MICROSCOPY RESEARCH AND TECHNIQUE 41:416–430 (1998) Localization of Focal Adhesion Kinase in Differentiating Schwann Cell/Neuron Cultures CRISTINA FERNANDEZ-VALLE,1, 2* PATRICK M. WOOD,3 AND MARY BARTLETT BUNGE3 1Department of Molecular Biology and Microbiology and Center for Diagnostics and Drug Development, University of Central Florida, Orlando, Florida 32816 2Orlando Regional Healthcare System/Health Research Institute, Orlando, Florida 32806 3The Chambers Family Laboratory of Electron Microscopy, The Miami Project to Cure Paralysis and the Departments of Neurological Surgery and Cell Biology and Anatomy, University of Miami School of Medicine, Miami, Florida 33136 KEY WORDS Schwann cell; basal lamina; myelination; focal adhesion kinase; immunogold localization ABSTRACT Previous studies have shown that Schwann cells (SCs) differentiate into myelinforming or ensheathing cells only under conditions which allow the deposition of basal lamina and extracellular collagen [Bunge (1993) Peripheral Neuropathy, pp. 299–316]. SC adhesion to basal lamina is mediated by b1 integrins and function blocking antibodies to b1 integrins inhibit myelination [Fernandez-Valle et al. (1993) Development 119:867–880]. Recently, focal adhesion kinase (FAK), a cytoplasmic non-receptor tyrosine kinase, was found to mediate b1 integrindependent signalling in a variety of cultured cell types adhering to ECM components such as fibronectin [reviewed in Schwartz et al. (1995) Ann. Rev. Cell Biol. 11:549–599; Ilic et al. (1997) J. Cell Sci. 110:401–407]. In the present study, we have determined more precisely the respective time courses of ECM deposition and myelination. In addition, we have studied by immunocytochemistry, immuno-gold labelling, and electron microscopy the expression and subcellular localization of FAK in nondifferentiating SCs and in SCs differentiating into myelinating cells. We show that the development of basal lamina and extracellular collagen fibrils precedes by 3 days the appearance of the first myelin sheaths. FAK was detected by immunocytochemistry or immuno-gold labelling only in SCs differentiating in the presence of ascorbic acid. Localization of FAK to the abaxonal plasma membrane was dependent upon ECM deposition. Cytochalasin D did not prevent or disrupt localization of FAK to the plasma membrane. These data support the possibility that FAK acts as an intermediate in the pathway by which basal lamina regulates SC differentiation. Microsc. Res. Tech. 41:416–430, 1998. r 1998 Wiley-Liss, Inc. INTRODUCTION Extracellular matrix (ECM) is a rich, gelatinous network of proteins and proteoglycans that cushions cells, provides information about the environment, forms a reservoir of trophic factors, and regulates cell behavior by binding and activating cell surface receptors. In peripheral nerve, Schwann cells (SCs) and the axons they ensheathe or myelinate are surrounded by an ECM consisting of basal lamina and collagen fibrils. During development, assembly and adhesion to basal lamina regulates SC differentiation in response to axons. The regulatory role of ECM is clearly seen during differentiation of Schwann cells in vitro. SCs cultured with sensory axons in a serum-free or serumrich (15%) medium contact axons but remain round and undifferentiated (Moya et al., 1980; Fernandez-Valle et al., 1993). Addition of ascorbate (50 µg/ml) to the serum-containing medium induces SC differentiation as evidenced by changes in cell shape and expression of P0 mRNA that occurs within 2–5 days of ascorbate supplementation (Eldridge et al., 1987; reviewed in Bunge, 1993; Fernandez-Valle et al., 1993). Ascorbate is essential for hydroxylation of proline and lysine residues in procollagen polypeptides. This post-translational modification allows procollagen molecules to r 1998 WILEY-LISS, INC. form triple helical structures that resist thermal and proteolytic degradation when the mature collagen molecule is secreted into the extracellular space for selfassembly into collagen fibrils. Evidence that this is the mechanism by which ascorbate promotes SC differentiation is provided by work using cis-hydroxyproline and biochemical analysis of type IV collagen synthesized by SCs grown in medium with and without ascorbate (Eldridge et al., 1988, 1989). This work showed that SC/axon interactions in the absence of ECM are not sufficient to stimulate SC differentiation and suggests that secondary regulatory or signaling mechanisms are initiated by adhesion to ECM. LAMININ AND b1 INTEGRINS PARTICIPATE IN INITIATION OF SCHWANN CELL DIFFERENTIATION How does adhesion to basal lamina influence SC differentiation? Basal lamina is composed of many *Correspondence to: Department of Molecular Biology and Microbiology, Biology Bldg. Rm. 306, University of Central Florida, Orlando, FL 32816–2360. E-mail: email@example.com Contract grant sopnsor: PHS; Contract grant numbers: 34499, 09923; Contract grant sponsor: The Miami Project to Cure Paralysis; Contract grant sponsor: State of Florida. Received 4 December 1997; Accepted 12 December 1997 FAK LOCALIZATION IN DIFFERENTIATING SCHWANN CELLS well-characterized molecules including collagen type IV, laminin, entactin or nidogen, and heparan sulfate proteoglycan. Various types of cell surface receptors bind individual ECM components and mediate attachment to and signaling from the ECM. The identity of the critical basal lamina component responsible for signaling to the SC interior was investigated by adding Matrigel (a commercial basal lamina preparation) or purified basal lamina components to ascorbate-deficient medium in an effort to promote myelination. Either Matrigel or purified laminin alone stimulated myelination but collagen and heparan sulfate proteoglycan did not (Carey et al., 1986; Eldridge et al,. 1989; Guenard et al., 1995). This suggests that laminin is the component of basal lamina that induces or regulates myelination. The integrin family is the largest and most studied group of ECM receptors. Integrins bind specific ECM ligands and are capable of initiating signaling cascades that regulate cell behavior (Clark and Brugge, 1995; Juliano and Haskill, 1993; Kornberg and Juliano, 1992; Richardson and Parsons, 1995; Schwartz et al., 1995). Integrins are heterodimeric receptors consisting of alpha and beta subunits. The possibility of many different combinations of subunits selected from 14 alpha and 9 beta subunits provides a large repertoire of ligand specificities for these receptors. For instance, beta 1 integrin paired with alpha 5 binds fibronectin; beta 1 paired with alpha 6 binds laminin; and beta 1 paired with alpha 1 binds both collagen and laminin. The major integrin receptors expressed by undifferentiated and differentiating SCs are a6b1 integrin, a laminin receptor, a1b1 integrin, a dual collagen/ laminin receptor, av, b3, and b8 integrins (Einheber et al., 1993; Feltri et al., 1994; Fernandez-Valle, et al., 1994; Milner et al., 1997; Niessen et al., 1994; Sonnenberg et al., 1990). When SCs are grown with neurons in ascorbate-rich medium containing function-blocking b1 integrin polyclonal antibody, basal lamina is assembled but is not bound to the SC surface and SCs fail to myelinate axons (Fernandez-Valle et al., 1994). Sister cultures grown in the presence of function-blocking a1b1 monoclonal antibody myelinate normally. This result is consistent with the hypothesis that laminin binding to b1 integrin induces a signaling cascade that ‘‘primes’’ the SC to respond to instructive axonal signals that specify the path of differentiation. Myelinating SCs down-regulate a6b1 integrin and replace it with another integrin laminin receptor, a6b4 integrin. a6b1 integrin has recently been shown to promote much of the migration of mouse SCs on laminin (Milner et al., 1997). FOCAL ADHESION KINASE TRANSDUCES SIGNALS FROM b1 INTEGRINS IN OTHER CELL TYPES Our knowledge of the immediate molecular consequences of b1 integrin activation is rapidly increasing, and yet how these molecular events coordinate changes in cell shape and gene expression during development remain unclear. It is now recognized that b1 integrins, upon binding their native ligands, initiate signal transduction cascades that affect cellular behavior and gene expression in many cell types (Boudreau et al., 1995; Juliano and Haskill, 1993; Kornberg et al., 1991; Streuli 417 et al., 1991; Werb et al., 1989; Yurochko et al., 1992). b1 integrins bind ECM ligands, aggregate, and induce tyrosine phosphorylation of associated cytoplasmic proteins and recruitment of many signaling and structural proteins to the plasma membrane at specialized sites of ECM adhesion known as focal adhesions (Burridge et al., 1988; Burridge and Chrzanowska-Wodnicka, 1996; Jockusch et al., 1995). Focal adhesion kinase (FAK, pp125) is a non-receptor tyrosine kinase identified as the major protein tyrosine phosphorylated in response to integrin activation in platelets and fibroblasts (Ilic et al., 1997; Lipfert et al., 1992; Schaller et al., 1992). There is indirect evidence that FAK is involved in the transduction of signals from integrin receptors, which, unlike many growth factor receptors, do not contain inherent kinase domains. This evidence derives from in vitro binding studies using peptides that mimic b1 cytoplasmic domains and extracts of cells over-expressing FAK (Schaller et al., 1995). FAK-b1 integrin interactions, however, have not been demonstrated in whole cells adhering to ECM in vitro or in cells undergoing differentiation. Recent studies have identified numerous proteins that bind to FAK and participate in down-stream signaling events (Schlaepfer et al., 1994; Miyamoto et al., 1995). These include the src kinase family (e.g., as fyn kinase) and regulatory or docking proteins such as Cas and paxillin (Hanks and Polte, 1997). Recruitment of additional proteins such as Grb2 to the focal adhesion complex allows interaction with the well known ras/mitogen-activated kinase pathway (Schlaepfer et al., 1994; Miyamoto et al., 1996; Schlaepfer and Hunter, 1996). Therefore, FAK is likely to play important roles in regulating growth and differentiation of many cell types (Hanks and Polte, 1997; Ilic et al., 1997). ROLE FOR F-ACTIN IN b1 INTEGRIN-DEPENDENT SIGNALING MECHANISMS The actin cytoskeleton provides the immature cell with motility and the mature cell with structural integrity. Actin is a dynamic structure that, through interactions with other proteins, forms an organizational network determining the localization of signal transduction molecules (Carraway and Carraway, 1995; Mochly-Rosen, 1995) that bind to cell surface receptors involved in regulating cellular growth and differentiation. Recruitment of F-actin to the juxtamembrane focal adhesion complex occurs during b1 integrinmediated signal transduction via this complex. b1 integrins interact with the actin cytoskeleton and this interaction appears to be an integral part of signaling (Lipfert et al., 1992; Otey et al., 1990; Schaller et al., 1992; Shaw et al., 1990). In all cell types studied, disruption of F-actin by cytochalasin D (CD) inhibits tyrosine phosphorylation of FAK, an early and key step in b1 integrin-dependent signal transduction that triggers recruitment of additional signaling molecules to the focal adhesion complex (Lipfert et al., 1992; Miyamoto et al., 1995; Richardson and Parsons, 1995; Schaller et al., 1992; Schaller, 1996). F-actin, however, is not required for FAK-b1 integrin co-localization (Miyamoto et al., 1995). When a low CD concentration is added to SC/neuron cultures, SCs assemble and adhere to basal lamina, 418 C. FERNANDEZ-VALLE ET AL. elongate, and engulf axons, but do not myelinate or express myelin specific mRNAs encoding cyclic nucleotide phosphodiesterase, MAG, and P0 (Fernandez-Valle et al., 1997). This finding suggests that the initial changes in cell morphology indicative of SC differentiation can be uncoupled from changes in gene expression. A possible mechanism for this effect is the disruption, by CD, of FAK activation as explained above. As an initial test of this possibility we have studied the expression and co-localization of FAK in SCs cocultured with neurons in undifferentiating and differentiating conditions and in the presence of low doses of CD. A striking increase in FAK imunostaining was observed when SC/neuron cultures were treated with ascorbate to induce differentiation. Ultrastructural immunogold staining was employed to determine the localization of FAK in differentiating SC/neuron cocultures. Gold labelling was observed in the cytosol and juxtamembrane at sites of basal lamina adhesion in differentiating and myelinating SCs and in CD-treated SCs that form basal lamina but do not myelinate (this work has appeared in abstract form, Fernandez-Valle, 1996.) MATERIALS AND METHODS Antibodies Two FAK antibodies, a mouse monoclonal (catalog no. 05–182, Upstate Biotechnology, Inc., UBI, Lake Placid, NY) and a rabbit polyclonal (catalog no. sc557, Santa Cruz Biotechnology, Inc., SCBI, Santa Cruz, CA) were used for immunocytochemistry and immuno-gold labelling. Tissue Culture Primary SC Cultures. The Brockes et al. (1979) method was used to isolate SCs from sciatic nerves removed from E21 or newborn Sprague-Dawley rats (Charles River, Raleigh, NC). SCs were expanded in vitro in medium containing Dulbecco’s modified Eagle’s medium (DMEM), 10% heat inactivated fetal bovine serum (FBS; Gibco/BRL; Grand Island, NY), forskolin (Sigma; St. Louis, MO), and pituitary extract (BTI; Stoughton, MA) on poly-L-lysine (Sigma) coated 100 mm tissue culture dishes. SC cultures were passaged no more than four times before the SCs were plated onto sensory neuron cultures. Sensory Neuron Cultures. Neurons were isolated from cervical dorsal root ganglia of Sprague-Dawley rat embryos at 15 days of gestation by dissociation with trypsin. Cells were plated on poly-L-lysine and laminincoated (Gibco/BRL) glass coverslips (Carolina, Burlington, NC) and maintained in Eagles’ minimum essential medium (EMEM) containing human placental serum (a generous gift from Dr. R. Devon, Saskatoon, Canada) or FBS, nerve growth factor (NGF) and glucose as described in Fernandez-Valle et al. (l993). SCs and fibroblasts were eliminated using one pulse of the antimitotic agent, fluorodeoxyuridine (10 µm). Additional details of the culture procedure are provided in Kleitman et al. (1991). Neuron-only cultures were maintained 7–14 days in serum-containing medium (EMEM, 10% FBS) before immunostaining for FAK. SC/Neuron Cultures. SCs were removed from culture dishes using trypsin, washed extensively in L-15 containing 10% serum to inactivate trypsin, and rinsed in L-15 (Gibco/BRL). Approximately 30,000 SCs were seeded onto purified neuron cultures and maintained for 1 week in serum-only medium to allow the cultures to become fully populated with SCs (Fernandez-Valle et al., 1994). Cultures were switched from serum-only medium to serum-plus-ascorbate medium (EMEM, 15% FBS, NGF, 50 µg/ml ascorbate) for 7–10 days to allow myelination. Cytochalasin D Treatment. SC/neuron cultures were grown in serum-only medium for 1 week to expand the SC population on axons. All cultures at this time have equivalent SC densities. Cultures were then fed serum-plus-ascorbate medium with or without CD (Sigma, St. Louis, MO) at 0.25 µg/ml. Medium was replenished every other day until myelin became visible in the control cultures (approximately 7–8 days later) and then cultures were processed as described below. Immunocytochemistry Cultures were processed for immunochemistry as previously described (Fernandez-Valle et al., 1997). Briefly, all cultures were rinsed in PBS, fixed in PBS/4% paraformaldehyde (PFA) for 10 minutes, and permeabilized with 0.2% Triton X-100 in PBS/4% PFA. SC/ neuron cultures grown in serum-plus-ascorbate medium were additionally permeabilized with a 50%/100%/ 50% series of acetone at 220°C to increase antibody penetration through ECM. After blocking non-specific reactions by incubating in PBS/10% normal goat serum, FAK monoclonal antibody (UBI) and/or P0 antibody (a generous gift from Dr. J. Brockes) at 1:100 dilution was added for 1 hour, then rinsed, and goat anti-mouse or rabbit secondary antibody conjugated to rhodamine or fluorescein (Organon Teknika Corp., West Chester, PA) was added for 30 minutes. Control staining was carried out by omitting the primary antibody. Cultures were rinsed, post-fixed, and mounted in an anti-fade solution (Citifluor, Canterbury, England) containing Hoechst nuclear dye 33342 (Sigma). Cultures were viewed on a Zeiss (Thornwood, NY) universal microscope. Electron Microscopy Standard Fixation and Embedding. Cultures were fixed in buffered glutaraldehyde followed by osmium tetroxide, dehydrated in ethanol, and embedded in Embed (Electron Microscopy Services, Fort Washington, PA; Ratner et al., 1986). Areas for examination were selected, thin sectioned, and stained with uranyl acetate and lead citrate. Sections were viewed with a Philips (Mahwah, NJ) CM10 electron microscope. Quantitation of Developmental Time Course. SC/neuron cultures were switched from serum-only to serum-plus-ascorbate medium to initiate differentiation into myelin-forming cells. At daily intervals thereafter, cultures were fixed and processed for electron microscopy and the percentage of SCs with myelin, basal lamina, and fibrillar collagen was determined. SC differentiation was assessed by analysis of 10 electron micrographs/time point. A minimum of 10 of the largest axons/micrograph were selected and the presence of myelin, basal lamina, and collagen fibrils associated with the SC contacting each selected axon was recorded. FAK LOCALIZATION IN DIFFERENTIATING SCHWANN CELLS Immunogold Fixation and Embedding. SC/neuron cultures were fixed and embedded in LRGold (Electron Microscopy Sciences, Fort Washington, PA) using the method reported by Gillespie et al. (1994). Briefly, cultures were immersed in 4% PFA in 0.1 M phosphate buffer (pH 7.4) containing 0.01 M sodium periodate, 0.075 M lysine, and 3% sucrose for 2 hours at room temperature. Fixed cultures were washed several times in 0.1 M phosphate buffer with 3% sucrose (rinse buffer) and stained in 0.25% tannic acid in the same buffer for 30 minutes at 4° C. Subsequent steps were carried out at 4° C. Cultures were rinsed several times in rinse buffer and then incubated for 30 minutes in rinse buffer containing 50 mM ammonium chloride to quench aldehydes. Cultures were washed 3 times in 0.1 M maleate buffer (pH 6.2) containing 4% sucrose and then stained with 2% uranyl acetate in maleate/sucrose buffer for 30 minutes. Cultures were dehydrated in a graded ethanol series from 50 to 90% at 220°C. Cultures were infiltrated for 30 minutes in a sequential series of 1:1 ratio of LRgold:ethanol, 7:3 ratio LRgold: ethanol and, finally, LRgold with 0.05% benzoin methyl ether for 30 minutes before embedding by inverting the coverslip cultures onto a gelatin capsule filled with resin and polymerizing at -20°C for 24 hours with UV light (365 nm). Glass coverslips were removed by immersion in hydrofluoric acid. Areas of the embedded cultures were thin sectioned and collected onto nickel grids. Immunogold Staining. Grids were incubated for 30 minutes to 2 hours in a blocking solution containing 10 mM Tris pH 7.4 buffer, 500 mM sodium chloride, 1% bovine serum albumin (BSA), 0.5% fish skin gelatin, 0.05% Triton X-100, and 0.05% Tween 20 at room temperature. Primary antibodies were diluted in blocking buffer (UBI monoclonal, 1:140; SCBI polyclonal, 1:500) and grids were inverted over a 20 µl drop of anti-FAK antibody solution and incubated overnight at 4°C in a humidified chamber. The following day grids were washed several times in blocking buffer and incubated in a 1:25 dilution of 10 nm gold-conjugated goat anti-mouse or anti-rabbit secondary antibody (Ted Pella, Inc., Redding, CA) for 1 hour. Grids were washed extensively with distilled water, fixed with 2.5% glutaraldehyde in PBS for 10 minutes, washed in a stream of distilled water and then exposed to osmium vapors for 15 minutes. Grids were stained in lead citrate for 2 minutes and viewed with a Philips CM10 electron microscope. Quantitation of Gold Particles. Two to five areas from SC/neuron cultures grown in serum-only or in serum-plus-ascorbate medium with and without CD for 8 days were selected to assess immunogold labelling with FAK antibodies. Areas were randomly selected and photographed, all gold particles present in 5–25 electron micrographs of each section were counted, and the distribution was recorded. The distribution of all gold particles in multiples of 2 or more are reported. Evaluation of Nonspecific Labelling. With the monoclonal UBI antibody, 14–35% of single gold particles fell on the plastic resin and not on cellular structures. In 5 of 7 labelling experiments, clusters of two or more gold particles were not observed on the resin and, in the other two experiments, only 8 and 9% 419 of aggregates of 2 or more gold particles were observed on the plastic resin. SCBI polyclonal antibody used at 1:500 dilution resulted in low background staining and specifically labelled the same structures observed with the monoclonal antibody in PBS. Single gold particles were most often found dispersed on the plastic sections and in cultures grown in serum-only medium. Their distribution was not included in this report to reduce the possibility of including non-specifically labelled sites. Changing the blocking buffer from TBS plus BSA and detergents to PBS with 0.1% acetylated BSA, 0.1% Tween 20, and 0.05% Triton X-100 to reduce nonspecific antibody binding abolished all labelling. Additional experiments are required to carefully titrate detergent and acetylated BSA concentrations to completely eliminate nonspecific antibody binding. Both antibodies labelled SC nuclei and occasionally the nuclear envelope. Because nuclear labelling was observed in undifferentiated SC/neuron cultures, it also was considered non-specific binding of the primary antibodies. All labelling was abolished when primary antibody was omitted, indicating that background labelling observed was not due to the gold-conjugated secondary antibody. Structures grouped under ‘‘other’’ in Tables 1–3 included unrecognizable structures, cell surface, and collagen fibrils. RESULTS Extracellular Matrix Deposition Precedes Myelination ECM deposition, visible as basal lamina and collagen fibrils, precedes SC differentiation. This was clearly seen in the results of a 7-day ultrastructural time course study (Fig. 1). The results demonstrated that basal lamina and collagen fibril deposition occurred simultaneously and preceded the onset of myelination by 3 days. At the time of ascorbate addition (day 1), SCs had been growing in continuous contact with axons in serum-containing but ascorbate-lacking medium for 7 to 10 days; fewer than 20% of the SCs exhibited ECM and SCs had not yet myelinated axons. Ultrastructural analysis of these cultures revealed that SCs contacted axons but extended few processes into axon fascicles (Fig. 2). Previous work in our laboratories has shown that even after 3–5 weeks of co-culture with sensory neurons in serum-only medium, SCs do not differentiate. Instead, they continue to proliferate in response to axons and only a few myelin segments (0–10 myelin segments/culture) form. By 4 days after ascorbate addition to the medium, basal lamina and collagen were observed in periodic patches along the SC surface in nearly 40% of the SC/axon units (Fig. 1). Axon fascicles had begun to be subdivided by SC processes and substantial amounts of basal lamina and collagen fibrils had formed, increasing the space between axons and neighboring SCs. Typically, a number of SCs were engulfing axons and isolating them from other axons (Fig. 3). A rare thin myelin segment was observed. During the subsequent 24 hours, there was a dramatic increase in ECM deposition, so that at day 5, 80% of the axon/SC units displayed basal lamina and collagen fibrils. During this rapid increase in ECM deposition, several additional thin myelin segments appeared (Fig. 4a). Many more 420 C. FERNANDEZ-VALLE ET AL. Fig. 1. Deposition of basal lamina and collagen fibrils is concurrent and precedes myelination in Schwann cell/neuron cultures. Quantitative development time course showing the appearance of basal lamina, collagen fibrils, and myelin sheaths following addition of serum-plus-ascorbate to SC/N cultures. SCs were differentiating and were engulfing, spiraling, or in one-to-one relationships with axons. There was increased space between SCs occupied by basal lamina and collagen fibrils and large axon fascicles were less frequently observed. The number of axon/SC units with myelin increased slowly with time (Figs. 1 and 4b), reaching 20% of the total by 7 days, a time by which 90% of axon/SC units exhibited ECM. The myelin sheaths present were now of greater width than those observed at 4 days, and basal lamina covered the myelinating SC exterior, which was associated with increased fibrillar collagen. The developmental sequence of basal lamina and collagen deposition followed by initiation of myelination suggests that interaction of the SC with ECM signals or regulates further differentiation. To explore this possibility, we have studied the pattern of expression and localization of FAK during the period preceding and the period coinciding with myelination (Fernandez-Valle, 1996). FAK Is Detected by Immunostaining in Differentiating and Myelinating SCs Cultures of sensory neurons grown alone and SC/ neuron cultures grown in serum-only or serum-plusascorbate medium were immunostained with FAK antibodies to determine if SCs express FAK and whether FAK expression is dependent upon ECM deposition. Immunocytochemical analysis failed to detect FAK in the soma or neurites of dorsal root ganglion neurons cultured alone (Fig. 5a,b) or in SCs cultured with neurons in serum-only medium for 2 days (Fig. 5c,d). FAK was readily detected in the perinuclear cytoplasm and bipolar cytoplasmic extensions of differentiating SCs grown 5–8 days in serum-plus-ascorbate medium (Fig. 5e,f). FAK continued to be expressed at high levels in myelinating SCs and was present along the extent of the myelin internode (Fig. 5g,h). FAK was also observed in the perinuclear cytoplasm of non-myelinating SCs in cultures grown 21 days in serum-plus-ascorbate medium (Fig. 5i,j). Double immunostaining confirmed that FAK and P0, the major protein in peripheral myelin, were co-expressed in myelinating Ss (Fig. 6). The striking increase in FAK immunostaining that occurs following ascorbate addition suggests that FAK expression might be induced or up-regulated by ascorbate. However, in a companion study, Western blot analysis was used to verify FAK expression. Using this technique and a different FAK antibody suitable for immunoblot analysis of denatured protein, FAK was found to be expressed in SCs cultured alone, and in undifferentiated SC/neuron cultures growing in serumonly medium, as well as in SC/neuron cultures grown 4, 7, 14, and 21 days in serum-plus-ascorbate medium (Chen and Fernandez-Valle, unpublished data). FAK was not detected in sensory neuron cultures, confirming the immunostaining results. The Western blot results indicated, therefore, that FAK was expressed at all stages of SC development in vitro and its expression was not dependent upon ECM deposition or axonal contact. FAK Is Present at Juxtamembrane Sites of Basal Lamina Adhesion in Differentiating and Myelinating Schwann Cells A comprehensive evaluation of FAK-gold labelling was carried out to determine the subcellular localization of FAK. FAK localization was studied in cultures grown in serum-only medium and in serum plus ascorbate medium. In serum-only cultures (Fig. 7a), multiple gold particles only infrequently labelled the abaxonal juxtamembrane regions or SC processes, in contrast to differentiating and CD treated cultures (Table 1). Gold labelling was never observed in sections from differentiating 8 day SC/neuron cultures reacted with only the gold-conjugated secondary antibody (Fig. 7b). FAK LOCALIZATION IN DIFFERENTIATING SCHWANN CELLS 421 Fig. 2. Schwann cells grown with sensory neurons in serum-containing but ascorbate-lacking medium do not differentiate. Fascicles of axons with few intervening SC processes are observed. Only scant basal lamina and collagen fibrils are present, allowing close apposition of neighboring SCs (arrows). X31,000. In differentiating cultures, the consistently labelled subcellular structures were the cytosol and abaxonal juxtamembrane regions of SCs and SC processes of differentiating SCs that were associated with basal lamina and collagen fibrils (Fig. 8a,b). At times, the gold particles labelled sub-plasmalemmal densities as well. Because these structures were only rarely labelled in undifferentiated SC/neuron cultures grown in serumonly medium, this FAK labelling was considered to be specific. The labelled plasma membrane domain was usually adherent to a nascent basal lamina patch. In one oblique section through a myelinating Schwann cell, a large gold aggregate was observed in the cytoplasm adjacent to plasma membrane that was apposed to collagen fibrils and, by inference, basal lamina (Fig. 8a, inset). In myelinating SCs, multiple gold particles were observed in the outer cytoplasmic collar, just exterior to the outermost myelin lamella (Fig. 8c). The distribution of all gold particles in multiples of two or more in cultures 8 days after adding differentiation medium is shown in Table 2. In Table 2, the organelle labelled by the SCBI polyclonal antibody was a clathrin- 422 C. FERNANDEZ-VALLE ET AL. Fig. 3. Schwann cells grown with sensory neurons for 4 days in serum plus ascorbate medium begin differentiation. Many large and small diameter axons are ensheathed by SCs. There are two examples (*) of a SC ensheathing only one axon, a stage prefatory to myelination. Basal lamina and collagen fibrils have appeared in relation to the SC surface (arrows). X31,000. coated vesicle associated with a 14 gold particle cluster. Occasional labelling of clathrin-coated vesicles and granular endoplasmic reticulum was observed using the UBI monoclonal antibody. The significance of this labelling is not known and more extensive experiments will be necessary to determine whether labelling of these organelles is specific. Occasional axon and myelin labelling was also observed. FAK LOCALIZATION IN DIFFERENTIATING SCHWANN CELLS 423 Fig. 4. Schwann cells begin to form thin myelin sheaths 3 to 4 days after ascorbate addition. a: Five days after ascorbate addition, a few Schwann cells elaborated myelin around the larger diameter axons, and other Schwann cells are in a 1:1 relation with axons or are ensheathing multiple small diameter axons. Increasing amounts of basal lamina and collagen fibrils are present. b: Eight days after ascorbate addition, many well-myelinated axons are present. Basal lamina and collagen fibrils completely surround the axon/Schwann cell unit. With time, progressively more Schwann cells will complete differentiation into myelinating or ensheathing cells. X23,000. FAK Is Present at Juxtamembrane Sites of Basal Lamina Adhesion in Schwann Cells Grown With Sensory Neurons in the Presence of Cytochalasin D Immunogold staining to detect FAK was also carried out in cultures grown in serum-plus-ascorbate medium containing CD that disrupts actin polymerization and myelination but does not disrupt basal lamina deposition. SC/neuron cultures grown in low CD concentrations fail to develop myelin, but assemble basal lamina and begin differentiation as assessed by changes in SC morphology, i.e., elongation and formation of the first Fig. 5. FAK LOCALIZATION IN DIFFERENTIATING SCHWANN CELLS 425 Fig. 6. FAK and P0 are co-expressed in myelinating Schwann cells. Phase (A) and fluorescent double immunostaining for FAK (B) and P0 r in SC/neuron cultures grown 21 days in serum plus ascorbate medium. The majority of P0-positive SCs express FAK that is localized along the entire length of the myelin internode. X400. SC membrane spiral around axons (Fernandez-Valle et al., 1997). In these cultures gold particle labelling of elongated SCs was frequently observed. Triplets of gold particles labelled the abaxonal juxtamembranous regions of SCs engulfing axons or in the process of elongation (Fig. 9). The labelled plasma membrane domains adhered to basal lamina patches, which were in turn associated with collagen fibrils. Subplasmalemmal densities in SCs were also frequently labelled by multiple gold particles (Fig. 9, inset). The distribution of gold particles is shown in Table 3. As in SC/neuron cultures grown in the absence of CD, the predominant structures labelled in CD-treated SCs were the cytosol, cellular processes and abaxonal juxtamembranous regions of elongating or spiraling SCs. DISCUSSION The results reported here provide new clues about the role of ECM in regulating SC differentiation. These results include: (1) the demonstration that basal lamina and collagen fibrils form at the same time, 3 days preceding myelination; (2) the demonstration that SCs treated with ascorbate exhibit strongly increased immunostaining for FAK; (3) the demonstration that differentiating SCs exhibit cytosolic and juxtamembrane localization of FAK; and (4) the demonstration that myelinating SCs display FAK localization in the outer cytoplasmic collar of the myelin sheath. Fig. 5. FAK is expressed in differentiating and myelinating Schwann cells. Phase (a,c,e,g,i) and fluorescent (b,d,f,h,j,) images of FAK immunostaining in neuron-only cultures (a,b) and in SC/neuron cultures grown 2 days in serum-only medium (c,d) or in serum-plusascorbate medium for 5 days (e,f) and 21 days (g–j). Neurons grown alone and SCs grown with neurons for 2 days in serum-only medium do not express FAK. SCs express FAK after 5–7 days of growth in serum-plus-ascorbate medium (single SC at arrow in f). In myelinating SCs, FAK staining is present along the myelin internode (arrow in h) and, in non-myelinating SCs, FAK is observed in the perinuclear area (arrow in j). a–d, X1,000; e–j, X600. Whereas there is compelling evidence that basal lamina deposition regulates SC myelination, a quantitative time course study to determine the chronological relationship between the appearance of basal lamina and myelin sheaths had not been done. Additionally, the temporal development of collagen fibrils in the culture system had not been studied previously. The results of this study demonstrated that basal lamina and collagen fibrils form at the same time, and that their appearance begins within 1 day of ascorbate addition to the culture medium. Initially, ECM appears as intermittent patches along the SC surface and with time more SCs relate to ECM and the ECM becomes more complete. There is a 3-day delay between initiation of ECM deposition and the appearance of the first myelin segments. ECM assembly clearly precedes myelination; this sequence of developmental stages is consistent with the hypothesis that ECM assembly functions to regulate steps of SC differentiation that occur later. FAK Expression in Schwann Cells Immunostaining results suggested that FAK was not expressed by sensory neurons or by SCs cultured with sensory neurons in serum-only medium that does not support myelination. Immunoblot analysis of culture extracts confirmed that sensory neurons do not express FAK, but indicated that FAK is present in SC cultures and in SC/neuron cultures grown in serum-only medium (Chen and Fernandez-Valle, unpublished data). Our interpretation of these results is that undifferentiated SCs express FAK when in contact with neurons, but that the antigenic epitope recognized by the antibodies used for immunostaining is masked in some way. FAK was detected by Western blot analysis using a different antibody recognizing denatured FAK. The time course for the appearance of FAK immunostaining in SC/neuron cultures following addition of ascorbate to serum-only medium parallels the onset of SC differentiation and suggests that the FAK epitopes become 426 C. FERNANDEZ-VALLE ET AL. Fig. 7. a: FAK labelling is not observed in Schwann cells grown with sensory neurons in serum-only medium. Single gold particles are dispersed within SC cytoplasm and nucleus but specific multiples of gold are not observed adjacent to plasma membrane (b). Photomicro- graph of an elongating SC grown in serum-plus-ascorbate medium for 8 days and incubated with only the secondary gold conjugated antibody. No gold labelling is observed in the absence of the FAK antibody. X45,000. unmasked as SCs deposit basal lamina. This ‘‘unmasking’’ could result from a change in FAK localization or association with other cytoplasmic proteins. Results of the Western blot experiments did not indicate an increase in FAK expression associated with switching SC/neuron cultures from serum-only to serum-plus ascorbate medium. The distribution of FAK immunogold labelling near the plasma membrane at sites of basal lamina adhesion in differentiating SCs is consistent with the expected 427 FAK LOCALIZATION IN DIFFERENTIATING SCHWANN CELLS TABLE 1. Distribution of gold particles ($2) in Schwann cell/neuron cultures grown 8 days in serum-containing but ascorbate-lacking medium Exp. no. 1 2 Ab Total gold/# photos Axon Cytosol SC process JxPlasma membr. Organelles Nucleus Other UBI UBI 26/11 41/8 2 (7%) 4 (10%) 2 (7%) 5 (12%) 0 8 (20%) 0 2 (5%) 0 3 (7%) 18 (69%) 6 (15%) 4 (15%) 13 (32%) Fig. 8. FAK is present in juxtamembranous sites of basal lamina adhesion in differentiating and myelinating Schwann cells. a: A large cluster of gold particles is located just within the cell at the site of basal lamina adhesion in a myelinating SC grown 8 days in serum-plusascorbate medium. This area, enclosed in the rectangle, is enlarged in the inset. Collagen fibrils, located just beyond the basal lamina, are shown in oblique section (see inset). b: In a cross-sectional view of a SC cytoplasmic process engulfing an axon, 4 gold particles are present just inside the abaxonal plasma membrane associated with ECM. Axons are occasionally labelled with single gold particles. c: Gold particles are seen in the cytoplasmic collar just outside the myelin. a,b, X55,000; c, X100,000. location of FAK in other cell types. Cultured cells form focal adhesions at sites of cell-ECM attachment through integrins (Burridge et al., 1988). At this site, structural and signalling proteins accumulate beneath the plasma membrane and initiate signalling cascades that regulate cell growth and differentiation (Jockusch et al., 1995). FAK is a cytoplasmic non-receptor tyrosine kinase that is recruited to the membrane and is rapidly phosphorylated on tyrosine residues as a result of integrin binding of ECM components such as fibronectin and laminin (Burridge et al., 1992; Schaller et al, 1992; Schaller, 1996). Previous work in this field indicates that FAK and b1 integrins co-localize at the plasma membrane and autophosphorylation of FAK 428 C. FERNANDEZ-VALLE ET AL. TABLE 2. Distribution of gold particles ($2) in Schwann cell/neuron cultures grown 8 days in serum- and ascorbate-containing medium Exp. no. 1 2 3 4 Ab Total gold/# photos Axon Cytosol SC process JxPlasma membr. Organelles Nucleus Other UBI UBI UBI SCBI 289/26 81/23 61/5 87/7 24 (8%) 0 4 (7%) 0 49 (17%) 23 (28%) 23 (38%) 25 (29%) 16 (6%) 9 (11%) 4 (7%) 8 (9%) 59 (20%) 7 (9%) 6 (10%) 10 (11%) 22 (8%) 4 (5%) 4 (7%) 14 (16%) 56 (19%) 24 (30%) 16 (26%) 18 (21%) 37 (13%) 14 (17%) 4 (7%) 5 (6%) Fig. 9. FAK is present in juxtamembranous sites of basal lamina adhesion in Schwann cells grown with sensory neurons in the presence of cytochalasin D. Frequently, three linearly arranged gold particles may be seen just within the plasma membrane of elongated SCs or SC processes engulfing or spiraling around axons. Multiple gold particles (2–7) often labeled subplasmalemmal densities at sites of basal lamina (inset). X100,000; inset, X110,000. leads to recruitment of additional signalling molecules to the focal adhesion complex (Burridge and Chrzanowska-Wodnicka, 1996; Miyamoto et al., 1995). CD inhibits FAK phosphorylation and down-stream signalling events but not the co-localization of b1 integrin and FAK (Lipfert et al., 1992; Miyamoto et al., 1995). The distribution of gold labelling is consistent with the known function of FAK. Gold was observed in the cytosol, consistent with a cytoplasmic kinase, and at the membrane at sites of adhesion to ECM. Labelling of sub-plasmalemmal domains adhering to basal lamina in CD-treated cultures is consistent with the observation that FAK-b1 integrin co-localization is maintained in cells treated with CD. The occasional labelling of organelles may represent additional sites of FAK recruitment as has been shown for other kinases such as src which is found in endosomes (Mochly-Rosen, 1995). Our findings support a role for FAK in b1 integrinmediated signalling involved in regulating SC differentiation into myelin-forming cells. FAK expression in myelinating SCs that do not express b1 integrin was surprising because, thus far, FAK has only been reported to interact with b1 integrin. FAK localization in myelinating SCs was observed at the plasma membrane, and sometimes in the outer cytoplasmic collar adjacent to the myelin lamellae. These results suggest 429 FAK LOCALIZATION IN DIFFERENTIATING SCHWANN CELLS TABLE 3. Distribution of gold particles ($2) in Schwann cell/neuron cultures grown 8 days in serum-, ascorbate-, and cytochalasin D-containing medium Exp. no. 1 2 Ab Total gold/# photos Axon Cytosol SC process JxPlasma membr. Organelles Nucleus Other UBI UBI 86/18 13/7 14 (16%) 0 20 (23%) 0 7 (8%) 3 (23%) 20 (23%) 2 (15%) 0 0 19 (19%) 5 (38%) 6 (6%) 3 (23%) that FAK has a role independent of b1 integrinmediated signalling and could be involved in maintaining the myelinating phenotype. Additional work will be necessary to determine the function of FAK in myelinating SCs. ACKNOWLEDGMENTS We thank April Bradford, a MARC summer fellow, Antje Feucht, Jason Tanner, and Marcella Schlein for assistance with experiments; Margaret L. Bates, J.-P. Brunshwig, Ernesto Cuervo, and Anna M. Gomez for technical help; Robert Camarena for photographic reproduction; and Charlaine Rowlette for word processing. We thank Dr. J. Brockes for his generous gift of Po antibody. This work was supported by PHS grants 34499 to C.F.V. and 09923 to R.P.B., The Miami Project to Cure Paralysis, and the State of Florida. REFERENCES Bailey, N., and Bissell, M. (1991) Control of mammary epithelial differentiation: Basement membrane induces tissue-specific gene expression in the absence of cell-cell interaction and morphological polarity. J. Cell Biol., 115:1383–1395. Boudreau, N., Myers, C., and Bissell, M.J. (1995) From laminin to lamina: Regulation of tissue-specific gene expression by the ECM. TICB, 5:1–4. Brockes, J. P., Fields, K. L., and Raff, M.C. (1979) Studies on cultured rat Schwann cells. I. Establishment of purified populations from cultures of peripheral nerve. Brain Res., 165:105–118. Bunge, M.B. (1993) Schwann cell regulation of extracellular matrix biosynthesis and assembly. In: Peripheral Neuropathy, 3rd ed. Dyck, P.J., Thomas, P.K., Griffin, Low, P.A., Poduslo, J., eds. Saunders, Philadelphia, pp. 299–316. Burridge, K., and Chrzanowska-Wodnicka, M. (1996). Focal adhesions, contractility and signaling. Ann. Rev. Cell Dev. Biol., 12:463–519. Burridge, K., Fath, K., Kelly, T., Nuckolls, G., and Turner, C. (1988) Focal adhesions: Transmembrane junctions between the extracellular matrix and the cytoskeleton. Ann. Rev. Cell Biol., 4:487–525. Burridge, K., Turner, C.E .and Romer, L.H. (1992). Tyrosine phosphorylation of paxillin and pp125FAK accompanies cell adhesion to extracellular matrix: a role in cytoskeletal assembly. J. Cell Biol.,119: 893–903. Carey, D.J., Todd, M.S., and Rafferty, C.M. (1986). Schwann cell myelination: Induction by exogenous basement membrane-like extracellular matrix. J. Cell Biol., 102:2254–2263. Carraway, K.L., and Carraway, C.A.C. (1995) Signaling, mitogenesis and the cytoskeleton: Where the action is. BioEssays, 17:171–175. Clark, E. A., and Brugge, J. S. (1995) Integrins and signal transduction pathways; the road taken. Science 268:233–239. Einheber, S., Milner, T. A., Giancotti, F. and Salzer, J. L. (1993) Axonal regulation of Schwann cell Integrin expression suggests a role for a6b4 in myelination. J. Cell Biol., 123:1223–1236. Eldridge, C.F.,Wood, P.M., and Bunge, R.P. (1987) Differentiation of axon-related Schwann cells in vitro. I. Ascorbic acid regulates basal lamina assembly and myelin formation. J. Cell Biol., 105:1023– 1034. Eldridge, C.F. , Bunge, R.P. and Bunge, M.B. (1988) Effects of cis-4-hydroxy-L-proline, an inhibitor of Schwann cell differentiation, on the secretion of collagenous and non-collagenous proteins by Schwann cells. Exp. Cell Res., 174:491–501. Eldridge, C. F., Bunge, M. B., and Bunge, R. P. (l989). Differentiation of axon-related Schwann cells in vitro: II. Control of myelin formation by basal lamina. J. Neurosci., 9:625–638. Feltri, M.L., Scherer, S.S., Nemni, R., Kamholz, J., Vogelbacker, H., Scott, M.O., Canal, N., Quaranta, N. and Wrabetz, L. (1994) Beta 4 integrin expression in myelinating Schwann cells is polarized, developmentally regulated and axonally dependent. Development, 120:1287–1301. Fernandez-Valle, C. (1996) Focal adhesion kinase associates with b1 integrin in differentiating Schwann cell/neuron co-cultures. 36th American Society for Cell Biology Annual Meeting, H190, p14 (Addendum). Fernandez-Valle, C., Fregien, N., Wood, P. M. and Bunge, M. B. (1993) Expression of the protein zero myelin gene in axon-related Schwann cells is linked to basal lamina formation. Development, 119: 867– 880. Fernandez-Valle, C., Gwynn, L., Wood, P. M., Carbonetto, S. and Bunge, M. B. (1994). Anti-b1 integrin antibody inhibits Schwann cell myelination. J. Neurobiol., 25:1207–1226. Fernandez-Valle, C., Gorman, D., Gomez, A.M. and Bunge, M.B. (1997) Actin plays a role in both changes in cell shape and gene-expression associated with Schwann cell myelination. J. Neurosci., 17: 241–250. Gillespie, C. S., Sherman, D. L., Blair, G.E. and Brophy, P. J. (1994) Periaxin, a novel protein of myelinating Schwann cells with a possible role in axonal ensheathment. Neuron, 12:497–508. Guenard, V., Gwynn, L.A. and Wood, P.M. (1995) Transforming growth factor-b blocks myelination but not ensheathment of axons by Schwann cells in vitro. J. Neurosci., 15: 419–428. Hanks, S.K., and Polte, T. R. (1997) Signaling through focal adhesion kinase. BioEssays, 19:137–145. Ilic, D., Furuta, Y., Sobue, K., Nakatsuji , N., Nomurs, S., Fujimoto, J., Okada, M., Yamamoto, T., and Alzawa, S. (1995) Reduced cell motility and enhanced focal adhesion contact formation in cells from FAK-deficient mice. Nature, 377:539–544 Ilic, D., Damsky, C., and Yamamoto, T. (1997) Focal adhesion kinase: at the crossroads of signal transduction J. Cell Sci., 110:401–407. Jockusch, B. M., Bubeck , P., Giehl, K., Kroemker, M. , Moschner, J., Rothkegel, M., Rudiger, M., Schluter, K., Stanke, G., and Winkler, J. (1995) The molecular architecture of focal adhesions. Ann. Rev Cell Dev. Biol., 11:379–416. Juliano, R.L., and Haskill, S. (1993) Signal transduction from the extracellular matrix. J. Cell Biol., 120:577–585. Kleitman, N., Wood, P. M. , and Bunge, R.P. (1991) Tissue culture methods for the study of myelination. In: Culturing Nerve Cells. G. Banker and K. Goslin, eds. MIT Press, Cambridge, MA, pp. 337– 377. Kornberg, L., and Juliano, R.L. (1992) Signal transduction from the extracellular matrix: The integrin-tyrosine kinase connection. TIPS, 13:93–95. Kornberg, L.J., Earp, H.S., Turner, C.E., Prockop, C., and Juliano, R.L. (1991) Signal transduction by integrins:Iincreased protein tyrosine phosphorylation caused by clustering of b1 integrins. J. Cell Biol., 88:8392–8396. Lipfert, L., Haimovich, B., Schaller, M.D., Cobb, B.S., Parsons, J.T., and Brugge, J.T. (1992) Integrin-dependent phosphorylation and activation of the protein tyrosine kinase pp125FAK in platelets. J. Cell Biol., 119:905–912. Milner, R., Wilby, M., Nishimura, S., Boylen, K., Edwards, G., Fawcett, J., Streuli, C., Patella, R., and ffrench-Constant, C. (1997) Division of labor of Schwann cell integrins during migration on peripheral nerve extracellular matrix ligands. Dev. Biol., 185:215–228. Miyamoto, S., Teramoto, H., Coso, O., Gutkind, J., Burbelo, P. D, Akiyama, S.K. and Yamada, K. M. (1995) Integrin function: Molecular hierarchies of cytoskeletal and signaling molecules. J. Cell Biol., 131:791–805. Miyamoto, S., Teramoto, H., Gutkind, J.S. and Yamada, K. M. (1996) Integrins can collaborate with growth factors for phosphorylation of receptor tyrosine kinase and MAP kinase activation: roles of integrin aggregation and occupancy of receptors. J. Cell Biol., 135:1633– 1642. Mochly-Rosen, D. (1995) Localization of protein kinases by anchoring proteins: a theme in signal transduction. Science, 268:247–251. Moya, F., Bunge, M.B., Bunge, R.P. (1980) Schwann cells proliferate 430 C. FERNANDEZ-VALLE ET AL. but fail to differentiate in defined medium. Proc. Natl. Acad. Sci. U.S.A., 77:6902–6906. Niessen, C.M., Cremona, O., Daams, H., Ferraresi, S., Sonnenberg, A. and Marchinsio, P.C. (1994) Expression of the integrin a6b4 in peripheral nerves: Localization in Schwann cells and perineural cells and different variants of the b4 subunit. J. Cell Sci., 107:543– 552. Otey, CA, Pavalko, F.M. and Burridge, K. (1990) An interaction between a-actinin and the b1 integrin subunit in vitro. J Cell Biol., 111:721–729. Richardson, A., and Parsons, J.T. (1995) Signal transduction through integrins: A central role for focal adhesion kinase. BioEssays, 17:229–236. Schaller, M. D. (1996) The focal adhesion kinase. J. Endocrinol., 150:1–7. Schaller, M.D., Borgman, C.A., Cobb, B.S., Vines, R.R., Reynolds, A.B. and Parsons, J.T. (1992) PP 125 FAK, a structurally distinctive protein tyrosine kinase associated with focal adhesions. Proc. Natl. Acad. Sci. U.S.A., 89:5192–5196. Schaller, M.D., Otey, C.A., Hildebrand, J. D. and Parsons, J. T. (1995) Focal adhesion kinase and paxillin bind to peptides mimicking a-integrin cytoplasmic domains. J. Cell Biol., 130:1181–1187. Schlaepfer, D. D., Hanks, S. K., Hunter, T. and van der Geer, P. (1994) Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature, 372:786–791. Schlaepfer, D. D., and Hunter, T. (1996) Evidence of in vivo phosphory- lation of the Grb2 SH2 domain binding site on focal adhesion kinase by src-family protein tyrosine kinases. Mol. Cell Biol., 16:5623–5633. Schwartz, M. A., Schaller, M. D., and Ginsberg, M. H. (1995) Integrins: Emerging paradigms of signal transduction. Ann. Rev. Cell Biol., 11:549–599. Shaw, L.M., Messoer, J.M., and Mercurio, A.M. (1990) The activation dependent adhesion of macrophages to laminin involves cytoskeletal anchoring and phosphorylation of the a6b1 integrin. J. Cell Biol., 109:877–889. Sonnenberg, A., Linders, P.W., Modderman, P.W., Damsky, C.H., Aumailley, M., and Timpl, R. (1990) Integrin recognition of different cell-binding fragments of laminin (P1, E3, E8) and evidence that a6b1 but not a6b4 functions as a major receptor for fragment E8. J. Cell Biol.,110:2145–2155. Strueli, C.H., Bailey, N., and Bissell, M.J. (1991) Control of mammary epithelial differentiation: basement membrane induces tissuespecific gene expression in the absence of cell-cell interaction and morphological polarization. J. Cell Biol., 115:1383–1395. Werb, Z., Tremble, P.M., Behrendtsen, O., Crowley, E., and Damsky, C. (1989) Signal transduction through the fibronectin receptor induces collagenase and stromelysin gene expression. J. Cell Biol., 109:877–889. Yurochko, A.D., Liu, D.Y., Eirman, D., and Haskill, S. (1992) Integrins as a primary signal transduction molecule regulating monocyte immediate-early gene induction. Proc. Natl. Acad. Sci. U.S.A., 89:9034–9038.