|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
First published online 17 July 2006
doi: 10.1242/jcs.03057
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
1ß1-integrin engagement to distinct laminin-1 domains orchestrates spreading, migration and survival of neural crest cells through independent signaling pathways
1 Laboratoire de Biologie du Développement, CNRS et Université Pierre et Marie Curie, 9 quai Saint-Bernard, 75252 Paris Cedex 05, France
2 Unité Mixte de Recherche 6032, CNRS et Université de la Méditerranée, Facultés de Médecine et de Pharmacie, Marseille, France
3 IFR 128 BiosSciences Lyon-Gerland, Institut de Biologie et Chimie des Protéines, CNRS et Université Lyon-1, Lyon, France
Author for correspondence (e-mail: duband{at}ccr.jussieu.fr)
Accepted 18 May 2006
 |
Summary |
|---|
 Top Summary IntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
1 subunit and can be separated in the LN-1 elastase proteolytic fragments E1' and E8. Interaction with either domain was mediated by the same integrin 1ß1 but produced distinct, complementary responses through specific signaling cascades. FAK activation upon E8 binding induced spreading, formation of actin bundles and focal adhesions, stimulated oriented migration, but failed to support survival. Conversely, Erk activation upon E1' binding promoted long-term survival and random migration without actin reorganization. Consistent with this, interaction with laminin-5 or laminin-10/11, which do not harbor integrin-binding domains in the N-terminal side of their chains, failed to support survival. Thus, the signaling activity and function of integrins might depend on binding domains in their ligands, thereby revealing ligand control of integrin function as a possible mechanism for the modulation and coordination of cell response to adhesive signals.
Key words: Neural crest, Integrin, Laminin, Cell migration, Cell survival, Signaling pathways
|
Introduction |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
; de Arcangelis and Georges-Labouesse, 2000). Among the different families of adhesion molecules, integrins are primarily involved in physical aspects of ECM adhesion (Hynes and Zhao, 2000; Hynes, 1992). Central to their function is their ability to promote actin assembly to generate tension locally via the recuitment of molecules that activate the actin polymerization machinery or anchor it to adhesion sites (DeMali et al., 2003; Geiger et al., 2001). In addition to their structural and mechanosensory roles, integrins are able to activate a variety of tyrosine kinases and GTPases (e.g. focal adhesion kinase (FAK) and members of the Rho family of GTPases) to induce multiple downstream signaling pathways (Giancotti and Ruoslahti, 1999; Giancotti and Tarone, 2003). Furthermore, integrin signals cooperate with and modulate signaling events initiated by growth factor receptors (ffrench-Constant and Colognato, 2004). Thus, because of this dual physical and chemical signaling activity, integrins influence numerous aspects of cell behavior, including migration, proliferation, survival and differentiation. However, although much progress has been achieved through the identification of cellular partners of integrins, understanding how these receptors can coordinately regulate such a diversity of cellular events remains a major challenge.
The neural crest of the vertebrate embryo constitutes a powerful paradigm to study the multiple roles of integrins in the control of cell adhesion, migration, growth, and differentiation. This discrete cell population segregates from the neural epithelium and travels long distances to populate various areas of the embryo where it provides a large array of cell types (Le Douarin and Kalcheim, 1999). So far, most studies on the function of integrins during crest ontogeny focused primarily on the mechanical aspect during migration. Neural crest cells follow restricted pathways filled with ECM material and lined by the basal laminae of epithelia (Erickson and Perris, 1993). Numerous studies have provided evidence that the ECM encountered by crest cells serves as a scaffold onto which they migrate, and that integrins play a prominent role in this process. In vitro, crest cells are able to spread and migrate in an integrin-dependent manner onto a variety of ECM components, including fibronectin (FN), laminin-1 (LN-1), vitronectin (VN) and collagens (Delannet et al., 1994; Desban and Duband, 1997; Newgreen et al., 1982; Perris et al., 1989) and, in vivo, blocking agents to ECM molecules or integrins all perturb crest cell migration (Boucaut et al., 1984; Bronner-Fraser, 1986; Kil et al., 1996).
However, it is now increasingly clear that integrin functions during neural crest development are not limited strictly to substrate anchorage and cell motion. Crest cells express a multiplicity of integrins, and not all of them are implicated in adhesion and migration (Delannet et al., 1994; Desban and Duband, 1997; Testaz et al., 1999). Such a diversity of receptors certainly reflects the very changing nature of the environment to which crest cells are confronted during migration, but it also illustrates the necessity of additional roles for integrins not directly related to matrix adhesion. For example, integrins have been found to control surface distribution and activity of N-cadherin during migration (Monier-Gavelle and Duband, 1997). Integrins are also involved in maintenance of cell survival (Crump et al., 2004; Haack and Hynes, 2001; Testaz and Duband, 2001; Yang et al., 1993); the primary defect observed in crest cells of embryos depleted in individual integrins is selective cell death during migration, thereby revealing that anchorage-dependent survival signals elicited by integrins are of paramount importance for cells confronted with a continuously changing environment. Finally, changes in the integrin repertoire during crest differentiation and the numerous alterations observed in embryos with conditional ß1-integrin gene deletion argue for integrin implication in late neural crest development, during lineage segregation, cell differentiation and final maturation of the nervous system (Breau et al., 2006; Bronner-Fraser et al., 1992; Duband et al., 1992; Pietri et al., 2004). Thus, it seems likely that integrins are involved in multiple cellular events during neural crest development. However, these functions await further characterization and signaling cascades that are implicated for their coordination over time and space remain to be established.
In the present study, to uncover the different facets of integrin functions during neural crest development, we revisited in detail crest cell response to LN-1, one of the main constituents of basal laminae lining their migration routes (Duband and Thiery, 1987). We found that LN-1 promoted crest cell spreading, migration, and survival, through two distinct domains interacting with the same integrin that activated specific signaling cascades, resulting in distinct, complementary responses. These results reveal that signals conveyed by integrins might depend on the nature of their ligands. We propose that such a mechanism accounts for the coordination and diversity of integrin-dependent signals in cells.
|
Results |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
1 chain, E8 interacts with integrins through a globular module at the C-terminus of the 1 chain; this binding requires structural integrity and is sensitive to heat denaturation (Colognato and Yurchenco, 2000). To gain a better understanding of the cellular response of neural crest cells to LN-1, we examined their ability to interact with, spread, migrate and survive on LN-1, and investigated whether the E1' and E8 fragments are involved in each process using a series of specific in vitro assays. To avoid any bias due to serum-derived components, all assays were performed in defined, serum-free culture medium.
|
|
Crest cell spreading on LN-1 and LN-1 elastase fragments E1' and E8 was evaluated by visualizing cell morphologies after 1-hour-adhesion assays (Fig. 1C-F). Both LN-1 and E8 promoted substantial cell spreading, with maximal values being attained at 10 µg/ml. Cells exhibited their typical stellate morphology, and no striking difference was detected in cell shapes between the substrates. Cell spreading was poorer on E1
, even at high coating concentrations. Most cells remained round and the few spread cells were never entirely flattened, instead they showed a bipolar morphology with thin filopodia. Unlike adhesion, spreading to LN-1 was fully sensitive to denaturation, confirming that it was primarily mediated by E8 (Table 1).
Crest cell migration on LN-1 and LN-1 elastase fragments E1' and E8 was analysed in culture of neural tube explants. Movement of the whole cell population was estimated by measuring the linear distance between the periphery of the outgrowth and the edge of the neural tube and movement of individual cells was followed by videocinematography (Duband et al., 1991; Dufour et al., 1988). LN-1 supported crest cell migration with great efficacy (Fig. 1G,H,N). Migration was dose-dependent within the range of 1-10 µg/ml and was maximal at higher concentrations. The cell population travelled about 800 µm within 24 hours with an average velocity of 115 µm/hour and a persistence of movement of 0.3 for individual cells. Cells showed a well-spread morphology with long protrusions. The E8 fragment promoted cell migration as efficiently as LN-1 (Fig. 1G,I,N): values for the distance covered by the population and for the velocity of individual cells were almost identical to those on LN-1. The only notable difference was that cells were less compact and less numerous on E8. On E1', crest cells exhibited a significantly-reduced migratory potential (Fig. 1G,J,N). Cells migrated half of the distance covered on LN-1 and were compact. Individual cells exhibited round shapes with thin, short protrusions and their velocity was reduced compared with that on LN-1 or E8. Even more significant was the reduction of the persistence of movement, which reflected a random migration compared with the more-oriented migration observed on LN-1 or E8. In contrast to E1', migration on E8 was totally abolished upon its heat-denaturation (Fig. 1L,M, Table 1). Interestingly, heat-denaturation of LN-1 resulted in crest cell migration to E1' level (Fig. 1K, Table 1).
To analyse cell survival, we designed an assay derived from adhesion assays. Crest cells collected from neural tube primary cultures were deposited into dishes coated with LN-1 or LN-1 elastase fragments adsorbed at 1-100 µg/ml in the complete absence of external influence, e.g. from the neural tube, and were cultured for several days without further change of medium. At defined time points (4, 12, 24 and 48 hours), adherent cells were counted and their morphologies were scrutinized. This assay allowed us to reliably follow survival over time in culture. Coating concentrations of LN-1 or fragments below 10 µg/ml did not support crest cell survival because the number of adherent cells declined rapidly after 4 hours (Fig. 2A). These concentrations corresponded to those at which cell spreading was minimal. By contrast, LN-1 used at 10 µg/ml or more permitted survival with great efficiency, as revealed by the rather constant number of cells adhering to the dish throughout the course of the experiment (Fig. 2A). Analysis of cellular morphologies revealed that cells became more flattened after 24 hours and that, after 48 hours, a number of them differentiated into neurons and grew long neurites (Fig. 3A). On E8, cell survival was quite poor even at high concentrations of the substrate: the number of adherent cells decreased constantly after 4 hours and almost none could be detected on the dish after 48 hours (Fig. 2A, Fig. 3B). After 24 hours, a strong proportion of cells displayed multiple signs of degeneration: retraction of processes, intense blebbing and overall decrease of the cellular area (Fig. 3B). In contrast to E8, crest cells survived well on E1', although they were originally much less flattened. The profiles of the survival curves were similar to those on native LN-1 (Fig. 2A). Cells underwent progressive spreading onto the fragment and, after 48 hours, cell shapes became indistinguishable from those on LN-1, except that we never observed neuronal differentiation (Fig. 3C). Heat denaturation of LN-1 did not affect its survival-promoting activity, thereby confirming that this effect was essentially mediated by E1' (Fig. 2B). Finally, we verified that the ability of crest cells to spread gradually onto E1' and survive on it essentially relied on E1' and not on other ECM molecules released and deposited onto the substrate, such as FN or VN, because only antibodies against LN-1 totally abrogated spreading and survival on E1' (data not shown).
|
|
In the subsequent experiments, LN-1 and its fragments were used at the minimal concentration for which maximal activity was achieved. This corresponded for all assays to coating concentrations of 10-20 µg/ml of LN-1 (i.e. 10-20 nM), 10 µg/ml of E8 (i.e. 50 nM) and 10-20 µg/ml of E1' (i.e. 25-50 nM).
Cellular responses of neural crest cells to LN-5, and LN-10 and LN-11
Our results indicate that crest cells can adhere to, spread, migrate and survive on LN-1 and that binding to either binding domain located in E1' and E8 elicit distinct cellular responses. To further document this observation, using the same assays, we analysed the cell response to LN-5 and LN-10/11, two other members of the family that, toward the C-terminus of their 3 and 5 chains, harbor an integrin-binding domain equivalent to the one located in E8, but that differ from LN-1 in that no cell-binding site has been identified in the N-terminus (Fig. 4A). As shown in Fig. 4BC, adhesion to and spreading onto LN-5 were minimal compared with LN-1. In migration assays, although crest cells were able to escape from the neural tube and moved onto LN-5, the distance they migrated was much shorter than on LN-1, and cells were poorly spread with tiny processes (Fig. 4D,E). Finally, unlike LN-1, LN-5 did not support cell survival (Fig. 4G); cells were unable to adhere firmly to the substrate within 4 hours and, after 24 hours, they formed aggregates that ultimately detached from the dish and died (Fig. 3D).
|
These observations highlight the crucial role of the binding domain situated in the E1' fragment in the cellular response to LN-1, and reveal that the binding sites in the C-terminus of the 1 and 5 chain of LN-1 and LN-10/11, respectively, are functionally interchangeable for crest cells but distinct from that of the 3 chain of LN-5.
Cellular responses of neural crest cells to LN-1 are mediated by 1ß1 integrin
We have shown previously that truncal crest cells express 1ß1, 3ß1 and 6ß1 as integrin receptors for LN-1 (Desban and Duband, 1997). To identify which of them are involved in mediating spreading, migration and survival, we made use of specific antibodies as competitors for LN-1 binding. Inhibitory antibodies against 1ß1 almost completely abrogated cell spreading and migration on LN-1 as well as on its fragments (Fig. 5A,B). On E8, these antibodies caused detachment of the neural tube from the dish, leaving only a few poorly spread crest cells on the substratum (Fig. 5C), whereas on E1', crest cells were unable to disperse although the neural tube remained adherent to the dish (Fig. 5D). Moreover, survival was severely compromised in cells detaching from the substrate after treatment with the antibodies against 1ß1, because they systematically exhibited signs of apoptosis and died after detachment (data not shown). Spreading and migration on LN-10/11 were also almost completely blocked by antibodies against 1ß1 (Fig. 5A,B,E). Non-inhibitory antibodies against 1ß1 and 3ß1 or 6ß1 were, by contrast, totally ineffective on cell spreading, migration and survival on LN-1 and its fragments as well as on LN-10/11 (data not shown).
|
To further evaluate the contribution of integrins to cell spreading, motility and survival on LN-1, we used the fact that antibodies against integrins can mimic the effect of integrin ligands and elicit integrin-dependent cellular responses when immobilized on the substrate or coupled to beads (Duband et al., 1991; Miyamoto et al., 1995). Antibodies against 1ß1 adsorbed to the dish supported cell adhesion, spreading and migration at rates comparable with those of LN-1, whereas antibodies against 6ß1 promoted cell adhesion with poor efficiency and totally failed to mediate spreading and migration (Fig. 6A-C). On antibodies against 1ß1, crest cells migrated significant distances and, after 24 hours in culture, they organized into outgrowths with cell densities and shapes approaching those obtained on LN-1 (Fig. 6D). By contrast, on antibodies against 6ß1 cells were able to detach from the neural tube, but once they left its close vicinity they became round and migrated only a few cell diameters away from the tube (Fig. 6E). Antibodies against 1ß1 but not 6ß1 favored cell survival. The dose-response curves of survival over time as a function of the concentration of the antibody against 1ß1 adsorbed to the dishes matched those on LN-1 or E1' (Fig. 6F). In addition, cells exhibited spread shapes and spatial organization indistinguishable from those on LN-1 and, as observed for LN-1, neuronal differentiation was often observed in the cultures (Fig. 6G). On antibodies against 6ß1, cells first were round with tiny processes. Instead of spreading after prolonged culture, they formed clusters of round cells and ultimately died after detachment from the dish (Fig. 6H). These observations indicate that crest cell responses to LN-1 are mediated essentially by a single integrin, 1ß1, interacting with specific binding sites in the E1' and E8 fragments.
|
Cellular responses to LN-1 domains are mediated by distinct signaling cascades
To understand how integrin binding to separate LN domains can elicit different cellular responses, we investigated downstream signaling pathways. First, we compared the organization of adhesion sites and cytoskeleton in crest cells cultured on each fragment for 4-24 hours. On LN-1 and E8, cells formed broad lamellipodia soon after spreading, and ß1-integrins were concentrated as a thin band at the periphery of the lamellipodium, in protruding spikes and in more centrally found focal contacts (Fig. 7A,B). On E1', by contrast, cells were able to organize large processes only after about 12-24 hours, and ß1-integrins were never observed in focal contacts and at the cell periphery. Instead, they displayed a uniform, punctate pattern throughout the cell body (Fig. 7C). It should be noted that non-ß1 integrins, such as the VN receptors vß3 and vß5, that could compensate for the lack of ß1 integrins in focal contacts were not found on cells cultured on E1' (data not shown). Consistent with integrin distribution, focal contacts seen on cells cultured on LN-1 or E8 were decorated by antibodies against paxillin (Fig. 7D,E), FAK, vinculin, -actinin and talin (data not shown), and actin was assembled into stress fibers irradiating toward the cell periphery (data not shown). On E1', no such concentrations of focal contact components were present and few actin bundles were visualized (Fig. 7E).
|
). Since phosphorylation of tyrosine on position 397 (Y397) is a hallmark of FAK activation, we investigated its phosphorylation state and cellular distribution in crest cells cultured on LN-1, or fragments E1' or E8. On LN-1 and E8 but not on E1', phosphorylation of FAK on Y397 was detected in focal contacts coinciding with ß1-integrins. However, whereas immunostaining studies did not reveal significant differences between LN-1 and E8, immunoblotting revealed that FAK was considerably more phosphorylated in extracts of cells cultured on E8 than on LN-1, and almost not phosphorylated on denatured LN-1 or E1' (Fig. 7G). FAK was also activated in crest cells cultured on antibodies against 1ß1 (Fig. 7H). These results therefore suggest that integrin binding to E8, but not to E1', recruits FAK and associated components, correlating with an elaborated actin cytoskeletal network.
To identify putative signals downstream of E1' binding, we analysed Erk activation in crest cells cultured over LN-1 or fragments. As shown in Fig. 8A, a strong band corresponding to phosphorylated Erk was detected in extracts of cells cultured on E1' or denatured LN-1. On intact LN-1, the intensity of the band was slightly less than on E1', whereas on E8, it was barely detectable. Erk was also strongly phosphorylated in cells cultured on antibodies against 1ß1 (Fig. 8B). Immunofluorescence analyses revealed intense nuclear staining for phosphorylated Erk in cells cultured on LN-1 and E1', thereby confirming that Erk was activated and subsequently translocated to the nucleus (Fig. 8C,D). Consistent with the faint band detected in cellular extracts, only poor nuclear staining for phosphorylated Erk was visualized in cells cultured on E8 (Fig. 8E). To further assess the implication of the MAP kinase (MAPK) pathway in crest cell response to LN-1 and E1', we analysed the effects of PD98059, a cell-permeable inhibitor of MAP kinase kinase (MEK). PD98059 reduced significantly adhesion to E1' and abrogated almost completely spreading and migration (Fig. 8F-H,J). By contrast, it was not very effective on E8 (Fig. 8F-H,K). In survival assays, addition of PD98059 to cells cultured on E8 did not induce premature cell death, whereas it readily caused massive cell death on E1' (Fig. 8M,N). On LN-1, PD98059 caused a two-fold reduction of adhesion and spreading but did not affect migration capacity, although it caused a severe decrease in the number of cells (Fig. 8F-I). Finally, it prevented survival on LN-1 almost as dramatically as on E1' (Fig. 8L). These observations indicate that binding to E1' but not to E8 in crest cells activates the MAPK signaling cascade.
|
|
Discussion |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
1ß1, interacting with two distant, independent domains of the LN-1 molecule located in its E8 and E1' fragment. Binding to E8 allowed rapid cell spreading, associated with the presence of actin microfilament bundles and numerous focal adhesions, and stimulated active, oriented cell migration. However, it failed to promote cell survival in the long term (Fig. 9A). Conversely, interaction with E1' promoted long-term cell survival without leading to major cytoskeletal remodeling: it could not support rapid cell spreading, and the microfilament network it induced in the long term was poorly elaborated with few focal contacts. Furthermore, E1' supported only moderate and random migration (Fig. 9A). Consistent with these observations, interaction of crest cells with LN-5 or LN-10/11, that do not harbor an integrin-binding domain in the N-terminal part of their chains, failed to support survival (Fig. 9B,C). In addition, although we have not investigated the implication of all signaling pathways potentially activated by integrins in these cellular events, we identified two major cascades that were alternatively triggered upon integrin binding to E8 and E1', and may account for the differences in cell responses. Interaction with E8 caused a massive recruitment of FAK to focal contacts and its activation, whereas binding to E1' activated the MAPK pathway. Furthermore, inhibition of the MAPK cascade by PD98059 totally abrogated all E1'-dependent cellular responses.
|
Although data on the precise location of the various LN isoforms in embryos are still sketchy, previous studies clearly established the widespread occurrence of LN in the vicinity of neural crest cells throughout their ontogeny (Duband and Thiery, 1987; Miner et al., 1998; Rogers et al., 1986; Tosney et al., 1994). Thus, the 1 chain, essentially characteristic of LN-1, has been found to be abundant along the migration routes in mouse (Miner et al., 1998). By contrast, the 3 subunit of LN-5 has not been reported to be associated with crest cells at any phase of their migration or later (Aberdam et al., 1994; Lu et al., 2001), thereby explaining the limited range of cellular responses it promoted in vitro. Interestingly, the 5 chain, constituting LN-10/11, has been identified in a screen for genes that are upregulated in response to induction of crest cells and it is expressed along their migration routes but only at the initial and terminal phases (Coles et al., 2006; Gammill and Bronner-Fraser, 2002; Miner et al., 1998). Our data, showing that LN-10/11 can elicit spreading, immediate but not sustained migration as well as neuronal differentiation, are therefore fully consistent with those describing the distribution of the 5 chain, and the abnormalities in crest migration and delayed differentiation observed in the 5 mutant mice (Coles et al., 2006). Among the other laminin isoforms examined so far, LN-2 and LN-4 (sharing the 2 chain) appear late during migration and are likely to be implicated only in neuronal differentiation, whereas LN-8 (containing the 4 chain) has been tentatively proposed to be present during migration although firm evidence for this is still lacking (Colognato and Yurchenco, 2000; Perris and Perissinotto, 2000). Therefore, it is intriguing that only LN-1 is associated with crest cells throughout their development, whereas the other isoforms - notably LN-10/11 and LN-5 that do not harbor an integrin domain in the N-terminus of their chains and do not support survival - are either absent or present only transiently. This suggests that LN-1 is fundamental for neural crest development in vivo and supports the view that, the ECM encountered by crest cells does not only play a permissive role by serving as a scaffold for migration but also instructs cells with guidance, survival and differentiation cues.
Erk and FAK have been demonstrated to function as essential integrators of signals emanating from growth factor and adhesion receptors, respectively, to regulate gene expression, cell growth and survival, and to induce cell shape, spreading and motility (Howe et al., 2002; Mitra et al., 2005). However, several studies established possible connections between the MAPK and the FAK signaling pathways, and dissected their functional specificities in cell migration and growth. Thus, Erk has also been shown to contribute to the regulation of cell motility by phosphorylation of myosin-light-chain kinase but, unlike FAK, it plays no significant role in spreading (Klemke et al., 1997). Using cells transfected with PTEN, a potent inhibitor of cell migration, Gu and colleagues showed that overexpression of FAK, the adaptor protein Shc or an constitutively-activated form of MEK-1 can restore cell motility (Gu et al., 1999). Moreover, in a strikingly similar fashion to crest cells cultured on E8 and E1', Shc and FAK were found to regulate cell movement through two different, independent mechanisms: one is a pathway starting at Shc via the MAPK pathway, leading to the stimulation of random migration associated with a poor actin cytoskeletal organization and few focal adhesions. The other starts at FAK proceeds via Cas and leads to extensive organization of actin microfilaments and numerous focal adhesions, and to directionally persistent migration. Cho and Klemke showed that migration and survival of COS cells are coordinately regulated through activation of Erk and FAK pathways downstream of integrins and growth-factor receptors (Cho and Klemke, 2000). Finally, two subsets of integrins were distinguished on the basis of their ability to recruit different intracellular partners and to activate distinct signaling pathways that control cell cycle progression (Wary et al., 1996; Wary et al., 1998). 1ß1, 5ß1 and vß3 integrins selectively activate Shc independently of FAK; this event is necessary and sufficient to activate the MAPK signaling pathway, promoting cell-cycle progression. 1ß1, 5ß1 and vß3 integrins are constitutively linked to the Fyn kinase through caveolin-1 and are coupled to the Ras-Erk pathway. By contrast, other integrins (such as 2ß1, 3ß1 and 6ß1) that are not linked to Shc, fail to activate the MAPK pathway, resulting in cell-cyle arrest and apoptosis. However, they are able to recruit and activate FAK and promote cell spreading.
On the basis of these reports, we propose that neural crest cell response to LN-1 is orchestrated by engagement of 1ß1 integrin to distinct cell-binding domains within LN-1. Interaction with the E8 domain produces cell spreading and promotes persistent migration through activation of a FAK-mediated signaling cascade, whereas interaction with E1' allows cell survival and induces random migration via the MAPK pathway. Whether and how FAK and Erk signaling cascades are interconnected for coordinated cellular response is still an open question. The biochemical analysis of FAK and Erk phosphorylation indicates that both pathways are activated independently: In cells plated on E8, we observed only weak phosphorylation of Erk, whereas FAK was massively recruited to focal contacts and strongly activated. Conversely, in cells confronted with E1', there was a potent activation of Erk, although FAK was not significantly phosphorylated and recruited to focal contacts. However, careful comparison of cell responses to LN-1 and fragments suggests more intricate relationships between both pathways. In fact, crest cells were able to survive on E8 for about 12-24 hours, and it cannot be excluded that FAK signals contribute to some extent to the maintenance of survival. Also, LN-1 systematically produced a more robust migration than E8 and E1' with a greater number of crest cells, suggesting that both FAK and Erk pathways converge or cooperate for optimal migration. Finally, it is striking that the magnitude of the signals elicited on LN-1 was lower than those on E1' or E8; in particular, FAK phosphorylation was much stronger on E8 than on LN-1. Although competition between 1ß1 integrin heterodimers to bind E1' and E8 is a plausible explanation for this, it cannot be excluded that signals downstream of integrin binding to E1' and E8 may counterbalance each other through inhibitory activities. Interestingly, previous studies have identified a suppressor site in the E1' fragment that selectively interferes with E8 activity, but the cellular mechanisms involved in this repression have not been identified (Calof et al., 1994).
An intriguing feature of our study is that crest cell responses to LN-1 were mediated by only one integrin. To our knowledge, this is the first observation of a single integrin simultaneously activating distinct signaling cascades upon binding to separate domains of the LN-1 molecule. It should be stressed that, although 1ß1 integrin is best known as a LN-1 receptor interacting with the N-terminus of the 1 chain (Colognato-Pyke et al., 1995), its ability to interact with a site in the E8 fragment is not unprecedented. Biochemical analyses convincingly demonstrated that in rat hepatocytes, 1ß1 integrin binds both E8 and E1' (Forsberg et al., 1990). Interestingly, 1ß1 binding to E8 in crest cells was found to activate the same signaling pathway and to induce the same range of cellular responses as 6ß1 in endothelial cells (Wary et al., 1996), raising the possibility that the 1ß1 and 6ß1 integrins recognize the same site in the E8 fragment and implying that signaling activity of an integrin is essentially driven by its ligand. This view is further supported by the fact that, like 6ß1 integrin, binding of 1ß1 to E8 is sensitive to heat denaturation. However, the limited capacity of crest cells to adhere to LN-5 means that, in contrast to 6ß1 integrin, 1ß1 does not constitute a receptor for this LN isoform and argues in favor of 1ß1 and 6ß1 binding to non-overlapping regions of E8.
The presence of multiple cell-binding domains located in distant regions of the molecule is a common feature to the major ECM components and it is therefore likely that other integrins behave like 1ß1. Similarly to LN-1, FN has been found to stimulate both Ras-MAPK-dependent and RhoA-dependent signals to control cell-cycle progression in 3T3 cells (Danen et al., 2000). In addition, it harbors several domains, each able to interact with more than one integrin: 4ß1 binds FN primarily in the heparin-binding domain situated towards the C-terminus of the FN chain but it can also interact with the RGD sequence in the central region of the molecule (Sanchez-Aparicio et al., 1994). Likewise, 5ß1 may recognize the RGD sequence or a matrix-assembly site in the N-terminal half of the molecule (Hocking et al., 1998) but it should be stressed that, contrasting with the situation found for crest cell interaction with LN-1, binding to the these sites seems to be exclusive and does not occur coincidently in the same cells. The question remains how this process is achieved at the molecular level. A possible explanation is that, owing to their different biochemical features, the LN-1 domains interacting with 1ß1 induce different conformational alterations within the integrin dimer. Structural analyses established that integrin chains form in the ectodomain, a `head' and two long `legs' spanning the membrane, and that conformation changes in the head region induced by ligand binding propagate to the legs and disrupt their interactions, allowing the integrin molecule to deploy at the cell surface (Xiao et al., 2004). An alternative is that the LN-1 domains interact with two separate regions in the integrin head, also resulting in different conformational changes. In this respect, 3ß1 has been proposed to regulate cell-cell contact and matrix adhesion through distinct ligand-binding sites in its head region (Zhang et al., 2003). Regardless of the molecular mechanism implicated, differences in conformational changes in integrins caused by E1' and E8 are likely to induce recruitment of distinct molecular partners followed by segregation of the resulting complexes in specific cellular compartments or membrane domains. It is then conceivable that interaction of E1' would drive the 1ß1 heterodimer in rafts via caveolin-1 to recruit Shc and activate the MAPK pathway, whereas interaction with E8 would instead lead to 1ß1 clustering in association with FAK in focal contacts. Other plausible candidates can be found among tetraspanins that are commonly engaged in associations with LN-binding integrins and act as modulators of cell spreading, motility and survival (Hemler, 2003). Finally, 1ß1 interaction with either LN-1 domains may selectively recruit distinct cytoskeletal elements, as reported previously for paxillin, which binds the 4 subunit tail with a strong affinity to promote cell migration (Liu et al., 1999).
In conclusion, our study reveals that separate regions within an ECM ligand can simultaneously activate distinct signaling cascades through the same integrin to elicit different cellular responses. Such a mechanism may account for the diversity and specificity of integrin signals in cells and probably provides an additional level of flexibility for cells to adapt rapidly to brutal changes in their environment. Thus, partial degradation of ECM components or physical constraints may locally release or, conversely, unmask cell-binding domains within these molecules, resulting in the selective activation or repression of signaling cascades in cells, followed by modifications in their response without extensive changes in the integrin repertoire on their surface. A challenging task for the future is to develop new experimental designs and strategies to identify such subtle environmentally-controlled alterations in the ECM structure and organization.
|
Materials and Methods |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
). LN-1 elastase fragments E1' and E8 were generated as described (Goodman et al., 1987). Monoclonal antibody (mAb) against the chicken ß1-integrin subunit was from K. Yamada (NIH, Bethesda, MD). Inhibitory or non-inhibitory Ab against chicken 1ß1 integrin was from M. Paulsson (University of Bern, Switzerland). Ab against chicken 6 integrin was from I. de Curtis (DIBIT, Milan, Italy). Anti-chicken FAK mAb was from Transduction Laboratories. Ab against phosphorylated FAK phosphorylated on Tyr397 (anti-FAK-pY397) was from Biosource International. Ab against the dual phosphorylated, active form of ERK was from Promega and mAb against pan ERK from Transduction Laboratories. mAb against chicken paxillin was from Chemicon, Abs against chicken -actinin and chicken talin from Sigma, and Ab against chicken vinculin from B. Geiger (Weizmann Institute, Rehovot, Israel). PD98059 was from Calbiochem.
Neural crest cell cultures and assays for cell adhesion and migration
Cultures of trunk crest cells were generated as described (Duband et al., 1995). Cell adhesion and migration assays were performed at 37°C in serum-free medium containing 0.1% ovalbumin, transferrin and insulin at 10 µg/ml as described previously (Desban and Duband, 1997), in bacteriological dishes (and not in cell culture dishes to avoid non-specific cellular adhesion) coated with LN-1 and fragments at 1-100 µg/ml. At these concentrations, LN-1 and fragments were found to adsorb to the dish with comparable efficiencies, proportional to the coating concentration, excluding the possibility of differential adsorption (Goodman et al., 1987). Video-microscopy analyses were performed in Terasaki plates as described (Dufour et al., 1988).
Determination of cell proliferation and cell death
Cell proliferation was immunohistochemically by detecting BrdU incorporation on crest cell cultures with the BrdU labeling and detection kit from Roche. Apoptosis was detected by terminal transferase dUTP nick end labeling (TUNEL) assay with the in situ cell-death detection kit from Roche.
Immunolabeling
For immunofluorescent labeling, crest cell cultures were fixed using different procedures adapted for each antibody to visualize surface, cytoskeleton-associated, cytoplasmic or nuclear antigens: (1) in chilled methanol for 5 minutes followed by chilled acetone for 1 minute; (2) in 3.7% formaldehyde-5% sucrose in PBS for 1 hour followed by permeabilization with 0.2% Triton X-100 in PBS for 3 minutes; (3) in 3.7% formaldehyde-0.5% Triton X-100-5% sucrose in PBS for 5 minutes followed by a 1-hour incubation in 3.7% formaldehyde in PBS; and (4) in 2% paraformaldehyde in PBS for 30 minutes followed by a 20-minute incubation in 0.5% Triton X-100. Staining was revealed with appropriate fluorescein-conjugated secondary antibodies (Southern Biotechnology Associates) or with biotinylated secondary antibodies and cyanin-3-conjugated streptavidin (NEN).
Immunoblotting
For immunoblotting, crest cells obtained from at least 100 neural tubes cultured for 18 hours were plated on dishes coated with adhesive proteins. Cells were harvested with a solution of 0.01% trypsin, followed by addition of 0.01% soybean trypsin inhibitor (Sigma). Cells were collected by sedimentation, resuspended in Tris buffer, counted and lysed under reducing conditions at 90°C with SDS sample buffer. Lysates were subjected to SDS-PAGE in Laemmli buffer system on slab 7.5% polyacrylamide gels and electroblotted for 1.5 hours onto nitrocellulose. The nitrocellulose membranes were then saturated with 1% BSA in Tris buffer, 0.05% Tween-20, incubated with the primary antibodies for 12 hours at 4°C, then with peroxidase-conjugated goat anti-rabbit or goat anti-mouse IgG for 30 minutes. Proteins were visualized with enhanced chemo-luminescence (Amersham).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
Aberdam, D., Aguzzi, A., Baudouin, C., Galliano, M.-F., Ortonne, J.-P. and Meneguzzi, G. (1994). Developmental expression of nicein adhesion protein (laminin-5) subunits suggests multiple morphogenetic roles. Cell Adhes. Commun. 2, 115-129.[Medline]
Bökel, C. and Brown, N. H. (2002). Integrins in development: moving on, responding to, and sticking to the extracellular matrix. Dev. Cell 3, 311-321.[CrossRef][Medline]
Boucaut, J.-C., Darribère, T., Poole, T. J., Aoyama, H., Yamada, K. M. and Thiery, J. P. (1984). Biological active synthetic peptides as probes of embryonic development: a competitive peptide inhibitor of fibronectin function inhibits gastrulation in amphibian embryos and neural crest cell migration in avian embryos. J. Cell Biol. 99, 1822-1830.
Breau, M. A., Pietri, T., Eder, O., Blanche, M., Brakebusch, C., Fässler, R., Thiery, J. P. and Dufour, S. (2006). Lack of ß1 integrins in enteric neural crest cells leads to a Hirschsprung-like phenotype. Development 133, 1725-1734.
Bronner-Fraser, M. (1986). An antibody to a receptor for fibronectin and laminin perturbs cranial neural crest development in vivo. Dev. Biol. 117, 528-536.[CrossRef][Medline]
Bronner-Fraser, M., Artinger, M., Muschler, J. and Horwitz, A. F. (1992). Developmentally regulated expression of 6 integrin in avian embryos. Development 115, 197-211.[Abstract]
Calof, A. L., Campanero, M. R., O'Rear, J. J., Yurchenco, P. D. and Lander, A. D. (1994). Domain-specific activation of neuronal migration and neurite outgrowth-promoting activities of laminin. Neuron 13, 117-130.[CrossRef][Medline]
Cho, S. Y. and Klemke, R. L. (2000). Extracellular-regulated kinase activation and CAS/Crk coupling regulate cell migration and suppress apoptosis during invasion of the extracellular matrix. J. Cell Biol. 149, 223-236.
Coles, E. G., Gammill, L. S., Miner, J. H. and Bronner-Fraser, M. (2006). Abnormalities in neural crest cell migration in laminin 5 mutant mice. Dev. Biol. 289, 218-228.[CrossRef][Medline]
Colognato, H. and Yurchenco, P. D. (2000). Form and function: the laminin family of heterodimers. Dev. Dyn. 218, 213-234.[CrossRef][Medline]
Colognato-Pyke, H., O'Rear, J. J., Yamada, Y., Carbonetto, S., Cheng, Y.-S. and Yurchenco, P. D. (1995). Mapping of network-forming, heparing-binding, and 1ß1 integrin-recognition sites within the -chain short arm of laminin-1. J. Biol. Chem. 270, 9398-9406.
Crump, J. G., Swartz, M. E. and Kimmel, C. B. (2004). An integrin-dependent role of pouch endoderm in hyoid cartilage development. PLoS Biol. 2, 1432-1445.
Danen, E. H. J., Sonnenweld, P., Sonnenberg, A. and Yamada, K. M. (2000). Dual stimulation of Ras/mitogen-activated protein kinase and RhoA by cell adhesion to fibronectin supports growth factor-stimulated cell cycle progression. J. Cell Biol. 151, 1413-1422.
de Arcangelis, A. and Georges-Labouesse, E. (2000). Integrins and ECM functions: role in vertebrate development. Trends Genet. 16, 389-395.[CrossRef][Medline]
Delannet, M., Martin, F., Bossy, B., Cheresh, D. A., Reichardt, L. F. and Duband, J.-L. (1994). Specific roles of the Vß1, Vß3 and Vß5 integrins in avian neural crest cell adhesion and migration on vitronectin. Development 120, 2687-2702.
DeMali, K. A., Wennerberg, K. and Burridge, K. (2003). Integrin signaling to the actin cytoskeleton. Curr. Opin. Cell Biol. 15, 572-582.[CrossRef][Medline]
Desban, N. and Duband, J.-L. (1997). Avian neural crest cell migration on laminin: interaction of the 1ß1 integrin with distinct laminin-1 domains mediates different adhesive responses. J. Cell Sci. 110, 2729-2744.[Abstract]
Duband, J.-L. and Thiery, J. P. (1987). Distribution of laminin and collagens during avian neural crest development. Development 101, 461-478.[Abstract]
Duband, J.-L., Dufour, S., Yamada, S. S., Yamada, K. M. and Thiery, J. P. (1991). Neural crest cell locomotion induced by antibodies to ß1 integrins: a tool for studying the roles of substratum molecular avidity and density in migration. J. Cell Sci. 98, 517-532.
Duband, J.-L., Belkin, A. M., Syfrig, J., Thiery, J. P. and Koteliansky, V. E. (1992). Expression of 1 integrin, a laminin-collagen receptor, during myogenesis and neurogenesis in the avian embryo. Development 116, 585-600.[Abstract]
Duband, J.-L., Delannet, M. and Monier, F. (1995). Neural crest cells: strategies to generate lineage diversification in vitro. In Neural Cell Culture: A Practical Approach. Vol. 163 (ed. J. Cohen and G. P. Wilkin), pp. 133-152. Oxford: Oxford University Press.
Dufour, S., Duband, J.-L., Humphries, M. J., Obara, M., Yamada, K. M. and Thiery, J. P. (1988). Attachment, spreading, and locomotion of avian neural crest cells are mediated by multiple adhesion sites on fibronectin molecules. EMBO J. 7, 2661-2671.[Medline]
Erickson, C. A. and Perris, R. (1993). The role of cell-cell and cell-matrix interactions in the morphogenesis of the neural crest. Dev. Biol. 159, 60-74.[CrossRef][Medline]
ffrench-Constant, C. and Colognato, H. (2004). Integrins: versatile integrators of extracellular signals. Trends Cell Biol. 14, 678-686.[CrossRef][Medline]
Fincham, V. J., Frame, M., Frame, M. C. and Winder, S. J. (2000). Active Erk/MAP kinase is targeted to newly forming cell-matrix adhesions by integrin engagement and v-src. EMBO J. 19, 2911-2923.[CrossRef][Medline]
Forsberg, E., Paulsson, M., Timpl, R. and Johansson, S. (1990). Characterization of a laminin receptor on rat hepatocytes. J. Biol. Chem. 265, 6376-6381.
Gammill, L. S. and Bronner-Fraser, M. (2002). Genomic analysis of neural crest induction. Development 129, 5731-5741.[CrossRef][Medline]
Geiger, B., Bershadsky, A., Pankov, R. and Yamada, K. M. (2001). Transmembrane extracellular matrix-cytoskeleton crosstalk. Nat. Rev. Mol. Cell Biol. 2, 793-805.[CrossRef][Medline]
Giancotti, F. G. and Ruoslahti, E. (1999). Integrin signaling. Science 285, 1028-1032.
Giancotti, F. G. and Tarone, G. (2003). Positional control of cell fate through joint integrin/receptor protein kinase signaling. Annu. Rev. Cell Dev. Biol. 19, 173-206.[CrossRef][Medline]
Goodman, S. L., Deutzmann, R. and Von der Mark, K. (1987). Two distinct cell-binding domains in laminin can independently promote nonneuronal cell adhesion and spreading. J. Cell Biol. 105, 589-598.
Gu, J., Tamura, M., Pankov, R., Danen, E. H. J., Takino, T., Matsumoto, K. and Yamada, K. M. (1999). Shc and FAK differentially regulate cell motility and directionality modulated by PTEN. J. Cell Biol. 146, 389-403.
Haack, H. and Hynes, R. O. (2001). Integrin receptors are required for cell survival and proliferation during development of the peripheral glial lineage. Dev. Biol. 233, 38-55.[CrossRef][Medline]
Hemler, M. E. (2003). Tetraspanin proteins mediate cellular penetration, invasion, and fusion events and define a novel type of membrane microdomain. Annu. Rev. Cell Dev. Biol. 19, 397-422.[CrossRef][Medline]
Hocking, D. C., Sottile, J. and McKeown-Longo, P. J. (1998). Activation of distinct 5ß1-mediated signaling pathways by fibronectin's cell adhesion and matrix assembly domains. J. Cell Biol. 141, 241-253.
Howe, A. K., Aplin, A. E. and Juliano, R. L. (2002). Anchorage-dependent Erk signaling: mechanisms and consequences. Curr. Opin. Genet. Dev. 12, 30-35.[CrossRef][Medline]
Hynes, R. and Zhao, Q. (2000). The evolution of cell adhesion. J. Cell Biol. 150, F89-F95.[CrossRef][Medline]
Hynes, R. O. (1992). Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69, 11-25.[CrossRef][Medline]
Kil, S. H., Lallier, T. and Bronner-Fraser, M. (1996). Inhibition of cranial neural crest adhesion in vitro and migration in vivo using integrin antisens oligonucleotides. Dev. Biol. 179, 91-101.[CrossRef][Medline]
Klemke, R. L., Cai, S., Gianini, A. L., Gallagher, P. J., de Lanerolle, P. and Cheresh, D. A. (1997). Regulation of cell motility by mitogen-activated protein kinase. J. Cell Biol. 137, 481-492.
Le Douarin, N. M. and Kalcheim, C. (1999). The Neural Crest (2nd edn). New York: Cambridge University Press.
Liu, S., Thomas, S. M., Woodside, D. G., Rose, D. M., Kiosses, W. B., Pfaff, M. and Ginsberg, M. H. (1999). Binding of paxillin to 4 integrins modifies integrin-dependent biological responses. Nature 402, 676-681.[CrossRef][Medline]
Lu, W., Miyazaki, K., Mizushima, H. and Nemoto, N. (2001). Immunohistochemical distribution of laminin-5 
2 chain and its developmental change in human embryonic and foetal tissues. Histochem. J. 33, 629-637.[CrossRef][Medline]
Miner, J. H., Cunningham, J. and Sanes, J. R. (1998). Roles for laminin in embryogenesis: exencephaly, syndactyly and placentopathy in mice lacking the laminin 5 chain. J. Cell Biol. 143, 1713-1723.
Mitra, S. K., Hanson, D. A. and Schlaepfer, D. D. (2005). Focal adhesion kinase: in command and control of cell motility. Nat. Rev. Mol. Cell Biol. 6, 56-68.[CrossRef][Medline]
Miyamoto, S., Akiyama, S. and Yamada, K. M. (1995). Synergistic roles for receptor occupancy and aggregation in integrin transmembrane function. Science 267, 883-885.
Monier-Gavelle, F. and Duband, J.-L. (1997). Cross-talk between adhesion molecules: control of N-cadherin activity by intracellular signals elicited by ß1 and ß3 integrins in migrating neural crest cells. J. Cell Biol. 137, 1663-1681.
Newgreen, D. F., Gibbins, I. L., Sauter, J., Wallenfels, B. and Wütz, R. (1982). Ultrastructural and tissue-culture studies on the role of fibronectin, collagen and glycosaminoglycans in the migration of neural crest cells in the fowl embryo. Cell Tissue Res. 221, 521-549.[Medline]
Parsons, J. T. (2003). Focal adhesion kinase: the first ten years. J. Cell Sci. 116, 1409-1416.
Perris, R. and Perissinotto, D. (2000). Role of the extracellular matrix during neural crest cell migration. Mech. Dev. 95, 3-21.[CrossRef][Medline]
Perris, R., Paulsson, M. and Bronner-Fraser, M. (1989). Molecular mechanisms of avian neural crest cell migration on fibronectin and laminin. Dev. Biol. 136, 222-239.[CrossRef][Medline]
Pietri, T., Eder, O., Breau, M. A., Topliko, P., Blanche, M., Brakebusch, C., Fässler, R., Thiery, J. P. and Dufour, S. (2004). Conditional ß1-integrin gene deletion in neural crest cells causes severe developmental alterations of the peripheral nervous system. Development 131, 3871-3883.
Rogers, S. L., Edson, K. J., Letourneau, P. C. and McLoon, S. C. (1986). Distribution of laminin in the developing peripheral nervous system of the chick. Dev. Biol. 113, 429-435.[CrossRef][Medline]
Rousselle, P. and Aumailley, M. (1994). Kalinin is more efficient than laminin in promoting adhesion of primary keratinocytes and some other epithelial cells and has a different requirement for integrin receptors. J. Cell Biol. 125, 205-214.
Sanchez-Aparicio, P., Dominguez-Jiménez, C. and Garcia-Pardo, A. (1994). Activation of the 4ß1 integrin through the ß1 subunit induces recognition of the RGDS sequence in fibronectin. J. Cell Biol. 126, 271-279.
Testaz, S. and Duband, J.-L. (2001). Central role of the 4ß1 integrin in the coordination of avian truncal neural crest adhesion, migration, and survival. Dev. Dyn. 222, 127-140.[CrossRef][Medline]
Testaz, S., Delannet, M. and Duband, J.-L. (1999). Adhesion and migration of avian neural crest cells on fibronectin require the cooperating activites of multiple integrins of the ß1 and ß3 families. J. Cell Sci. 112, 4715-4728.[Abstract]
Tosney, K. W., Dehnbostel, D. B. and Erickson, C. A. (1994). Neural crest cells prefer the myotome's basal lamina over the sclerotome as a substratum. Dev. Biol. 163, 389-406.[CrossRef][Medline]
Wary, K. K., Mainiero, F., Isakoff, S., Marcantonio, E. E. and Giancotti, F. G. (1996). The adaptator protein Shc couples a class of integrins to the control of cell cycle progression. Cell 87, 733-743.[CrossRef][Medline]
Wary, K. K., Mariotti, A., Zurzolo, C. and Giancotti, F. G. (1998). A requirement for caveolin-1 and associated kinase fyn in integrin signaling and anchorage-dependent cell growth. Cell 94, 625-634.[CrossRef][Medline]
Xiao, T., Takagi, J., Coller, B. S., Wang, J. H. and Springer, T. A. (2004). Structural basis for allostery in integrins and binding to fibrinogen-mimetic therapeutics. Nature 432, 59-67.[CrossRef][Medline]
Yang, J. T., Rayburn, H. and Hynes, R. O. (1993). Embryonic mesodermal defects in 5 integrin-deficient mice. Development 119, 1093-1105.[Abstract]
Zhang, F., Tom, C. C., Kugler, M. C., Ching, T.-T., Kreidberg, J. A., Wei, Y. and Chapman, H. A. (2003). Distinct ligand binding sites in integrin 3ß1 regulate matrix adhesion and cell-cell contact. J. Cell Biol. 163, 177-188.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Otis, S. Campbell, M. D Payet, and N. Gallo-Payet Expression of extracellular matrix proteins and integrins in rat adrenal gland: importance for ACTH-associated functions J. Endocrinol., June 1, 2007; 193(3): 331 - 347. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
