino acids totally abolishes static adhesion but allows efficient firm arrest to VCAM-1 under flow conditions
Deletion of the carboxyl-terminal five amino acids of the cytoplasmic domain of the 1 tail has previously been shown to completely inhibit basal and stimulated adhesion of Jurkat T cells to FN (31). Stable A1 transfectants expressing the 1(793) deletion were also unable to adhere to VCAM-1 in a static adhesion assay, when compared with A1-1 cells (Fig. 7, A, C, and E), following either PMA stimulation (Fig. 7A) or direct 1 integrin activation by the mAb TS2/16 (Fig. 7C). Only very weak stimulation with an optimal concentration of PMA and of coated VCAM-1 was detected. Similarly, stimulation of A1-1(793) cells with integrin-activating the CD2 mAb pair or CD47-specific mAb did not enhance adhesion above unstimulated controls, both on VCAM-1 and FN-coated surfaces (Fig. 7, E and F, respectively). Similar results for static adhesion assays were obtained on FN-coated plates (Fig. 7, B, D, and F) in the presence of increasing concentrations of PMA (Fig. 7B), TS2/16 (Fig. 7D), or activating 1 integrin costimuli (Fig. 7F).
Although the 1(793) deletion had a dramatic effect on static T cell adhesion to both VCAM-1 and FN, A1-1(793) were able to firmly arrest on VCAM-1 under shear flow conditions (Fig. 7G). When whole data from at least three different experiments performed were statistically analyzed, the number of arrested cells was comparable between A1-1(793) cells and control A1-1 cells at both flow rates tested. The same pattern was observed when cells were stimulated with TS2/16 at 10 μg/ml (data not shown).
Mutation of tyrosine residues of the NPIY and NPKY motifs increases the firm arrest of T cells on VCAM-1 under flow conditions, but has no effect in static adhesion
The tyrosine residues in the NPIY(Y783) and NPKY(Y795) motifs have been implicated in downstream signaling activities and integrin functions, such as cell spreading (45), tissue invasion (46), and bacterial internalization (47). To elucidate the importance of these motifs in adhesion, transfectants were generated expressing mutant 1 subunits with tyrosine to phenylalanine substitution in each NPXY motif (A1-Y783F and A1-Y795F) or in both motifs (A1-Y783,795F).
In a static adhesion assay, A1-1 Y783F, A1-1 Y795F, and A1-1 Y783,795F cells all supported adhesion to the same extent as A1-1 wild-type cells. Similar levels of adhesion were observed under all activating conditions tested (data not shown). This was in agreement with data obtained previously with transient transfectants on FN-coated microwells (31). However, in shear flow assays, both A1-Y783F (Fig. 8A) and A1- Y795F (Fig. 8B) cells exhibited statistically significant higher levels of arrest than wild-type A1-1 at high shear stress (2 dyn/cm2). Furthermore, A1 cells expressing the double tyrosine mutant Y783,795F exhibited higher levels of arrest at both low and high shear rates (Fig. 8C). The same pattern of firm adhesion increase was observed when tyrosine-mutated cells were stimulated with TS2/16 mAb, in comparison to basal conditions. In particular, A1-1 Y783F and A1-1 Y795F firm arrest was significantly higher than in A1-1 control cells at 2 dyn/cm2. Moreover, firm arrest of A1-1 Y783,795F cells was significantly higher than A1-1 cells at both 1- and 2-dyn/cm2 shear stress (data not shown).
Discussion
We used the 1 integrin-deficient A1 Jurkat T cell line to define the structural requirements in the 1 integrin cytoplasmic domain that regulate VLA-4-mediated adhesion under both static conditions and in the presence of shear flow. The A1 cell line represents a unique cellular reagent for these studies, since the lack of endogenous 1 integrin expression in this cell line allowed us to stably express wild-type and mutant 1 integrin subunits without the complications of endogenous wild-type 1 integrin expression. Our studies reveal that integrin function under static and shear flow conditions are differentially regulated by three regions of the 1 integrin tail (Table II): 1) the membrane-proximal NPIY motif regulates VLA-4-mediated adhesion under both static and shear conditions, 2) the carboxyl-terminal region of the 1 tail has a specific role in regulating static adhesion, but not arrest under shear flow, and 3) the tyrosines in the NPXY motifs do not play a major role in regulating static adhesion but appear to negatively regulate adhesion under shear flow.
The 1 integrin NPIY motif appears to be particularly critical for optimal VLA-4 function, since deletion of this motif inhibits both activation-dependent VLA-4-mediated adhesion under static conditions and firm arrest under flow conditions. The shear flow rates used in our experiments likely mimic physiological inflammatory conditions in microvessels, when microangectasias induced by inflammation slow the shear stress to these rates (48).
It is interesting to note that deletion of the NPIY motif altered firm arrest at 1 dyn/cm2, but not at 2 dyn/cm2. This was also observed in a more physiological model, on human activated endothelial EA cells. This may be related to a potential role of the NPIY motif in regulating affinity states of VLA-4, since Chen et al. (28) demonstrated that low-affinity VLA-4 states preferentially mediate transient tethering and rolling of PBL on VCAM-1, whereas high-affinity VLA-4 states preferentially support spontaneous firm arrest of both PBL and Jurkat T cells on VCAM-1. It is possible that cells arresting at higher shear stress conditions (2 dyn/cm2) is mediated by high-affinity VLA-4, whereas cell arrest at 1 dyn/cm2 is mediated predominately by intermediate-/low-affinity VLA-4. The reduced firm adhesion of the A1-1NPIY cells was also confirmed on VCAM-1-coated Lab-Tek slides at very low shear stress, such as 0.5 dyn/cm2. This suggests that the NPIY motif might be particularly critical in regulating intermediate-/low-affinity states of VLA-4 rather than high-affinity states.
The NPIY motif has also been implicated in the binding of the cytoskeleton protein talin to the 1 cytoplasmic domain (49), and talin interaction with integrin tails is proposed to play a critical role in integrin activation (42, 50). Thus, the inability of the A1-1NPIY cells to adhere to VCAM-1 following integrin-activating signals may be due to disruption of the interaction between talin and the integrin 1 subunit. In addition, our results with adhesion under shear flow suggest a role for talin in optimal firm arrest, but only under low shear flow rates. Integrin activation by TS2/16 or cosignal stimulation by CD47 mAb was not affected in A1-1NPIY cells, since an increase in firm arrest was observed at levels comparable to those of A1-1 control cells.
The C-terminal region of the integrin 1 tail that contains the NPKY motif was found to be particularly important for static adhesion but not for adhesion under shear flow. This was most evident with the A1-1(793) cells, which exhibited defective static adhesion to both VCAM-1 and FN under static adhesion conditions, but had surprisingly normal patterns of cell arrest under shear flow conditions. In addition, like A1-1NPIY, A1-1NPKY cells showed reduced adhesion after stimulation with the integrin-activating mAb TS2/16 and complete absence of response to other inside-out stimuli that activate 1 integrins. The impairment in the adhesion properties of the NPXY-deleted cells seen in both static and detachment assays was statistically significant, although small in some cases, rather than an "all or none" effect. The existence of different pools among Jurkat T cells of cells bearing different states of VLA-4 affinity for VCAM-1 could explain these smaller differences observed with VCAM-1 compared with the effects observed on FN, where the affinity states of VLA-4 may be more homogeneous.
Like A1-1(793) cells, arrest of cells expressing the A1-1NPKY under shear flow conditions was comparable to A1 cells expressing wild-type 1. Our results also show that TS2/16 stimulation is affected following deletion of the NPXY motifs or deletion of the five C-terminal amino acids of the 1 integrin cytoplasmic tail. Although TS2/16 is thought to activate 1 integrin primarily via alterations in the 1 extracellular domain, recent studies have shown that TS2/16-induced adhesion is inhibited when Rap1 signaling is impaired (51). Thus, our results are consistent with a potential requirement for intracellular signaling in regulating integrin activation induced by the TS2/16 mAb.
In other studies, the carboxyl-terminal amino acids of the 1 tail were found to be critical for 1 integrin-dependent adhesion, as well as migration and metastasis. In an in vivo model, Stroeken et al. (46) showed that expression of a 1(793) construct in 1 integrin-deficient double knockout ESb lymphoma cells (ESb-DKO) impaired invasion and metastasis in vivo when injected in mice and greatly reduced adhesion in vitro. Since firm cell arrest on endothelium is critical for invasion in vivo, our results suggest that the carboxyl-terminal end of the 1 integrin tail may be particularly critical for regulating integrin function subsequent to firm arrest, such as transendothelial migration.
The mechanism by which the carboxyl-terminal end of the 1 integrin tail reduces integrin function remains unclear. Studies with chimeric integrins expressed in Chinese HO cells (45) suggest that the 1(793) mutant is still capable of interacting with talin. Our results suggest that structural integrity at this region of the 1 tail is particularly critical for static adhesion regulated by inside-out signaling, but not firm adhesion under hydrodynamic shear flow conditions.
The third structural feature of the integrin 1 tail is that tyrosine residues 783 and 795 in the NPXY motifs appear to play a critical role in regulating adhesion under shear flow, but not under static conditions. When either of these tyrosine residues is substituted with phenylalanine, adhesion at high shear stress rates was actually enhanced when compared with wild-type A1-1 cells. In contrast, adhesion under static conditions was unaffected by these mutations, consistent with previous results (31). Similar results were observed when both tyrosines were substituted with phenylalanine, although now enhanced firm arrest was also observed at low shear rates.
Although the role of tyrosine phosphorylation of the 1 tail remains unresolved, our results nevertheless suggest a potential function for phosphorylation of these tyrosine residues in regulating adhesion under shear flow. A role for these tyrosines in migration has also been suggested, since a tyrosine to phenylalanine substitution at Y795 impairs the migration of mouse GD25 fibroblasts (52). Furthermore, the Y783/795F double mutation markedly affects cell migration and cytoskeletal architecture, but not adhesion (52). However, when expressed in 1-deficient lymphoid ESb-DKO cells, the Y783/795F double mutation impaired adhesion and migration in vitro, but not in in vivo invasion in the liver (46). Other studies have also implicated these tyrosine residues in integrin affinity modulation (40), as tyrosine to alanine substitutions at these sites impairs LIBS mAb binding. However, tyrosine to phenylalanine substitutions did not affect binding of this mAb to 1 and 3 integrins in this study, suggesting a structural role for these tyrosine residues in this system. Some of these differences may be due to the different cell types used and potential differences in integrin-associated proteins that may regulate integrin function via binding to this region of the 1 tail. Our data show that the deletion of the individual tyrosine residues is markedly different from deletion of the entire motif. Deletion of the entire NPXY motif may result in important structural changes in the conformation of the cytoplasmic tail of 1 integrin that may not occur when individual tyrosines are mutated. Moreover, NPXY tyrosines could be implicated in complex signaling pathways that regulate adhesion-dependent responses. The substitution of these amino acids may lead to the disruption of a possible negative control pathway for firm adhesion mediated by 1 integrin.
Leukocyte trafficking from blood into peripheral tissues occurs through a sequential process that involves tethering, rolling, firm adhesion, and transendothelial migration. The VLA-4 integrin has been implicated in all of these phases of trafficking. In addition, the functional activity of VLA-4 is dynamically regulated, because circulating lymphocytes maintain their integrins in a nonadhesive state to avoid nonspecific adhesion to blood vessels. When inflammation occurs, VCAM-1 is expressed on endothelial cells in postcapillary venules and activated T lymphocytes, expressing high-affinity integrins, are capable of tethering and arrest. In tissue sites, VLA-4 also mediated adhesion. For example, the adhesion of hemopoietic progenitors in bone marrow is mediated in part by VLA-4 binding to VCAM-1 expressed on bone marrow stromal cells. Our structural analysis in this report demonstrates that distinct regions of the integrin 1 tail regulate VLA-4 function under static and shear flow conditions. These findings have relevance to our understanding of how VLA-4 mediates adhesion in distinct anatomic compartments, such as in the bone marrow or in blood vessels. Furthermore, our work suggests a potential structural basis for modulating VLA-4-mediated adhesive functions in tissue sites while maintaining the ability of VLA-4 to mediate firm arrest and attachment under conditions of vascular shear flow.
Acknowledgments
We thank Dr. Serge Tempesta for statistical analysis and Dr. Marcel Deckert for helpful discussion.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported with grants from the Institut de la Santé et de la Recherche Médicale, the Association Pour la Recherche Contre le Cancer, the Fondation de France, the Fondation pour la Recherche Médicale, the Ligue Contre le Cancer, the Conseil Régional PACA, the Etablissement Fran?ais des Greffes (to A.B.) and with National Institutes of Health Grant A18474 (to Y.S.).
2 Address correspondence and reprint requests to Prof. Alain Bernard, Institut de la Santé et de la Recherche Médicale Unité U576, H?pital de l’Archet 1, Route de St Antoine de Ginestière-BP 3079, 06202 Nice, France. E-mail address: abernard@unice.fr
3 Abbreviation used in this paper: FN, fibronectin.
Received for publication February 12, 2004. Accepted for publication May 13, 2005.
References
Butcher, E. C.. 1991. Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell 67: 1033-1036.
Springer, T. A.. 1995. Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu. Rev. Physiol. 57: 827-872.
Butcher, E. C., L. J. Picker. 1996. Lymphocyte homing and homeostasis. Science 272: 60-66.
Campbell, J. J., E. C. Butcher. 2000. Chemokines in tissue-specific and microenvironment-specific lymphocyte homing. Curr. Opin. Immunol. 12: 336-341.
Lawrence, M. B.. 1999. Selectin-carbohydrate interactions in shear flow. Curr. Opin. Chem. Biol. 3: 659-664
Kansas, G. S.. 1996. Selectins and their ligands: current concepts and controversies. Blood 88: 3259-3287.
Murdoch, C., A. Finn. 2000. Chemokine receptors and their role in inflammation and infectious diseases. Blood 95: 3032-3043
Constantin, G., M. Majeed, C. Giagulli, L. Piccio, J. Y. Kim, E. C. Butcher, C. Laudanna. 2000. Chemokines trigger immediate 2 integrin affinity and mobility changes: differential regulation and roles in lymphocyte arrest under flow. Immunity 13: 759-769.(Maria Alessandra Rosentha)
The VLA-4 integrin supports static cell-cell, cell-matrix adhesion, and dynamic interactions with VCAM-1. Although functions for well-conserved 1 integrin cytoplasmic domains in regulating static cell adhesion has been established, the molecular basis for 1 integrin-dependent arrest on VCAM-1 under flow conditions remains poorly understood. We have transfected the 1 integrin-deficient A1 Jurkat T cell line with 1 cDNA constructs with deletions of the NPXY motifs and specific mutations of tyrosine residues. Deletion of either NPXY motif impaired static adhesion induced by CD2 or CD47 triggering or direct 1 integrin stimulation. In contrast, PMA-induced adhesion to VCAM-1 was unaffected by deletion of the NPIY motif and only slightly impaired by deletion of NPKY. Moreover, deletion of the NPIY motif resulted in enhanced rolling and reduced arrest on VCAM-1 under shear flow conditions. In contrast, deletion of the NPKY motif did not alter arrest under flow. Although tyrosine to phenylalanine substitutions within two NPXY motifs did not alter static adhesion to VCAM-1, these mutations enhanced arrest on VCAM-1 under flow conditions. Furthermore, although deletion of the C'-terminal 5 AA of the 1 cytoplasmic domain dramatically impaired activation-dependent static adhesion, it did not impair arrest on VCAM-1 under flow conditions. Thus, our results demonstrate distinct structural requirements for VLA-4 function under static and shear flow conditions. This may be relevant for VLA-4 activity regulation in different anatomic compartments, such as when circulating cells arrest on inflamed endothelium under shear flow and when resident cells in bone marrow interact with VCAM-1- positive stromal cells.
Introduction
Lymphocyte recruitment into sites of inflammation requires multiple adhesion and signaling interactions with vascular endothelium. These interactions involve functionally specialized macromolecules on the surface of leukocytes and endothelial cells, acting in a multistep pathway (1, 2, 3, 4). L-selectin (CD62L), constitutively expressed by most leukocytes, and P- and E-selectins, up-regulated on activated endothelial cells, mediate the initial leukocyte tethering and rolling along the endothelium (5, 6, 7). This labile adhesion to the vessel walls brings the cells into proximity with chemoattractants displayed on the endothelial surface, which, by coupling to G proteins (8), transduce signals that activate integrin-mediated firm adhesion (9, 10, 11). Several integrins, including VLA-4 (41) and LFA-1 (L2), are involved in firm adhesion. VLA-4 (12), the mucosal homing receptor 47 (13, 14) and, under special circumstances, L2 (15, 16) are also able to mediate rolling.
VLA-4 is a transmembrane 41 heterodimeric receptor that mediates cell-cell and cell-matrix interactions (17, 18). It is widely expressed on mononuclear leukocytes and plays a major role in several cellular functions, including cell spreading and migration and recruitment of leukocyte subsets into inflammatory sites. The VLA-4 ligand VCAM-1 is expressed on vascular endothelium at sites of inflammation and on other selected sites, such as bone marrow stroma, where interaction with VLA-4 mediates stem cell attachment and mobilization (19, 20, 21, 22, 23, 24). VLA-4 mediates static adhesion in the bone marrow hemopoietic compartment by interacting with VCAM-1, which is constitutively expressed on bone marrow stromal cells (25). Moreover, interaction between leukemic cell VLA-4 and stromal fibronectin is crucial for minimal residual disease and acute myeloid leukemia progression (26). Whereas most leukocyte integrins support stationary cell-cell or cell-matrix adhesion, VLA-4 can also mediate versatile dynamic interactions with VCAM-1 under shear flow (12, 14). The interaction of VLA-4 integrin with VCAM-1 can support all of the adhesive steps required for the arrest of lymphocytes on inflamed vascular endothelium (12, 14, 27). Integrins sharing the 4-chain, such as 41 and 47, retain basal recognition of ligand at adhesive contacts under shear stress and can facilitate the spontaneous firm arrest of leukocytes on VCAM-1-expressing endothelium, bypassing the requirement for chemokine stimulation (28, 29, 30). Thus, VLA-4 mediates cell adhesion in distinct environmental conditions.
The molecular basis for support of 1 integrin-dependent rolling and spontaneous arrest on VCAM-1 under flow conditions are poorly understood. In the current study, we used the 1-negative Jurkat T cell line A1 (31) to investigate the role of specific intracytoplasmic motifs of the 1 subunit of the VLA-4 molecule to identify the specific domains critical for both static adhesion and dynamic adhesion under conditions of shear flow. We show that specific intracytoplasmic 1 integrin sequences, including two highly conserved motifs (NPIY and NPKY) and the carboxyl-terminal end of the 1 tail are differentially involved in the regulation of 1 integrin adhesion under dynamic or static conditions. These differences may reflect different roles played by VLA-4, such as in inflammation or in hemopoietic progenitor maturation and mobilization.
Materials and Methods
Reagents and Abs
The CD29 (K20) and CD2 (D66 and X11) mAbs were produced in our laboratory and were described elsewhere (32, 33, 34, 35). 4 and 5 mAbs were purchased from Immunotech; CD47 (B6H12) was produced in our laboratory from the American Type Culture Collection cells (American Type Culture Collection). TS2/16, which stimulates direct 1 integrin activation, was a kind gift from Dr. Sanchez Madrid (Hospital de la Princesa, Madrid, Spain). PMA was purchased from Sigma-Aldrich; VCAM-1, fibronectin (FN),3 and TNF- were obtained from R&D Systems.
Cells
The Jurkat T cell line JE6.1 (American Type Culture Collection) and the 1 integrin-deficient Jurkat T cell line A1 (31) were both cultured in RPMI 1640 (Invitrogen Life Technologies) supplemented with 5% FCS (Invitrogen Life Technologies), 50 U/ml penicillin, 50 μg/ml streptomycin, 2 mM L-glutamine, and 1 mM pyruvate (Invitrogen Life Technologies). The A1 cell line lacks expression of the 1 integrin subunit and is unable to adhere to 1 integrin ligands, including VCAM-1 and FN (31). The transformed human umbilical endothelial cell line EA.hy926 was kindly provided by Dr. Cora Jean Edgell (University of North Carolina, Chapel Hill, NC) and cultured in DMEM (Invitrogen Life Technologies) supplemented with 20% FCS. They were seeded on Lab-Tek chamber slides (Nalge Nunc International), allowed to grow to confluence, and activated with 25 ng/ml TNF- for 24 h before use.
cDNA, transfections, and expression
The wild-type and mutant 1 integrin constructs used in this study (see Table I) have been previously described (31). Areas targeted included tyrosine residues Y783 and Y795, representing potential sites of phos-phorylation, the two NPXY internalization motifs, and the last five amino acids of the 1 cytoplasmic tail. Stable transfections were produced using Effectene reagent (Qiagen) according to the manufacturer’s protocol and cells were selected with 1 g/L geneticin (Invitrogen Life Technologies). Single-color flow cytometry analysis was performed by staining cells with saturating amounts of mouse anti-human CD29 mAb K20, followed by rabbit anti-mouse FITC-conjugated secondary Ab (DakoCytomation). Stained cells were sorted on a FACSVantage SE (BD Biosciences) flow cytometer.
Adhesion experiments
Adhesion assays were performed on 96-well microplates (Costar). Plates were incubated overnight at +4°C with 2.5 μg/ml VCAM-1 or 5 μg/ml FN. Unbound protein binding sites were subsequently blocked with PBS/0.1% BSA for 1 h at room temperature. Cells were labeled with green fluorescent dye CellTracker Green (CMFDA) at a final concentration of 1 μg/ml for 5 x 106 cells for 30 min at 37°C. Cells were then washed and resuspended in HBSS medium complemented with 1 mM CaCl2 and MgCl2 for stimulation with various effectors at 37°C for 20 min. PMA was used at 10 ng/ml and TS2/16, 4B4, CD2 mAb pair D66 and X11, and B6H12 mAbs were used at 10 μg/ml. Cells were allowed to settle in the microplates for 30 min at 37°C. Cell fluorescence was measured by a fluorescence plate reader (Cytofluor; Millipore) before and after washing in warm PBS with 1 mM CaCl2 and MgCl2. Control wells were coated with 0.1% BSA. Specific adhesion was calculated for each well as: (fluorescence before wash – control well fluorescence)/(fluorescence after wash – control well fluorescence) x 100. Data are expressed as mean ± SD of specific adhesion per three replicate wells.
Flow chamber and laminar flow adhesion assay
The flow chamber (Immunetics) used has been previously described elsewhere (36, 37, 38). Briefly, the chamber allows stabilized laminar flow between 0.1 and 2 dyn/cm2. It was mounted on an Axiovert 25 B/W epifluorescence inverted microscope (Zeiss) for direct visualization in real time of dynamic cell adhesion process using a x10 objective. The microscope was coupled to an Axiocam high-resolution numeric camera (Zeiss) to record 10 random fields/coverslip of 1 mm2. The system was directly linked to a Power Macintosh G3 (Apple Computer) with a video card (Micromotion DC30; Pinnacle Systems). Commercial software (Adobe Premiere and Adobe Photoshop; Adobe Systems) was used to count the number of arrested cells and mean values ± SD were calculated per square millimeter.
Cells were recovered in HBSS medium supplemented with 1 mM CaCl2 and MgCl2 and injected through the flow chamber using a withdrawal syringe pump (Harvard Apparatus), at a concentration of 1 x 106/ml, on VCAM-1-coated Lab-Tek chamber slides (Nalge Nunc International). Chamber slides were coated overnight at +4°C with VCAM-1 at 2.5 μg/ml and washed twice in PBS and unbound protein binding sites were blocked with BSA for 1 h at room temperature before their use. For some experiments, cells were injected through the flow chamber on TNF--activated EA cells adherent to Lab-Tek chamber slides.
All flow experiments were performed at 37°C. Temperature was maintained by warming the microscope plate (Warming Plate; Minitub). Each flow experiment has been done at least in triplicate. In rolling experiments, lymphocyte interactions with coated sVCAM-1 in different fields were video recorded for 1 min using the Axiocam video camera. Rolling velocity was measured by determining the number of frames (each 1/24th s length of time) it took 30 cells to cross a 350-μm length field.
Detachment assay
A1-1 control cells and A1-1NPIY or A1-1NPKY cells were stained with fluorescence dyes. Green fluorescent fluorescein diacetate and orange fluorescence tetramethylrodamine were used for control A1-1 and A1-1NPXY, respectively. Cell adhesion was evaluated without stimulation and after a 20-min incubation at 37°C with various stimuli at optimal concentration, namely, PMA (10 ng/ml) and TS2/16 stimulating CD29 mAb at 10 μg/ml as indicated for static adhesion. Cells were mixed in equal amounts at 1 x 106 cells/ml final concentration and allowed to adhere on Lab-Tek chamber slides coated with 2.5 μg/ml VCAM-1 and saturated at 37°C with 0.1% BSA. The strength of adhesion was evaluated by counting residual cells adhering on Lab-Tek slides after washing for 1 min with each increasing shear stress of warm medium (0.5–30 dyn/cm2) on chamber slides. The amount of cells resting adherent on 10 different fields at each shear stress conditions was counted. Comparisons between the A1-1 control cells and the various transfectants were performed using a variance analysis test, with p 0.05 as the statistically significant value.
Fluorescent dyes
For flow experiments, cells were stained with fluorescent dyes purchased from Molecular Probes. Green fluorescent fluorescein diacetate (CellTracker Green CMFDA) is a green tracer with excitation at 522 nm and emission at 529 nm; orange fluorescent tetramethylrhodamine (CellTracker Orange CMTMR) is a red tracer with excitation at 541 nm and emission at 547 nm. Transfected cells were labeled in the culture medium at a concentration of 5 x 106 cells/ml with the different probes (0.5–2 μg/ml) for 30 min at 37°C in the dark, then washed twice before use.
Statistical analysis
Results from static adhesion assays of at least three independent experiments in triplicate wells were pooled and expressed as mean ± SD. Whole data were analyzed by the EpiInfo 6.0 software (Centers for Disease Control and Prevention) using the ANOVA parametric test for variance comparison and by the Kruskal-Wallis test, with p 0.05 as the statistically significant value. For each condition, results of transfectants were compared with control A1-1 cells. For shear flow assays, the number of firm arrested cells, counted in 10 different fields, were recovered from at least three different experiments. Whole results of the firm arrest of A1-1 control cells and the various transfectants were compared using the Kruskal-Wallis test and the ANOVA test for variance comparison, with p 0.05 as the statistically significant value.
Results
Reconstitution of 1 integrin-deficient cell line
The A1 Jurkat T cell derivative lacks expression of the common 1 integrin subunit (31). Therefore, A1 cells do not express either VLA-4 or VLA-5 molecules and exhibit complete loss of adhesion to FN (31). A1 cells were still able to secrete IL-2 in response to classical T cell activation stimuli such as PMA plus CD3 mAb and suboptimal mitogenic concentrations of CD3 mAb did induce Ca2+ release from intracellular stores (data not shown).
We stably transfected A1 cells with the human 1A cDNA, which resulted in re-expression of VLA-4 and VLA-5 on the cell surface (Fig. 1). There was no change in expression of surface molecules capable of activating integrins, such as CD2, CD3, CD28, CD47, or CD99 (data not shown).
Deletion of these motifs also significantly impaired adhesion to both VCAM-1 (Figs. 3E and 4E) and FN (Figs. 3F and 4F) induced by integrin-activating pairs of CD2-specific mAb pairs or a CD47-specific mAb. Thus, our results demonstrate a role for the NPXY motifs in the 1 integrin tail in adhesion induced by the activating 1-specific mAb and inside-out signals mediated by CD2 and CD47. In addition, there is a role for the NPKY motif, but not the NPIY motif, in PMA-induced adhesion to VCAM-1 specifically.
The results obtained in the static adhesion assay on VCAM-1 were confirmed by using the detachment assay (Fig. 5). A1-1NPIY cells (Fig. 5, A and B) and A1-1NPKY cells (Fig. 5, C and D) were stimulated with PMA (Fig. 5, A and C, respectively) and TS2/16 (Fig. 5, B and D, respectively) and allowed to adhere in the Lab-Tek chamber slide. When increasing shear stresses of washing medium were administrated, significant differences (see Fig 5 legend) appeared between the NPXY-deleted cells and their control. The strength of adhesion of A1-1NPIY cells was lower than that of control cells after stimulation with the TS2/16 mAb (Fig. 5B), but not with PMA (Fig. 5A). For the A1-1NPKY cells, the strength of adhesion was lower than that of control cells after stimulation with both PMA (Fig. 5C) and TS2/16 (Fig. 5D).
Deletion of the 1 NPIY intracytoplasmic motif, but not NPKY, affects the firm arrest of T cells on VCAM-1 under flow conditions
Since static adhesion experiments showed that VLA-4 adhesion regulation was dependent on NPXY motifs, we next performed experiments to determine whether firm arrest mediated by VLA-4 was also related to these motifs. Under flow conditions at a shear stress of 1 dyn/cm2 as described in Materials and Methods, the number of A1-1NPIY cells that arrested on coated VCAM-1 was significantly lower then A1-1 control cells. However, no differences were observed at a higher shear stress (2 dyn/cm2) (Fig. 6A). The same significant impairment of A1-1NPIY firm arrest was observed at a very low shear stress of 0.5 dyn/cm2 (data not shown). When cells were stimulated with TS2/16 or CD47 mAbs, an increase of A1-1NPIY was obtained. However, adhesion of A1-1NPIY cells was significantly lower than A1-1 control cells both in basal conditions and after stimulation with TS2/16 and CD47 mAbs. At 2 dyn/cm2, a very moderate increase of firm adhesion was obtained after stimulation with TS2/16 and CD47 mAbs, but not in A1-1NPIY cells (Fig. 6A). A1-1NPIY-deleted cells rolled faster than A1-1 control cells (Fig. 6B). When cells were allowed to roll on TNF--activated human endothelial cells, firm adhesion of A1-1NPIY cells was significantly lower than that of A1-1 cells, both in basal conditions and after stimulation with TS2/16 mAb (Fig. 6C).
In marked contrast to A1-1NPIY-deleted cells, no difference was observed in the firm arrest of A1-1NPKY transfectants on VCAM-1 when compared with A1-1 cells at either shear stress tested (Fig. 6D). These results were confirmed when A1-1NPKY-deleted T cells were injected into the flow chamber on TNF--activated EA cells (data not shown). A significant increase of firm adhesion was obtained after TS2/16 stimulation at both 1 and 2 dyn/cm2 (Fig. 6E). Thus, our results demonstrate a role for the NPIY motif, but not the NPKY motif, in the regulation of T cell adhesion to VCAM-1 under shear flow conditions.
Deletion of the carboxyl-terminal five amino acids totally abolishes static adhesion but allows efficient firm arrest to VCAM-1 under flow conditions
Deletion of the carboxyl-terminal five amino acids of the cytoplasmic domain of the 1 tail has previously been shown to completely inhibit basal and stimulated adhesion of Jurkat T cells to FN (31). Stable A1 transfectants expressing the 1(793) deletion were also unable to adhere to VCAM-1 in a static adhesion assay, when compared with A1-1 cells (Fig. 7, A, C, and E), following either PMA stimulation (Fig. 7A) or direct 1 integrin activation by the mAb TS2/16 (Fig. 7C). Only very weak stimulation with an optimal concentration of PMA and of coated VCAM-1 was detected. Similarly, stimulation of A1-1(793) cells with integrin-activating the CD2 mAb pair or CD47-specific mAb did not enhance adhesion above unstimulated controls, both on VCAM-1 and FN-coated surfaces (Fig. 7, E and F, respectively). Similar results for static adhesion assays were obtained on FN-coated plates (Fig. 7, B, D, and F) in the presence of increasing concentrations of PMA (Fig. 7B), TS2/16 (Fig. 7D), or activating 1 integrin costimuli (Fig. 7F).
Although the 1(793) deletion had a dramatic effect on static T cell adhesion to both VCAM-1 and FN, A1-1(793) were able to firmly arrest on VCAM-1 under shear flow conditions (Fig. 7G). When whole data from at least three different experiments performed were statistically analyzed, the number of arrested cells was comparable between A1-1(793) cells and control A1-1 cells at both flow rates tested. The same pattern was observed when cells were stimulated with TS2/16 at 10 μg/ml (data not shown).
Mutation of tyrosine residues of the NPIY and NPKY motifs increases the firm arrest of T cells on VCAM-1 under flow conditions, but has no effect in static adhesion
The tyrosine residues in the NPIY(Y783) and NPKY(Y795) motifs have been implicated in downstream signaling activities and integrin functions, such as cell spreading (45), tissue invasion (46), and bacterial internalization (47). To elucidate the importance of these motifs in adhesion, transfectants were generated expressing mutant 1 subunits with tyrosine to phenylalanine substitution in each NPXY motif (A1-Y783F and A1-Y795F) or in both motifs (A1-Y783,795F).
In a static adhesion assay, A1-1 Y783F, A1-1 Y795F, and A1-1 Y783,795F cells all supported adhesion to the same extent as A1-1 wild-type cells. Similar levels of adhesion were observed under all activating conditions tested (data not shown). This was in agreement with data obtained previously with transient transfectants on FN-coated microwells (31). However, in shear flow assays, both A1-Y783F (Fig. 8A) and A1- Y795F (Fig. 8B) cells exhibited statistically significant higher levels of arrest than wild-type A1-1 at high shear stress (2 dyn/cm2). Furthermore, A1 cells expressing the double tyrosine mutant Y783,795F exhibited higher levels of arrest at both low and high shear rates (Fig. 8C). The same pattern of firm adhesion increase was observed when tyrosine-mutated cells were stimulated with TS2/16 mAb, in comparison to basal conditions. In particular, A1-1 Y783F and A1-1 Y795F firm arrest was significantly higher than in A1-1 control cells at 2 dyn/cm2. Moreover, firm arrest of A1-1 Y783,795F cells was significantly higher than A1-1 cells at both 1- and 2-dyn/cm2 shear stress (data not shown).
Discussion
We used the 1 integrin-deficient A1 Jurkat T cell line to define the structural requirements in the 1 integrin cytoplasmic domain that regulate VLA-4-mediated adhesion under both static conditions and in the presence of shear flow. The A1 cell line represents a unique cellular reagent for these studies, since the lack of endogenous 1 integrin expression in this cell line allowed us to stably express wild-type and mutant 1 integrin subunits without the complications of endogenous wild-type 1 integrin expression. Our studies reveal that integrin function under static and shear flow conditions are differentially regulated by three regions of the 1 integrin tail (Table II): 1) the membrane-proximal NPIY motif regulates VLA-4-mediated adhesion under both static and shear conditions, 2) the carboxyl-terminal region of the 1 tail has a specific role in regulating static adhesion, but not arrest under shear flow, and 3) the tyrosines in the NPXY motifs do not play a major role in regulating static adhesion but appear to negatively regulate adhesion under shear flow.
The 1 integrin NPIY motif appears to be particularly critical for optimal VLA-4 function, since deletion of this motif inhibits both activation-dependent VLA-4-mediated adhesion under static conditions and firm arrest under flow conditions. The shear flow rates used in our experiments likely mimic physiological inflammatory conditions in microvessels, when microangectasias induced by inflammation slow the shear stress to these rates (48).
It is interesting to note that deletion of the NPIY motif altered firm arrest at 1 dyn/cm2, but not at 2 dyn/cm2. This was also observed in a more physiological model, on human activated endothelial EA cells. This may be related to a potential role of the NPIY motif in regulating affinity states of VLA-4, since Chen et al. (28) demonstrated that low-affinity VLA-4 states preferentially mediate transient tethering and rolling of PBL on VCAM-1, whereas high-affinity VLA-4 states preferentially support spontaneous firm arrest of both PBL and Jurkat T cells on VCAM-1. It is possible that cells arresting at higher shear stress conditions (2 dyn/cm2) is mediated by high-affinity VLA-4, whereas cell arrest at 1 dyn/cm2 is mediated predominately by intermediate-/low-affinity VLA-4. The reduced firm adhesion of the A1-1NPIY cells was also confirmed on VCAM-1-coated Lab-Tek slides at very low shear stress, such as 0.5 dyn/cm2. This suggests that the NPIY motif might be particularly critical in regulating intermediate-/low-affinity states of VLA-4 rather than high-affinity states.
The NPIY motif has also been implicated in the binding of the cytoskeleton protein talin to the 1 cytoplasmic domain (49), and talin interaction with integrin tails is proposed to play a critical role in integrin activation (42, 50). Thus, the inability of the A1-1NPIY cells to adhere to VCAM-1 following integrin-activating signals may be due to disruption of the interaction between talin and the integrin 1 subunit. In addition, our results with adhesion under shear flow suggest a role for talin in optimal firm arrest, but only under low shear flow rates. Integrin activation by TS2/16 or cosignal stimulation by CD47 mAb was not affected in A1-1NPIY cells, since an increase in firm arrest was observed at levels comparable to those of A1-1 control cells.
The C-terminal region of the integrin 1 tail that contains the NPKY motif was found to be particularly important for static adhesion but not for adhesion under shear flow. This was most evident with the A1-1(793) cells, which exhibited defective static adhesion to both VCAM-1 and FN under static adhesion conditions, but had surprisingly normal patterns of cell arrest under shear flow conditions. In addition, like A1-1NPIY, A1-1NPKY cells showed reduced adhesion after stimulation with the integrin-activating mAb TS2/16 and complete absence of response to other inside-out stimuli that activate 1 integrins. The impairment in the adhesion properties of the NPXY-deleted cells seen in both static and detachment assays was statistically significant, although small in some cases, rather than an "all or none" effect. The existence of different pools among Jurkat T cells of cells bearing different states of VLA-4 affinity for VCAM-1 could explain these smaller differences observed with VCAM-1 compared with the effects observed on FN, where the affinity states of VLA-4 may be more homogeneous.
Like A1-1(793) cells, arrest of cells expressing the A1-1NPKY under shear flow conditions was comparable to A1 cells expressing wild-type 1. Our results also show that TS2/16 stimulation is affected following deletion of the NPXY motifs or deletion of the five C-terminal amino acids of the 1 integrin cytoplasmic tail. Although TS2/16 is thought to activate 1 integrin primarily via alterations in the 1 extracellular domain, recent studies have shown that TS2/16-induced adhesion is inhibited when Rap1 signaling is impaired (51). Thus, our results are consistent with a potential requirement for intracellular signaling in regulating integrin activation induced by the TS2/16 mAb.
In other studies, the carboxyl-terminal amino acids of the 1 tail were found to be critical for 1 integrin-dependent adhesion, as well as migration and metastasis. In an in vivo model, Stroeken et al. (46) showed that expression of a 1(793) construct in 1 integrin-deficient double knockout ESb lymphoma cells (ESb-DKO) impaired invasion and metastasis in vivo when injected in mice and greatly reduced adhesion in vitro. Since firm cell arrest on endothelium is critical for invasion in vivo, our results suggest that the carboxyl-terminal end of the 1 integrin tail may be particularly critical for regulating integrin function subsequent to firm arrest, such as transendothelial migration.
The mechanism by which the carboxyl-terminal end of the 1 integrin tail reduces integrin function remains unclear. Studies with chimeric integrins expressed in Chinese HO cells (45) suggest that the 1(793) mutant is still capable of interacting with talin. Our results suggest that structural integrity at this region of the 1 tail is particularly critical for static adhesion regulated by inside-out signaling, but not firm adhesion under hydrodynamic shear flow conditions.
The third structural feature of the integrin 1 tail is that tyrosine residues 783 and 795 in the NPXY motifs appear to play a critical role in regulating adhesion under shear flow, but not under static conditions. When either of these tyrosine residues is substituted with phenylalanine, adhesion at high shear stress rates was actually enhanced when compared with wild-type A1-1 cells. In contrast, adhesion under static conditions was unaffected by these mutations, consistent with previous results (31). Similar results were observed when both tyrosines were substituted with phenylalanine, although now enhanced firm arrest was also observed at low shear rates.
Although the role of tyrosine phosphorylation of the 1 tail remains unresolved, our results nevertheless suggest a potential function for phosphorylation of these tyrosine residues in regulating adhesion under shear flow. A role for these tyrosines in migration has also been suggested, since a tyrosine to phenylalanine substitution at Y795 impairs the migration of mouse GD25 fibroblasts (52). Furthermore, the Y783/795F double mutation markedly affects cell migration and cytoskeletal architecture, but not adhesion (52). However, when expressed in 1-deficient lymphoid ESb-DKO cells, the Y783/795F double mutation impaired adhesion and migration in vitro, but not in in vivo invasion in the liver (46). Other studies have also implicated these tyrosine residues in integrin affinity modulation (40), as tyrosine to alanine substitutions at these sites impairs LIBS mAb binding. However, tyrosine to phenylalanine substitutions did not affect binding of this mAb to 1 and 3 integrins in this study, suggesting a structural role for these tyrosine residues in this system. Some of these differences may be due to the different cell types used and potential differences in integrin-associated proteins that may regulate integrin function via binding to this region of the 1 tail. Our data show that the deletion of the individual tyrosine residues is markedly different from deletion of the entire motif. Deletion of the entire NPXY motif may result in important structural changes in the conformation of the cytoplasmic tail of 1 integrin that may not occur when individual tyrosines are mutated. Moreover, NPXY tyrosines could be implicated in complex signaling pathways that regulate adhesion-dependent responses. The substitution of these amino acids may lead to the disruption of a possible negative control pathway for firm adhesion mediated by 1 integrin.
Leukocyte trafficking from blood into peripheral tissues occurs through a sequential process that involves tethering, rolling, firm adhesion, and transendothelial migration. The VLA-4 integrin has been implicated in all of these phases of trafficking. In addition, the functional activity of VLA-4 is dynamically regulated, because circulating lymphocytes maintain their integrins in a nonadhesive state to avoid nonspecific adhesion to blood vessels. When inflammation occurs, VCAM-1 is expressed on endothelial cells in postcapillary venules and activated T lymphocytes, expressing high-affinity integrins, are capable of tethering and arrest. In tissue sites, VLA-4 also mediated adhesion. For example, the adhesion of hemopoietic progenitors in bone marrow is mediated in part by VLA-4 binding to VCAM-1 expressed on bone marrow stromal cells. Our structural analysis in this report demonstrates that distinct regions of the integrin 1 tail regulate VLA-4 function under static and shear flow conditions. These findings have relevance to our understanding of how VLA-4 mediates adhesion in distinct anatomic compartments, such as in the bone marrow or in blood vessels. Furthermore, our work suggests a potential structural basis for modulating VLA-4-mediated adhesive functions in tissue sites while maintaining the ability of VLA-4 to mediate firm arrest and attachment under conditions of vascular shear flow.
Acknowledgments
We thank Dr. Serge Tempesta for statistical analysis and Dr. Marcel Deckert for helpful discussion.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported with grants from the Institut de la Santé et de la Recherche Médicale, the Association Pour la Recherche Contre le Cancer, the Fondation de France, the Fondation pour la Recherche Médicale, the Ligue Contre le Cancer, the Conseil Régional PACA, the Etablissement Fran?ais des Greffes (to A.B.) and with National Institutes of Health Grant A18474 (to Y.S.).
2 Address correspondence and reprint requests to Prof. Alain Bernard, Institut de la Santé et de la Recherche Médicale Unité U576, H?pital de l’Archet 1, Route de St Antoine de Ginestière-BP 3079, 06202 Nice, France. E-mail address: abernard@unice.fr
3 Abbreviation used in this paper: FN, fibronectin.
Received for publication February 12, 2004. Accepted for publication May 13, 2005.
References
Butcher, E. C.. 1991. Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell 67: 1033-1036.
Springer, T. A.. 1995. Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu. Rev. Physiol. 57: 827-872.
Butcher, E. C., L. J. Picker. 1996. Lymphocyte homing and homeostasis. Science 272: 60-66.
Campbell, J. J., E. C. Butcher. 2000. Chemokines in tissue-specific and microenvironment-specific lymphocyte homing. Curr. Opin. Immunol. 12: 336-341.
Lawrence, M. B.. 1999. Selectin-carbohydrate interactions in shear flow. Curr. Opin. Chem. Biol. 3: 659-664
Kansas, G. S.. 1996. Selectins and their ligands: current concepts and controversies. Blood 88: 3259-3287.
Murdoch, C., A. Finn. 2000. Chemokine receptors and their role in inflammation and infectious diseases. Blood 95: 3032-3043
Constantin, G., M. Majeed, C. Giagulli, L. Piccio, J. Y. Kim, E. C. Butcher, C. Laudanna. 2000. Chemokines trigger immediate 2 integrin affinity and mobility changes: differential regulation and roles in lymphocyte arrest under flow. Immunity 13: 759-769.(Maria Alessandra Rosentha)