Extracellular Matrix Proteins, Regulators of T-Cell ... - NCBI - NIH

Download PDF
7 downloads 0 Views 186KB Size Report
Comment
Antigen-dependent T-cell activation requires engagement of the T-cell receptor (TCR)-CD3 complex with peptides pre- sented by major histocompatibility ...
CLINICAL AND DIAGNOSTIC LABORATORY IMMUNOLOGY, Nov. 1995, p. 646–651 1071-412X/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 2, No. 6

Extracellular Matrix Proteins, Regulators of T-Cell Functions in Healthy and Diseased Individuals A. GORSKI1*

AND

J. W. KUPIEC-WEGLINSKI2

Department of Immunology, Transplantation Institute, Warsaw Medical School, 02006 Warsaw 22, Poland,1 and Surgical Research Laboratories, Harvard Medical School, Boston, Massachusetts 022152 Antigen-dependent T-cell activation requires engagement of the T-cell receptor (TCR)-CD3 complex with peptides presented by major histocompatibility complex molecules. However, stimulation of T cells through the complex of T-cell antigen receptor and CD3 in the absence of other signals is insufficient for optimal interleukin-2 (IL-2) production and proliferation and induces T-cell anergy. Therefore, additional costimulatory signals are needed for an effective T-cell-mediated immune response. Such signals can be provided by various interactions between T cells and ligands on antigen-presenting cells (e.g., CD2-CD58 [LFA-3] and CD28-CD80 [B7] cells [2, 59]). Furthermore, recent data indicate that the extracellular matrix (ECM) proteins can mediate strong costimulation in the process of T-cell activation in the absence of monocytes, so one could view the ECM protein as an equivalent of an antigen-presenting cell. This view is further strengthened by observations that ECM proteins can interact with both an integrin and other T-cell-activating agents (e.g., cytokines) (26, 29). Lymphocyte extravasation into the lymphatic tissue constitutes a normal physiological process by which the cells return finally to blood through the thoracic duct. This migration is associated with lymphocyte interaction with the ECM proteins present in the basement membranes (collagen type IV [C-IV] and laminin), because once the cells reach the underlying tissue they come into contact with interstitial ECM proteins (collagen type I [C-I] and fibronectin [FN]). The latter interactions appear to be especially relevant at sites of tissue invasion by inflammatory cells, because both C-I and FN are ubiquitous in interstitial ECM proteins. In addition, FN has been reported to be produced by mononuclear cells themselves (monocytes and T cells), and its increased concentrations are found at the foci of inflammation and in sites of delayed-type hypersensitivity (34, 58). Therefore, it could be expected that the ECM proteins may regulate T-cell functions under both physiological and pathological conditions. (The biochemical structures of major ECM proteins were recently reviewed [48].) Evidence that integrins mediate information transfer into T cells is accumulating: occupation of integrin receptors by their ligands leads to tyrosine phosphorylation, elevation of cytoplasmic pH and Ca21 concentration, activation of protein kinases, and inositol release. Moreover, FN binding via VLA-5 induces the AP-1 transcription factor necessary for IL-2 transcription. Integrin receptors, by recognizing ECM proteins, can provide the signal for the stimulation of cytokine genes and subsequent cytokine production by T cells (this was demonstrated for IL-1, IL-2, and IL-4 and tumor necrosis factor alpha). Recent minireviews have provided a detailed analysis of ‘‘inside-to-out’’ and ‘‘outside-to-in’’ signaling by integrins (10, 29, 32, 48).

Interestingly, recent data suggest that ECM protein-mediated cytokine gene activation can also occur in the absence of TCR engagement. This pathway of antigen-independent cytokine production by T cells interacting with the ECM proteins could be particularly important for gd T cells, frequently identified at sites of inflammation and thought to play an immunoregulatory rather than an effector role in the immune response (56). Few studies have investigated the functions of ECM proteins and integrins involved in T-cell development. These proteins are present in the thymic parenchyma and basement membranes and may be found to be associated with the epithelial cells which secrete them. Up to 15% of unseparated thymocytes adhere to FN, but 70% of the CD42 CD82 (double negative) cells do. In addition, thymocytes adhere to thymic stromal cells interacting with the FN present on their surfaces. Blocking of this adhesion (with anti-FN antibody and antiVLA-4 and anti-VLA-5 reagents) causes a potent inhibition of thymocyte differentiation in vitro. Thus, thymocyte-FN interactions may be critical in inducing and supporting the differentiation of immature T cells. In addition, another integrin (a6b4), with unknown ligand specificity, may play a role in intrathymic T-cell differentiation (63, 64). Recently, it was shown that human thymocyte proliferation induced by CD3 can be enhanced by ECM proteins in the presence of an additional signal: IL-2 or phorbol myristate acetate (61). COLLAGEN TYPE I No significant adhesion of freshly isolated T cells to C-I protein has been demonstrated by standard assays with C-Icoated culture plates. In fact, C-I is frequently used as a negative control in adhesion assays (54). In contrast, Sundqvist and colleagues (4, 57) reported that such cells can adhere to two-dimensional C-I substrata; however, very high concentrations of the protein (0.5 to 2.0 mg/ml versus the 1 to 10 mg/ml that is routinely used) had to be applied in order to promote T-cell attachment. Cell attachment was almost completely abolished by anti-b1 monoclonal antibody (MAb), indicating that the collagen receptor involved is a member of the integrin family (4, 57). When C-I is coimmobilized with anti-CD3 MAb, C-I causes a costimulatory effect demonstrable by T-cell proliferation (11, 40), although some investigators were unable to demonstrate such activity of the protein, which probably reflects methodological differences (12, 53). Furthermore, twoand three-dimensional collagen substrata augment DNA synthesis of preactivated T cells (57). In our hands, C-I is costimulatory in both serum-free and serum-containing media (19), thus confirming the data of Dang et al. (11). Those investigators showed that the costimulatory effect of C-I can be abolished by MAbs to CD49c (VLA-3) and CD26, thus suggesting that both integrin and nonintegrin surface receptors could be involved. Both receptors are detectable on freshly isolated T

* Corresponding author. Mailing address: Transplantation Institute, Warsaw Medical School, 02006 Warsaw 22, Poland. 646

VOL. 2, 1995

MINIREVIEWS

cells, of which approximately 10% can be VLA-31 and 40% can be CD261 (T-cell CD26 expression is heterogeneous and has high and intermediate levels of expression) (41, 52). In contrast to ab TCR T cells that are VLA-41 and VLA-51 and that express VLA-3, gd TCR T cells lack VLA-3 but express VLA-4 and VLA-5. However, those T cells increase their adhesion to FN upon activation and respond to FN-mediated costimulation (6). Although VLA-1, VLA-2, and VLA-3 are thought to be collagen-binding receptors, only VLA-2 is the integrin used as a collagen receptor by mononuclear cells (monocytes and cultured T cells) (18). Thus, T cells may use different integrin and nonintegrin receptors for adhesive interactions with C-I. COLLAGEN TYPE IV The interactions of T cells with C-IV are of special interest, since this type of collagen is localized solely within the basement membranes. Cultured T cells attach to C-IV by using VLA-2 in a manner similar to their attachment to C-I and FN (18). Likewise, C-IV enhances T-cell reactivity to OKT3 in a process that is sensitive to anti-VLA-3-mediated inhibition (11, 40). Our data indicate that in healthy individuals the costimulatory strength of C-IV is inferior to that of C-I (21). In contrast, activated T cells adhere to C-IV (but only very weakly to C-I [21, 22]), which resembles the adherence of murine T-cell lines that preferentially attach to C-IV (55). In our hands, T-cell attachment to C-IV is mostly CD26 mediated (33). These findings highlight the significance of different lymphocyte ECM receptors in mediating the reactivities of T cells with the ECM proteins. FIBRONECTIN When FN is coimmobilized with OKT3, FN costimulates vigorous T-cell proliferation which is higher than the response to C-I and C-IV (12, 21, 40, 53). Antibodies reactive with a common b1 subunit (CD29) of the VLA family of antigens could act synergistically with CD3 but not CD2, inducing T-cell proliferation (69). MAb and RGD peptide-blocking studies have revealed that CD49e (VLA-5), a classical FN receptor, is responsible for this effect (40), although others suggest the participation of CD49d (VLA-4) as well (12, 53). Whether the costimulatory effect provided by the VLA-5–FN interaction must be delivered in close proximity to the CD3-TCR crosslinking signal is a matter of some controversy, because studies involving OKT3 immobilized on beads and FN on plastic have yielded conflicting results (12, 53). Although cytokines in combination with immobilized OKT3 do not facilitate T-cell proliferation or can be only weakly costimulatory (53), they may be secreted as a result of mononuclear cell-ECM protein interactions (26). Furthermore, the level of tumor necrosis factor alpha secretion is greatly enhanced in the presence of physically damaged vascular ECM protein, which appears to transmit signals of tissue damage to the surrounding mononuclear cells (25). On the other hand, tumor necrosis factor alpha binds to FN by using the N-terminal domain of the ECM glycoprotein and induces an augmented adhesiveness of activated (but not of resting) T cells to that protein. FN-associated cytokine could serve as an enhancer of integrin function during postadhesion events (3). Those phenomena could play a role in the recruitment of additional immune cells to the inflammatory site, which may be important in the immunopathology of vasculitis. This assumption is strengthened by observations indicating that FN amplifies cytotoxic T-cell activation and augments their level of degranulation (45).

647

The contribution of different T-cell subsets in the costimulatory reaction mediated by FN is also debatable. It has been suggested that FN only enhances proliferation of the memory subset of CD41 cells (CD45RO1) (40); however, other investigators reported the responses of both memory and naive (CD45RA1) subsets of CD41 and CD81 cells (12), which is in accord with our own data (unpublished data). The response of the memory subset of CD41 cells is enhanced by lower FN concentrations, a finding in accord with a well-known hypoexpression of integrin molecules on naive cells compared with that on memory cells (52). While the costimulative response of T cells to FN is predominantly VLA-5 dependent, at least three different integrin receptors are involved in T-cell adhesion to that protein. Again, VLA-5 appears to be the major receptor on the surface of T cells for the RGD sequence within the FN molecule. In addition, VLA-4 reacts with the EILD VPST sequence of the alternatively spliced connecting segment 1 (CS1) (52). VLA-4 binding sites for FN and for VCAM-1 are overlapping, but the mechanisms of its interaction with those two different ligands appear to be different (39). Yet another potential integrin receptor on T cells that appears to be involved in cell attachment to FN is a4b7, which is weakly expressed by resting peripheral blood T cells (where it defines a CD45RO1 CD41 T-cell subset) but is induced upon activation (37). Interestingly, on B cells downregulation of a4b7 is observed upon activation (47). FN is present in tissues in a form of disulfide-cross-linked fibrils, in contrast to the form in which it is present in blood, in which it is dimeric. In vitro conversion of FN into a multimeric form (super FN) greatly enhances its adhesive properties, which are mediated both by integrin and by nonintegrin lymphocyte receptors. Since super FN resembles the FN matrix, the interactions of T cells with super FN may more accurately reflect the situation in vivo (42). Although resting T cells express at least three receptors known to bind to FN, their adhesion to that protein is only minimal, but it can rapidly (within minutes) be upregulated by phorbol esters (e.g., phorbol myristate acetate) or cross-linking with the CD3 molecule. Those agents are believed to induce a high-affinity state of integrins by virtue of still unclear qualitative changes within their molecules (14, 32). It was also demonstrated that activated T cells express FN at their surfaces, which probably reflects both increased uptake and increased production of FN by lymphocytes (23). Furthermore, T cells (but not monocytes) can degrade FN (7). The functional significance of surface-associated FN is unclear; it is possible that it may be of importance for cell mobility in tissues. In addition, binding of soluble FN to lymphocytes may also inhibit adhesion to the FN matrix (23). This confirms the general observation that T-cell activation resulting from integrin engagement by ECM proteins can occur only when they interact with T cells as insoluble (immobilized) ligands. In contrast, soluble ECM ligands (especially FN) are not costimulatory and can inhibit T-cell activation (35, 44). ELASTIN So far, no data on T-cell interactions with elastin have been reported. Our unpublished observations could be summarized as follows. (i) Resting T cells do not adhere to elastin, but adhesion can be induced upon short-term activation (30 min) with phorbol esters or phytohemagglutinin. Adhesion seems to be mediated by at least two surface receptors: the 67-kDa elastin-binding protein, already described for monocytes and other cell types (27), and an integrin (VLA-3?). The latter finding is rather surprising, because no RGD sequence was

648

MINIREVIEWS

found within the elastin molecule. (ii) Like other ECM proteins, elastin can provide costimulatory signals during T-cell activation via CD3. (iii) There is individual variation in the response to elastin. Both adhesion to and costimulation by elastin are statistically higher in young males than in young females. This difference is not observed in aged persons, who generally have high responses to elastin. (iv) Patients with advanced atherosclerosis with its overt clinical manifestations have high T-cell costimulative responses to elastin. In some cases these responses are highly specific for elastin (these responses are not seen for C-I, C-IV, and FN) (19a). We believe that the preliminary observations described above are of great potential importance for our understanding of the immunobiology of atherosclerosis, suggesting that T-cell interactions with elastin may play a major role in its pathogenesis. OTHER ECM PROTEINS VLA-6 is a member of the integrin family recognizing laminin, a basement membrane protein (52). The functions of laminin in the process of T-cell costimulation are unclear: both enhancing and inhibitory effects were reported (12, 38, 53); in addition, some investigators were unable to demonstrate any significant activity of laminin during this process (40). In contrast, T-cell adhesion to laminin is clearly VLA-6 dependent (52), while the role of the nonintegrin 67-kDa receptor remains to be determined (27, 50). Therefore, it appears that of the basement membrane proteins, only C-IV appears to have a definite activity in the regulation of T-cell proliferative responses. Another ECM protein that could play a regulatory role in mediating T-cell functions is tenascin (TN). It is found in low or undetectable levels in adult tissues (tissues that express TN in the embryo or neonate). In contrast, TN is present in the matrices of tumors and appears in large amounts in the extracellular spaces of developing and regenerating tissues (as a result of wound healing, for example) (13). Recently, TN has been shown to downregulate FN-mediated costimulation of T-cell proliferation triggered by OKT3 (24). This confirms earlier reports indicating that TN can suppress antigen- and alloantigen-specific T-cell responses (51). Thus, the action of an inducible protein like TN may be relevant in locations where it appears. For example, elaboration of TN into the matrices of solid tumors could inhibit tumor-specific T-cell responses. TN released at the site of an allograft could downregulate the action of locally present FN and therefore attenuate graft rejection. Thrombospondin, an ECM protein that is transiently expressed at high concentrations in damaged and inflamed tissues, may be another ligand for T cells. At least three receptors expressed on the surface of T cells can bind thrombospondin, of which two belong to the b1 integrin family (VLA-4 and VLA-5). Memory T cells show preferential adhesion to thrombospondin which is similar to that to the other ECM proteins recognized by those cells (36). ABNORMALITIES OF T-CELL–ECM PROTEIN INTERACTIONS AND DISEASES The fascinating connections between the biology of the ECM protein and its ability to actively influence the key steps in T-cell activation highlight the possible immunopathological role of disturbed T-cell–ECM interactions in diseases (Table 1). Vasculitis. (i) Rheumatoid arthritis. Garcia-Vicuna et al.

CLIN. DIAGN. LAB. IMMUNOL. TABLE 1. T-cell receptors mediating adhesion to and costimulation by ECM proteins Receptor mediatinga: ECM protein Adhesion

C-I C-IV FN Laminin Thrombospondin Elastin Tenascin

VLA-1 (CD49a) VLA-2 (CD49b) VLA-1 (CD49a) VLA-2 (CD49b) CD26 VLA-4 (CD49d) VLA-5 (CD49e) CD26(6) VLA-6 (CD49f) VLA-4 (CD49d) VLA-5 (CD49e) 67kD EBP VLA-3 (CD49c)(?) —

Costimulation

VLA-3 (CD49c) CD26 VLA-3 (CD49c) CD26 VLA-5 (CD49e) VLA-4 (CD49d)(6) VLA-6 (CD49f) — VLA-3 (CD49c)(?) VLA-4 (CD49d)2(?)

Symbols: 6, weak reactivity; ?, preliminary data; 2, inhibition of response; —, no data available. a

(15) have demonstrated that in patients with rheumatoid arthritis a significantly higher proportion of synovial fluid T cells than peripheral blood T cells was able to bind to FN. This enhanced attachment of synovial fluid T cells was independent of the level of surface expression of VLA-5. Similar data were reported by Rodriguez et al. (49). In addition, those investigators (49) described the upregulation of VLA-4 (CD49d) and b1 (CD29) on synovial fluid T cells. In contrast, Walle et al. (66) have shown higher levels of expression of both VLA-1 (CD49a) and VLA-5 (CD49e) on synovial fluid T cells. Recently, a novel integrin termed a4b7 has been implicated in synovial fluid infiltration and destruction. In contrast to ,5% in synovial fluid T cells and 9% in peripheral blood T cells, it was shown that the majority of synovial membrane T cells express high-density a4b7 (37). We have demonstrated heightened costimulatory responses of peripheral blood T cells of patients with rheumatoid arthritis to C-I, C-IV, and FN; interestingly, those abnormalities were found in patients with severe vasculitis with necrotizing lesions (22). In contrast to enhanced ECM protein-mediated costimulation, direct T-cell responses to ECM proteins were very low or absent. This suggests that upregulated T-lymphocyte integrin expression and function may be a more important phenomenon in the development and perpetuation of rheumatoid arthritis. (ii) Systemic lupus erythematosus. Takeuchi et al. (60) have found increased levels of VLA-4 expression on peripheral blood lymphocytes from patients with systemic lupus erythematosus who had signs of vasculitis. In addition, adhesion of those patients’ T cells to FN was also increased. (iii) Wegener’s granulomatosis. Dramatic costimulation by FN of peripheral blood T cells was found in a patient with Wegener’s granulomatosis. In addition, the patient’s T cells had much greater adhesiveness to C-IV, while their attachment to resting and inflamed endothelium was normal (22). (iv) Takayasu arteritis. In two patients with Takayasu arteritis studied so far, a lack of costimulative responses to ECM proteins was demonstrated. In addition, T-cell adhesion to C-IV, FN, and resting endothelium was also elevated. Those abnormalities could be at least partly corrected by immunosuppressive therapy (30). Skin diseases. T cells from patients with psoriasis have diminished responses to C-I, but their reactivities to C-IV may be

VOL. 2, 1995

normal or increased. In fact, there was about a twofold increase in the level of C-IV-induced costimulation of CD3 responses in patients with active psoriatic lesions that involved more than 40% of the body surface. In patients with widespread skin lesions, the costimulation of CD3 responses by FN was about threefold higher than that in patients with guttate psoriasis (17). Similar hyperactivity to C-IV-mediated signals was found in patients with atopic dermatitis with severe skin lesions (16). Thus, T-cell homing to skin and the transmigration of T cells through basement membranes may be associated with enhanced T-cell–C-IV interactions. Graves’ disease. The importance of enhanced levels of expression of T-cell integrins in the immunopathology of autoimmune diseases is also highlighted in patients with Graves’ disease, in which the levels of expression of VLA-4 and VLA-5 on CD41 cells in the thyroid gland were markedly higher than those in peripheral blood. No such differences were demonstrable on peripheral blood T-cell subsets. Those findings suggest that increased levels of expression of adhesion molecules on CD41 cells may be responsible for the migration of these cells into thyroid glands and cellular interactions between these cells and thyroid epithelial cells (31). Chronic hepatitis and cirrhosis. Peripheral blood T cells from patients with chronic hepatitis and cirrhosis show enhanced responsiveness to C-I, C-IV, and FN. Furthermore, when the T-cell responses of all patient groups were compared, reactivity to C-I was a discriminating factor between patients with chronic hepatitis and cirrhosis versus patients with acute hepatitis, while other parameters (response to OKT3 alone as well as costimulation by C-IV and FN) were comparable in all groups. This suggests that the enhanced sensitivity of patients’ T cells to C-I-mediated costimulation might be at least partially responsible for the progression to chronic hepatitis and cirrhosis (43). Interestingly, Torimoto et al. (62) have shown that most T cells infiltrating into the livers of patients with chronic hepatitis and cirrhosis express CD26, a functional collagen receptor responsible for costimulation of T cells by that protein. Sarcoidosis. In 14 patients with sarcoidosis an identical pattern of abnormality could be demonstrated: the patients’ T cells did not respond to costimulative signals. In contrast, ECM proteins caused inhibition of CD3-triggered activation (unpublished data). Allograft immunity. T cells from highly sensitized patients awaiting transplantation, who have increased numbers of circulating memory cells, respond poorly to ECM proteins (46). In contrast, acute kidney allograft rejection was demonstrated to be associated with enhanced peripheral blood T-cell proliferative responses to ECM protein-mediated signals. However, activated T-cell adhesion to FN and C-IV in patients with allograft immunity were lower than those in patients with stable graft function. Chronic rejection was associated with normal levels of costimulation but low levels of adhesion. The levels of circulating CD45RO1 (memory) T cells as well as those of CD291 and VLA-21 T cells were also decreased in some patients. A mirror-like phenotype was observed in T cells isolated from rejected allografts: the majority of graft-infiltrating T cells were CD45RO1, CD291, VLA-21, and CD691 cells. Adhesion assays revealed significant levels of graft-infiltrating T-cell adhesion to C-IV and FN and very high levels of adhesion to cultured endothelium. In addition, there was enhanced costimulation by C-IV, especially in patients with intense mononuclear cell infiltration of an allograft (33). Those findings strongly suggest that local interactions of memory T cells with the ECM proteins may be of considerable importance for rejection. In addition, they suggest that T-cell inter-

MINIREVIEWS

649

actions with the C-IV of the basement membrane may not only play a guiding role for T cells migrating to the allograft but may also upregulate their antigraft reactivities in situ, as suggested by Coito et al. (9). Cancer. Peripheral blood T cells from cancer patients and tumor-infiltrating cells respond poorly or are resistant to costimulative signals mediated by ECM proteins. Furthermore, the adhesive properties of those patients’ T cells are depressed. Moreover, tumor-infiltrating cells show a total lack of adhesiveness to C-IV and FN. Those severe functional deficits of T-cell–ECM interactions could not be explained by phenotypic deficiencies of T-cell integrin and nonintegrin receptors for ECM proteins. Those novel findings point to the possible role of deficient T-cell–ECM interactions in the disturbances of immune surveillance in patients with cancer (20). Human immunodeficiency virus infection. After human immunodeficiency virus infection, both chronically infected T-cell lines and acutely infected peripheral blood T cells acquire the ability to adhere to FN and C-IV but not to C-I. This is associated with the upregulation of b1 (CD29) and VLA-5 (CD49e) expression. Thus, enhanced levels of expression of integrins and upregulated adhesion may cause T-cell adherence to a variety of tissues, with increased infiltration of T cells into these tissues and the release of viral gene products, which may lead to progression to a fully blown syndrome (Table 2) (68). CONCLUSIONS The data presented in this minireview strongly suggest not only that ECM proteins regulate normal T-cell functions but also that the disturbances of those T-cell–ECM protein interactions may underlie the significant immune abnormalities that occur in human diseases. Further elucidation of the role of ECM protein components in regulating T-cell functions is warranted and may lead to the development of novel approaches including gene and antiadhesive therapies for a variety of human disorders. Indeed, cell adhesion-promoting synthetic peptides derived from distinct FN domains have recently been shown not only to inhibit lymphocyte adherence to FN and endothelial cell monolayers in culture but also to interrupt multiorgan cellular infiltration, inflammatory lesions, and development of the lethal wasting syndrome following their infusion into transforming growth factor b1-deficient ‘‘knock-

TABLE 2. Abnormalities in T-cell–ECM protein interactions in various diseases Abnormality ina: Disease

Rheumatoid arthritis Wegener’s granulomatosis Takayasu arteritis Psoriasis Atopic dermatitis Chronic hepatitis Cancer Graft rejection Acute Chronic Sarcoidosis Atherosclerosis

Adhesion

Costimulation

1(FN) 1(C-IV) 1(C-IV, FN) ND ND ND 2

1(C-I, C-IV, FN) 1(FN) None 2(C-I), 1(C-IV) 1(C-IV) 1(C-I) None or 2

2(C-IV, FN) 2(C-IV, FN) ND 1(elastin)

1(C-I, C-IV, FN) N 2(C-I, C-IV, FN) 1(elastin)

a Symbols and abbreviations: 1, enhanced; N, normal; 2, decreased (inhibition of CD3-triggered proliferation); none, ECM proteins do not enhance T-cell proliferation over OKT3-induced level; ND, not determined.

650

MINIREVIEWS

CLIN. DIAGN. LAB. IMMUNOL.

out’’ mice. The effectiveness of these peptides in rat models or in patients with erosive polyarthritis suggests that a role of in vivo interactions between mononuclear cells and FN ligands is vital for the entry of alloreactive cells into the inflamed tissue and offers potential novel sites for therapeutic intervention in patients in various pathological states (28, 65). Immunoregulatory lectin type 1, a galactose-specific lectin produced by subendothelial regions of lymph node vessels, has been shown to interact with the galactose-terminated glycoproteins expressed by T cells. Both prophylactic and therapeutic treatment with immunoregulatory lectin type 1 have been shown to be effective in murine autoimmune disease models of myasthenia gravis and encephalomyelitis. Furthermore, treatment of rats with immunoregulatory lectin type 1 abrogated acute rejection and prolonged the survival of cardiac and skin allografts. This suggests that immunomodulation of basement membrane components may affect host immune events leading to graft rejection (9). Recently, (June 1995) Sanchez-Madrid and colleagues (5) described the expression of activated conformations of b1 integrins (detected by an 15/7 MAb) in lymphocytes from the cellular infiltrates from patients with various chronic inflammatory diseases. The b1 integrin activation-dependent epitope was found to play an important role in T-cell adhesion to FN (the 15/7 MAb had an enhancing effect on VLA-5-mediated cell attachment to FN) and could be induced in vitro by several cytokines. Those data have very important implications suggesting that this epitope could be a possible target for immunointervention (5). While the data on targeting T-cell–ECM protein interactions are still very limited, evidence has accumulated that blocking of leukocyte-endothelial cell interactions may be beneficial as immunosuppressive and anti-inflammatory therapy. Three recent excellent reviews provide in-depth coverage of the progress made in this very promising field (1, 8, 67). ACKNOWLEDGMENTS Our own studies reported in this review were supported by grants from the Committee for Research (KBN) (4 P05B 047 09), INSERM (Paris, France) (to A.G.), and NIH (RO1 A 23847; to J.W.K.-W.). REFERENCES 1. Adams, D. H., and S. Shaw. 1994. Leukocyte-endothelial interactions and regulation of leucocyte migration. Lancet 343:831–836. 2. Allison, J. P. 1994. CD28-B7 interactions in T-cell activation. Curr. Opin. Immunol. 6:414–419. 3. Alon, R., L. Cahalon, R. Hershkowitz, et al. 1994. TNF-a binds to the N-terminal domain of fibronectin and augments the b1-integrin-mediated adhesion of CD41 T lymphocytes to the glycoprotein. J. Immunol. 152:1304– 1313. 4. Arencibia, I., and K. G. Sundqvist. 1989. Collagen receptor on T lymphocytes and the control of lymphocyte motility. Eur. J. Immunol. 19:929–934. 5. Arroyo, A. G., R. Garcia-Vicuna, M. Marazuela, T. A. Yednock, R. GonzalezAmaro, and F. Sanchez-Madrid. 1995. Expression and functional significance of an activation-dependent epitope on the b1 integrins in chronic inflammatory diseases. Eur. J. Immunol. 25:1720–1728. 6. Avdalovic, M., D. Fong, and B. Formby. 1993. Adhesion and costimulation of proliferative responses of human gamma delta T cells by interaction of VLA-4 and VLA-5 with fibronectin. Immunol. Lett. 35:101–108. 7. Bergstrom, S. E., D. Hauzenberger, and K. G. Sundqvist. 1991. T lymphocytes degrade fibronectin. Scand. J. Immunol. 33:453–459. 8. Carlos, T. M., and J. M. Harlan. 1994. Leukocyte-endothelial adhesion molecules. Blood 84:2068–2101. 9. Coito, A. J., M. de Sousa, and J. W. Kupiec-Weglinski. 1994. The role of cellular and extracellular matrix adhesion proteins in organ transplantation. Cell Adhesion Commun. 2:249–255. 10. Collins, T. L., P. D. Kassner, B. E. Bierer, and S. F. Burakoff. 1994. Adhesion receptors in lymphocyte activation. Curr. Opin. Immunol. 6:385–393. 11. Dang, N. H., Y. Torimoto, S. F. Schlossman, and C. Morimoto. 1990. Human CD4 helper T cell activation: functional involvement of two distinct collagen receptors, 1F7 and VLA integrin family. J. Exp. Med. 172:649–652. 12. Davis, L. S., N. Oppenheimer-Marks, J. L. Bednarczyk, B. M. McIntyre, and

P. E. Lipsky. 1990. Fibronectin promotes proliferation of naive and memory T cells by signalling through both the VLA-4 and VLA-5 integrin molecules. J. Immunol. 145:785–793. 13. Erickson, H. P., and M. A. Bourdon. 1989. Tenascin: an extracellular matrix protein prominent in specialized embryonic tissues and tumors. Annu. Rev. Cell Biol. 5:71–75. 14. Faull, R. J., N. L. Kovach, J. M. Harlan, and M. H. Ginsberg. 1994. Stimulation of integrin-mediated adhesion of T lymphocytes and monocytes: two mechanisms with divergent biological consequences. J. Clin. Invest. 179: 1307–1316. 15. Garcia-Vicuna, R., A. Humbria, A. A. Postigo, C. Lopez-Ellzaurdia, M. O. de Landazuri, F. Sanchez-Madrid, and A. Laffon. 1992. VLA family in rheumatoid arthritis: evidence for in vivo regulated adhesion of synovial fluid T cells to fibronectin through VLA-5 integrin. Clin. Exp. Immunol. 88:435–441. 16. Glinski, W., B. Stepien-Sopniewska, G. Korczak-Kowalska, M. GlinskaFerenz, and A. Gorski. Modulation of CD3 dependent lymphocyte proliferation by ECM proteins in atopic dermatitis. Acta Dermatol. Venerol., in press. 17. Glinski, W., B. Stepien-Sopniewska, S. Majewski, M. Glinska-Ferenz, and A. Gorski. 1993. Alterations of T-cell:extracellular matrix proteins interactions in psoriasis. Immunol. Lett. 35:153–158. 18. Goldman, R., J. Harvey, and N. Hogg. 1992. VLA-2 is the integrin used as a collagen receptor by leukocytes. Eur. J. Immunol. 22:1109–1114. 19. Gorski, A. 1994. The role of cell adhesion molecules in immunopathology. Immunol. Today 15:251–255. 19a.Gorski, A., et al. Unpublished data. 20. Gorski, A., V. Castronovo, B. Stepien-Sopniewska, et al. 1994. Depressed immune surveillance against cancer: role of deficient T cell:extracellular matrix interactions. Cell Adhesion Commun. 2:225–233. 21. Gorski, A., T. Mrowiec, L. Paczek, and B. Stepien-Sopniewska. 1993. Abnormal T-cell costimulation by proteins of the extracellular matrix in human renal allografting. Res. Immunol. 144:305–309. 22. Gorski, A., J. Piekarczyk, H. Skurzak, et al. 1993. Abnormalities in T cell interactions with the extracellular matrix proteins in a patient with Wegener’s granulomatosis. Clin. Immunol. Immunopathol. 69:149–154. 23. Hauzenberger, D., and K. G. Sundqvist. 1993. Fibronectin at the lymphocyte surface. Evidence for activation-dependent binding to VLA4 and VLA5 integrins. Scand. J. Immunol. 37:87–95. 24. Hemesath, T. J., L. S. Marton, and K. Stefansson. 1994. Inhibition of T cell activation by the extracellular matrix protein tenascin. J. Immunol. 152: 5199–5207. 25. Hershkowitz, R., L. Cahalon, D. Gilat, S. Miron, A. Miller, and O. Lider. 1993. Physically damaged extracellular matrix induces TNF-a secretion by interacting resting CD41 T cells and macrophages. Scand. J. Immunol. 37: 111–115. 26. Hershkowitz, R., D. Gilat, S. Miron, et al. 1993. Extracellular matrix induces tumour necrosis factor-a secretion by an interaction between resting rat CD41 T cells and macrophages. Immunology 78:50–57. 27. Hinek, A. 1994. Nature and the multiple functions of the 67-kD elastin/ laminin binding protein. Cell Adhesion Commun. 2:185–193. 28. Hines, K. L., A. B. Kulkarni, J. B. McCarthy, H. Tlan, J. M. Ward, M. Christ, N. L. McCartney-Francis, L. T. Furcht, S. Karisson, and S. M. Wahl. 1994. Synthetic fibronectin peptides interrupt inflammatory cell infiltration in transforming growth factor b-1 knockout mice. Proc. Natl. Acad. Sci. USA 91:5187–5191. 29. Hynes, R. O. 1992. Integrins: versatility, modulation, and signalling in cell adhesion. Cell 69:11–25. 30. Imiela, J., A. Gorski, B. Dybowska, R. Malecki, and J. Nosarzewski. 1994. Disturbances of T-cell interactions with endothelium and the extracellular matrix proteins in a patient with Takayasu arteritis. J. Intern. Med.235:609–612. 31. Ishikawa, N., K. Eguchi, Y. Uekii, et al. 1993. Expression of adhesion molecules on infiltrating T cells in thyroid glands from patients with Graves’ disease. Clin. Exp. Immunol. 94:363–370. 32. Juliano, R. L., and S. Haskill. 1993. Signal transduction from the extracellular matrix. J. Cell Biol. 120:577–585. 33. Korczak-Kowalska, G., B. Stepien-Sopniewska, T. Mrowiec, et al. 1995. Rejection-associated abnormalities of T-cell interactions with extracellular matrix. Transplant. Proc. 27:903–906. 34. Kupiec-Weglinski, J. W., A. Coito, A. Gorski, and M. de Sousa. 1995. Lymphocyte migration and tissue positioning in allograft recipients: the role played by extracellular matrix proteins. Transplant. Rev. 9:29–40. 35. Lause, D. B., J. E. Doran, J. A. Houston, and D. H. Beezhold. 1983. Modulation of rat lymphocyte transformation by plasma fibronectin. J. Reticuloendothel. Soc. 34:437–448. 36. Lawler, J., R. Weinstein, and R. O. Hynes. 1988. Cell attachment to thrombospondin: the role of ARG-GLY-ASP, calcium, and integrin receptors. J. Cell Biol. 107:2351–2361. 37. Lazarovitz, A. I., and J. Karsh. 1993. Differential expression in rheumatoid synovium and synovial fluid of a4b7 integrin. J. Immunol. 151:6482–6489. 38. Li, Y. Y., and H. T. Cheung. 1992. Basement membrane and its components on lymphocyte adhesion, migration, and proliferation. J. Immunol. 149: 3174–3181.

VOL. 2, 1995 39. Lobb, R. R., and M. E. Hemler. 1994. The pathophysiologic role of a4 integrins in vivo. J. Clin. Invest. 94:1722–1728. 40. Matsuyama, T., A. Yamaga, J. Kay, K. M. Yamada, S. K. Akiyama, S. F. Schlossman, and C. Morimoto. 1989. Activation of CD4 cells by fibronectin anti-CD3 antibody. A synergistic effect mediated by the VLA-5 fibronectin receptor complex. J. Exp. Med. 170:1133–1148. 41. Mattern, T., W. Scholz, A. C. Feller, A. D. Flad, and A. J. Ulmer. 1991. Expression of CD26 (dipeptidyl peptidase IV) on resting and activated human T-lymphocytes. Scand. J. Immunol. 33:737–748. 42. Moria, A., Z. Zhang, and E. Ruoslahti. 1994. Superfibronectin is a functionally distinct form of fibronectin. Nature (London) 367:193–196. 43. Mrowiec, T., J. Cianciara, W. Stanczak, B. Stepien-Sopniewska, and A. Gorski. 1994. Enhanced signalling by collagen type I in T cell responses in chronic hepatitis and cirrhosis. Int. J. Immunopathol. Pharmacol. 7:133– 137. 44. Olt, G., A. Berchuck, P. Soisson, C. M. Boyer, and R. C. Bast. 1992. Fibronectin is an immunosuppressive substance associated with epithelial ovarian cancer. Cancer 70:2137–2142. 45. O’Rourke, A. M., and M. F. Mescher. 1992. Cytotoxic T-lymphocyte activation involved a cascade of signalling and adhesion events. Nature (London) 358:253–255. 46. Podobin ˜ ska, I., B. Stepien-Sopniewska, T. Mrowiec, G. Korczak-Kowalska, P. Wierzbicki, K. Lis, and A. Gorski. 1995. Allogeneic sensitization alters T-cell reactivity to extracellular matrix proteins in prospective transplant recipients. Transplant. Proc. 27:494. 47. Postigo, A. A., P. Sanchez-Mateos, A. I. Lazarovits, F. Sanchez-Madrid, and M. O. de Landazuri. 1993. a4b7 integrin mediates B cell binding to fibronectin and vascular cell adhesion molecule-1. Expression and function of a4 integrins on human B lymphocytes. J. Immunol. 151:2471–2483. 48. Raghow, R. 1994. The role of ECM in post-inflammatory wound healing and fibrosis. FASEB J. 8:823–831. 49. Rodriguez, R. M., C. Pitzalis, G. H. Kingsley, E. Henderson, M. J. Humphries, and G. S. Panayi. 1992. T lymphocyte adhesion to fibronectin (FN): a possible mechanism for T cell accumulation in rheumatoid joint. Clin. Exp. Immunol. 89:439–445. 50. Romanov, V., M. E. Sobel, P. P. da Silva, S. Menard, and V. Castronovo. 1994. Cell localization and redistribution of the 67 kD laminin receptor and a6b1 integrin subunits in response to laminin stimulation: an immunogold electron microscopy study. Cell Adhesion Commun. 2:201–209. 51. Ruegg, T. J., R. Chiquet-Ehrismann, and S. S. Alkan. 1989. Tenascin, an extracellular matrix protein, exerts immunomodulatory activities. Proc. Natl. Acad. Sci. USA 86:7437–7441. 52. Shimizu, Y., and S. Shaw. 1991. Lymphocyte interactions with extracellular matrix. FASEB J. 5:2292–2299. 53. Shimizu, Y., G. A. van Seventer, K. J. Horgan, and S. Shaw. 1990. Costimulation of proliferative responses of resting CD4 T cells by the interaction of VLA-4 and VLA-5 with fibronectin or VLA-6 with laminin. J. Immunol. 145:59–67. 54. Shimizu, Y., G. A. van Seventer, K. J. Horgan, and S. Shaw. 1990. Regulated expression and binding of three VLA (beta 1) integrin receptors on T cells.

MINIREVIEWS

651

Nature (London) 345:250–253. 55. Simon, M. M., M. D. Kramer, M. Prester, and S. Gay. 1991. Mouse T-cell associated serine proteinase 1 degrades collagen type IV: a structural basis for the migration of lymphocytes through vascular basement membranes. Immunology 73:117–119. 56. Sturmho ¨fel, K., C. Brando, F. Martinon, E. M. Shevach, and J. E. Coligan. 1995. Antigen-independent, integrin-mediated T cell activation. J. Immunol. 154:2104–2111. 57. Sundqvist, K. G., L. Pedari, and D. Hauzenberger. 1993. Anchorage and lymphocyte function: extracellular matrix substrata control morphogenesis and interleukin production but have minor effects on DNA synthesis. Scand. J. Immunol. 37:295–307. 58. Sundqvist, K. G., L. Wanger, A. Heimdal, B. Lonnqvist, and D. Hauzenberger. 1991. Fibronectin-associated with infiltrating T lymphocytes. Evidence for in situ localization in biopsies and synthesis in vitro. Eur. J. Immunol. 21:299–303. 59. Suthanthiran, M., and B. T. Strom. 1994. Renal transplantation. N. Engl. J. Med. 331:365–376. 60. Takeuchi, T., K. Amano, H. Sekine, J. Koide, and T. Abe. 1993. Upregulated expression and function of integrin adhesive receptors in systemic lupus erythematosus patients with vasculitis. J. Clin. Invest. 92:3008–3016. 61. Ticchioni, M., M. Deckert, G. Bernard, et al. 1995. Comitogenic effects of VLA antigens on CD3-stimulated human thymocytes. J. Immunol. 154: 1207–1215. 62. Torimoto, Y., H. Sekaya, H. Kohda, and M. Namiki. 1993. Enhancement of CD26 (DPIV) expression on hepatocyte in chronic liver diseases and activated lymphocytes. Gastroenterology 104:1010A. 63. Utsumi, K., M. Sawada, S. Narumiya, et al. 1991. Adhesion of immature thymocytes to thymic stromal cells through fibronectin molecules and its significance for the induction of thymocyte differentiation. Proc. Natl. Acad. Sci. USA 88:5685–5689. 64. Wadsworth, S., M. J. Halvorson, and J. E. Coligan. 1992. Developmentally regulated expression of the b4 integrin on immature mouse thymocytes. J. Immunol. 149:421–428. 65. Wahl, S. M., J. B. Allen, K. L. Hines, T. Imamichi, A. M. Wahl, L. T. Furcht, and J. B. McCarthy. 1994. Synthetic fibronectin peptides suppress arthritis in rats by interrupting leukocyte adhesion and recruitment. J. Clin. Invest. 94:655–662. 66. Walle, T. K., T. Helve, I. Virtanen, and P. Kurki. 1994. Increased expression of VLA5 adhesion molecules on synovial fluid T lymphocytes in chronic polyarthritis: a consequence of T-cell activation. Scand. J. Immunol.39:189–194. 67. Ward, P. A., and M. S. Mulligan. 1994. Blocking of adhesion molecules in vivo as anti-inflammatory therapy. Ther. Immunol. 1:165–171. 68. Weeks, B. S., M. E. Klotman, S. Dhawan, et al. 1991. HIV-1 infection of human T lymphocytes results in enhanced a5b1 integrin expression. J. Cell Biol. 114:847–853. 69. Yamada, A., Y. Nojima, K. Sugita, N. H. Dang, S. F. Schlossman, and C. Morimoto. 1991. Cross-linking of VLA/CD29 molecule has a co-mitogenic effect with anti-CD3 on CD4 cell activation in serum-free culture system. Eur. J. Immunol. 21:319–325.