We have characterized the diversity of the chicken β1 integrin family and studied the expression of individual receptors during development. The diversity of the β1 integrin family was investigated by affinity purifying the β1 integrins from a variety of adult and embryonic tissues. These purifications reveal the relative levels of expression and also the differential expression of the α subunits in those tissues. Monoclonal antibodies were generated against the prominent ‘band 1’ of the embryonic chicken integrins and used to characterize the expression of this α subunit in embryonic and adult tissues. This α subunit is shown to be the chicken homologue of human α5 fibronectin receptor. The chicken α5β1 integrin is the most prominent β1 integrin in the embryo and is expressed on the majority of cell types through the day 17 stage. The distribution of this receptor in the embryo closely parallels the distribution of its ligand, fibronectin. In adult tissues, expression of this receptor is greatly diminished relative to the expression of other α subunits. The cell type distribution is highly restricted: limited primarily to the vasculature and to connective tissue regions. These studies reveal a prominent role for the α5β1 integrin in embryonic cell types and a down-regulation of this receptor on many cell types during development.

The β1 integrins are a family of cell surface receptors with specificity for molecules in the extracellular matrix (ECM) (Hynes, 1987; Ruoslahti and Pierschbacher, 1987; Buck and Horwitz, 1987). These receptors also associate with the cytoskeleton via their cytoplasmic domain and therefore function as transmembrane links between the cytoskeleton and the ECM (Horwitz et al. 1986; Otey et al. 1990; Argraves et al. 1989). The integrins are αβ heterodimers formed by two non-covalently linked glycoproteins both of which have a small cytoplasmic domain, a single membrane spanning region, and a large extracellular domain (Tamkun et al. 1986; Argraves et al. 1987). Diversity in the β1 integrin family is generated by a variety of α subunit, which associate with the common β1 subunit to form unique αβ heterodimers. As the fi subunit is constant in each receptor, the ligand specificity of the receptor is conferred by the variable α subunit. The β1 integrin family is best characterized in humans where at least seven different receptors (i.e. different α subunit) have been identified. Each heterodimer appears to function as a receptor for ECM molecules. Among them are receptors for collagen types I and IV, laminin and fibronectin (for reviews see Hemler, 1990; Albelda and Buck, 1990).

Many functions have been attributed to the β1 integrin family. By mediating the interaction of the cellular cytoskeleton with the ECM, the β1 integrins are central to issues of cell migration, proliferation, differentiation, morphology, and the maintenance of tissue architecture and integrity. Antibodies against the β1 subunit reveal its presence on virtually every cell type in the animal systems studied thus far (Buck and Horwitz, 1987). Anti-β1 antibodies have been shown to perturb cell adhesion to matrix molecules (Neff et al. 1982), alter cell morphology (Greve and Gottlieb, 1982), perturb cell migration (Bronner-Fraser, 1986) and neurite outgrowth (Bozycko and Horwitz, 1986; Tomaselli et al. 1986), and inhibit cell differentiation (Menko and Boettiger, 1987; Adams and Watt, 1989). The β1 integrins are also thought to play an important role in the morphogenetic events of embryonic development by conferring selective adhesive properties to cells.

Despite the many functions attributed to the β1 integrin family, little is known about the details of β1 integrin function as played out by the individual αβ heterodimeric receptors. The localization, expression pattern and in vivo role of each individual receptor is only beginning to be addressed. Recently, progress in this area has accelerated as antibodies for many of the human α subunits have been described (Hemler, 1990).

However, a comparable library of antibodies directed against αsubunits in systems appropriate for embryologic studies has not yet been developed. Consequently, our understanding of β1 integrin function during embryonic development is still relatively primitive.

We have initiated a study to characterize the diversity of the chicken β1 integrin family and study the regulation and functions of the individual α β heterodimers during development. In this paper, we use affinity purification of the β1 integrins to reveal the diversity of α subunits expressed in several embryonic and adult tissues. We then use monoclonal antibodies generated against the chick embryo ‘band 1’ (Knudsen et al. 1985; Chen et al. 1985) to identify a chicken α subunit homologous to the human α5 subunit (fibronectin receptor) (Pytela, 1985) and to study the expression of this receptor in embryonic and adult tissues. The relative expression of the different α subunits seen in the β1 integrin purifications reveals the prominence of this chicken α5 subunit in tissues of the embryo, and its general down-regulation in the adult. Immunohistologi-cal studies reveal the widespread cell type distribution of the α5β1 integrin in the embryo, a distribution comparable to its ligand, fibronectin. These studies also reveal the restricted expression of the α5β1 integrin in adult tissues, and its down-regulation on many cell types during development.

Purification of integrins

Chicken integrins were purified by the method of Knudsen et al. (1985) with minor modifications. One volume of tissue was added to four volumes of extraction buffer (10 HIM Tris – HOAc pH8.0, 0.15M NaCl, 0.5mM CaCl2, 0.5mM MgCl2, 0.5% NP-40, 0.1 mM PMSF, 1.0 mM o-phenanthroline, 15 nM pepstatin, 2.1 μM leupeptin) and homogenized twice for 30s in a Waring blender. The homogenate was centifuged at 27 000g for 20 min at 4 °C. The supernatant was brought to 20 mM acetic acid, stirred on ice for 40 min and centrifuged again at 27 000g for 20 min at 4°C. The final supernatant was neutralized to pH8.0 with 1.0M Tris base and, if necessary, filtered through Whatman No. 1 filter paper (Whatman Inc., Cliβ1on, NJ). This extract was then passed over a 3 ml CSAT mAb affinity column (Neff et al. 1982) at a flow rate of 30mlh−1 at 4°C. The column was washed with 20 column volumes of extraction buffer and eluted with 50 mM diethylamine in extraction buffer, pH 11.5, with the detergent decreased to 0.05 % NP-40. Fractions were neutralized to pH8.0 with 1.0M HC1, pooled, and dialyzed against extraction buffer. Samples were oβ1en concentrated before dialysis using Centriprep 10 concentrators (Amicon, Danvers, MA).

The 150×l03A7r chicken α subunit was purified by first denaturing the purified chick embryo integrin heterodimers and passing these proteins over an affinity column constructed from the anti-chick αsubunit monoclonal antibody (α5A2, see below). The embryonic integrin heterodimers, stored in extraction buffer, were denatured by adding guanidine isothiocyanate (Sigma, St Louis, MO) to a final concentration of 6.0M and heating the sample to 60°C for 15 min. The sample was dialyzed once against 200 volumes of extraction buffer before passing over the affinity column. The column was then washed and eluted by the method described above for the CSAT column.

CSAT and α5A2 affinity columns were constructed by coupling 5 mg ml−1 pure monoclonal antibody to CNBr-activated Sepharose 4B (Pharmacia, Piscataway, NJ). The day 11 chicken embryos were prepared from decapitated and eviscerated embryos rinsed in cold PBS and stored at –70°C. The individual embryonic tissues were dissected, rinsed and stored similarly. Most adult tissues were obtained from Pel-Freez Biologicals (Rogers, Ark) and stored at –20°C. The only exception was the adductor femoris skeletal muscle and small intestine which were dissected in our laboratory from adult K-strain Leghorn chickens.

Protein electrophoresis

SDS–PAGE was performed by the method of Laemmli (1970). Separating gels were 7% acrylamide and 0.12% bisacrylamide. Samples were prepared in sample buffer (62.5 mM Tris–HCl pH6.8°C, 2.0% SDS, 10% glycerol) and heated at 60°C for 15 min. All gels were run at 4°C and were visualized by silver staining using Bio-Rad Silver Stain kit (Bio-Rad, Richmond, CA).

Electroelution

Electroelution was performed according to the method of Hunkapiller et al. (1983) using the procedure described for elution at 4 °C. The electroelution apparatus was obtained from Isco, Inc. (Lincoln, NE). The eluant removed from the sample collection chamber was lyophilized in a Speedvac evaporator (Savant Instruments, Inc., Farmingdale, NY), resuspended in 100 μm deionized water and precipitated with 4 volumes methanol/acetone (1:1) overnight at 4 °C to remove salts. Protein yield was estimated by comparison on silver-stained protein gels with samples of known concentration.

Antibody production

Nine-week-old female Balb/c mice were immunized at 21 day intervals with 10–25 μg of protein electroeluted from the ‘Band 1’ region of the chick embryo integrins (Knudsen et al. 1985). For the first immunization the antigen was mixed (1:1) with Freunds complete adjuvant (Gibco Laboratories; Life Technologies Inc., Grand Island, NY) to a final volume of 0.5ml/mouse and injected into the interperitoneal cavity. Subsequent immunizations were identical except Freunds incomplete adjuvant was used. Two weeks after the second immunization the sera from both mice tested positive for ‘band 1’ on immunoblots. A fusion was performed using one of these mice three days after the third immunization. The fusion was performed using an SP2/0 myeloma cell line by the methods of Kennett et al. (1982). Six hybridomas were identified as specific for protein within the chicken embryo integrin ‘band 1’. Three were used in this study; they are designated α5A2, α5B2 and α5D7. Having been generated against denatured protein, these antibodies immunoblot well, immunostain fixed tissue sections best when the sections are first heat denatured, and immunoprecipitate the monomeric α subunit afterthe dimeric receptor has first been denatured, but, they do not appear to precipitate the native receptor.

Hybridoma screening

The hybridoma supernatants were screened using a dot blot apparatus (Bio-Dot Apparatus; Bio-Rad) and immunoblot detection. A nitrocellulose filter (Schleicher and Schuell, Inc., Keene, NH) was assembled into the apparatus and the purified day 11 embryo integrins were aliquotted into the wells at a concentration of 0.5, μg/well and drawn onto the filter by vacuum. The filter was then removed, blocked with wash buffer (10 HIM Tris–HOAc pH 8.0, 0.15 M NaCl) containing 3% gelatin, and reassembled into the apparatus. Individual supernatants to be screened were applied to the wells at a dilution of 1:1 with wash buffer containing 0.5% gelatin, and incubated with the filters for 30 min. Afterrinsing the supernatants from the wells the filter was removed and developed according to the remaining immunoblot procedure described below.

Western immunoblots

SDS–PAGE was performed as described above. The proteins were electrophoretically transferred from the gels to nitrocellulose membranes (Schleicher and Schuell, Inc.) at 4°C overnight at 40 volts using a Bio-Rad Trans-Blot Cell (BioRad). The transfer buffer consisted of 25 HIM Tris, 19 mM glycine, 20% methanol. Filters were blocked in wash buffer containing 3% gelatin for Ih. All antibodies were diluted in wash buffer containing 0.5 % gelatin and incubated with the filters for 30min. Primary antibody concentrations ranged from 10 to 25μg ml−1. The filters were washed for 30 min at 37°C with three changes of wash buffer. 0.05% Tween-20 (Bio-Rad) was included in the buffer of the second wash. The biotinylated secondary antibody, biotinylated alkaline phosphatase and avidin were obtained from the Vectastain ABCAP Kit (Vector Laboratories, Burlingame, CA) and used at half the concentration recommended in the kit instructions. The color developing reagent was the Alkaline Phosphatase Substrate KIT II (Vector).

Immunofluorescence staining

Sections of the day 4 chick embryo were provided by M. Bronner-Fraser (UC, Davis, CA). For other sections the tissues were frozen in liquid nitrogen, embedded in O.C.T Compound (Tissue-Tek; Miles Scientific, Naperville, IL) and sectioned on a Tissue-Tek cryostat at a thickness of 6–10 microns. Sections were placed on gelatin-coated slides and blocked with 10 % goat serum (Sigma) in PBS for 1 h. Primary antibodies were diluted with 5 % goat serum in PBS. For the primary antibody incubations, the slides were placed on a heating plate set at 60°C. The antibodies were added and incubated for 30 min on the heating slides which were covered to prevent excessive evaporation. This procedure was required for good staining with the anti-chicken α5 antibodies and did not appear to inhibit the staining with other antibodies. Afterincubation with the primary antibodies, the slides were removed from the heat and washed several times with PBS at room temperature. The FITC-conjugated antimouse secondary antibody (Cappel; Oreganon Teknika, West Chester, PA) was diluted with 5 % goat serum in PBS to 40μg ml−1. The slides were incubated-with the secondary antibody for 30 min and washed extensively with PBS before mounting. The primary antibodies were generally used at a concentration of 25μg ml−1. The best staining for chicken α5 subunit was obtained when three monoclonal antibodies were pooled (‱5A2, α5D7 and α5B2) each at a concentration of 25μg ml−1 and incubated on the heated sections. The negative control for all section staining was the P3 mAb used at 75μg ml−1 to match the total IgG concentration in the α5 staining. The negative controls were photographed and printed with exposures identical to those used for the α5 staining.

Antibodies

W1B10 mAb was prepared as described by Hayashi et al. (1990). CSAT mAb was prepared as described by Neff et al. (1982). Rabbit anti-human α5 cytoplasmic domain serum was a gift of L. Reichardt and K. Tomaselli (UC, San Francisco, CA) (Tomaselli et al. 1988). Rabbit anti-human α3 cytoplasmic domain serum was a gift of C. Buck (Wistar Institute, Philadelphia, PA). Monoclonal anti-FN (B3/D6) was purchased from the Developmental Studies Hybridoma Bank (Johns Hopkins Univ. School of Medicine, Baltimore, MD) and prepared as described by Gardner and Fambrough (1983).

Adult tissues show diversity and differential expression of the chicken β1 integrins

The diversity of the chicken β1 integrin family was investigated by purifying the β1 integrins from embryonic and adult chicken tissues. Extracts of tissues were passed over a CSAT (anti-β1) mAb affinity column which purifies the β1 subunit and associated α subunits. These purifications provide a profile of the β1 integrins expressed in each tissue. Previous purifications of the chicken β1 integrins from day 11 chick embryos and from chick embryo fibroblasts (CEFs) showed only three major protein bands: two putative α subunit bands at 150 and 130×l03Mr (bands 1 and 2, respectively) and the common β1 subunit at 95–110×l03Mr (band 3) (see lanes 1 and 2, Fig. 1A) (Knudsen et al. 1985; Chen et al. 1985).

Fig. 1.

β1 integrin purifications and immunoblot detection of the ‘band 1’ 150×l03Mrα subunit. (A) β1 integrin purifications from adult chicken tissues. Extracts from various tissues were applied to a CSAT (anti-β1) mAb affinity column. The purified proteins were run on 7% SDS–PAGE gels (non-reduced) and visualized by silver staining. Lane 1, whole day 4 embryo; Lane 2, eviscerated day 11 embryo; lane 3, adult adductor femoris (fast twitch) skeletal muscle; lane 4, adult breast (slow twitch) skeletal muscle; lane 5, adult heart; lane 6, adult gizzard; lane 7, adult sciatic nerve; lane 8, adult brain. The β1 subunit migrates as a broad band between 95 and 110×l03Mr. The putative α subunit are bracketed between 120 and 180×l03Mr. The 200×l03Mr protein purifies on control columns and is not an integrin-associated protein. The 70×l03Mr band appears to be an integrin-associated protein. (B) Immunoblot detection of the ‘band 1’ 150×l03Mr αsubunit from the β1 integrin purifications. Identical samples to those shown in Fig. 1A were run on 7% SDS–PAGE gels (non-reduced), transferred to nitrocellulose and probed with an anti-band 1 mAb (α5A2). This antibody detects a single band out of the many α subunit bands present in the purifications. This α subunit is virtually absent from the α1 integrins purified from the gizzard.

Fig. 1.

β1 integrin purifications and immunoblot detection of the ‘band 1’ 150×l03Mrα subunit. (A) β1 integrin purifications from adult chicken tissues. Extracts from various tissues were applied to a CSAT (anti-β1) mAb affinity column. The purified proteins were run on 7% SDS–PAGE gels (non-reduced) and visualized by silver staining. Lane 1, whole day 4 embryo; Lane 2, eviscerated day 11 embryo; lane 3, adult adductor femoris (fast twitch) skeletal muscle; lane 4, adult breast (slow twitch) skeletal muscle; lane 5, adult heart; lane 6, adult gizzard; lane 7, adult sciatic nerve; lane 8, adult brain. The β1 subunit migrates as a broad band between 95 and 110×l03Mr. The putative α subunit are bracketed between 120 and 180×l03Mr. The 200×l03Mr protein purifies on control columns and is not an integrin-associated protein. The 70×l03Mr band appears to be an integrin-associated protein. (B) Immunoblot detection of the ‘band 1’ 150×l03Mr αsubunit from the β1 integrin purifications. Identical samples to those shown in Fig. 1A were run on 7% SDS–PAGE gels (non-reduced), transferred to nitrocellulose and probed with an anti-band 1 mAb (α5A2). This antibody detects a single band out of the many α subunit bands present in the purifications. This α subunit is virtually absent from the α1 integrins purified from the gizzard.

The purifications of the β1 integrins from chicken tissues are shown in Fig. 1A. The common β subunit migrates as a broad band between 95 and 110×l03Mr. Differences are evident in the apparent molecular weights of the β1 subunits from different tissues: the β1 subunits of the gizzard and brain migrate faster than those of the other tissues, and some of the β1 subunit bands are broader than others. These differences among the β1 subunits appear to arise primarily by variable glycosylation as only one message is seen on northern blots and glycosidase-treated integrins migrated similarly on protein gels (data not shown).

All other bands of the β1 integrin purifications in Figure 1A are putative integrin α subunits or integrin-associated proteins by virtue of their association with the β1 subunit. The only major exception is the 200×l03Mr protein which purifies on control columns. An estimated 1 to 5 different α subunits, in the molecular weight range of 120 to 180×l03Mr, are seen in the purifications from each tissue. It is possible that some of these bands represent different glycosylated forms of a common α subunit; however, antibodies that react with chicken α subunit on immunoblots detect only single bands in each purification (see below). A greater number of α subunit may be present in these purifications than visualized in Fig. 1A. Some α subunit may co-migrate with other bands, and others may be present at levels too low to be detected. The number of distinct α subunit present in these purifications appears to exceed the seven identified in the human system thus far.

Unique sets of α subunit are expressed in each tissue in Fig. 1A. The day 4 and day 11 embryos each has the prominent α subunit at 150 and 130×l03Mr (bands 1 and 2 respectively). More bands are resolved in the day 11 embryo than in the day 4 embryo, but band 1 is the predominant α subunit band in both. Most of the α subunit from the adult tissues appear to represent receptors not detected in the purifications from the embryo. All of the adult tissues have prominent bands migrating at molecular weights higher than that of the embryo band 1. The brain and sciatic nerve also have minor bands migrating lower than the embryo band 1. The integrin profile from the gizzard contrasts all other tissues in that it has only a single major α subunit (at 155×l03Mr), and band 1 is essentially absent. In addition, skeletal and cardiac muscle both express a 70×l03Mr protein that co-purifies with the β1 integrins but whose molecular weight is not characteristic of integrin α subunit. Some of this 70×l03Aβ1 protein migrates with the integrin heterodimer when examined by gel filtration, suggesting that it is an integrin-associated protein (data not shown).

The most prominent α subunit in the chick embryo is homologous to the human α5 subunit

As band 1 represents the major β1 integrin a′subunit(s) in the embryo and is less prominent among the β1 integrins of the adult, we initiated a study of its expression and regulation during development. We first generated several (6) mAbs specific for band 1. All of these mAbs were subsequently shown to recognize a single protein. Immunoprecipitation experiments show this one protein to account for the majority (>75 %) of the protein in the embryonic band 1 (not shown). Fig. IB shows an immunoblot of the purified β1 integrins using one of these mAbs (α5A2). These antibodies detect a single band out of the many putative α subunit present in each tissue. The immunoblot signal corresponds closely in position and in relative intensity to the silver stained band seen at 150×l03Mr in all of the integrin purifications of Fig. 1A.

Immunofluorescent staining of fibroblasts using the anti-150×l03Mrα subunit mAbs shows that this receptor concentrates in the adhesion plaques of fibroblasts adhering to a fibronectin substrate (not shown). Several integrins, including the human fibronectin receptor 5β1), localize in the adhesion plaques of fibroblasts which are adhering to the receptors’ respective ligands (Burridge, 1986; Roman et al. 1989; Singer et al. 1988). Therefore, the observed localization of the chicken α subunit suggested that it functions as the α subunit of a fibronectin receptor. To test if the chicken protein was homologous to the human fibronectin receptor (α5β1), we obtained an antiserum against an 18 amino acid polypeptide corresponding to the cytoplasmic domain of the human <5 subunit (Tomaselli et al. 1988). The anti-human α5 antiserum recognizes specifically the 150× l03 Mrα subunit purified using our antibodies (Fig. 2). Thus, by immunologic relatedness and by localization in focal contacts of cells adhering to fibronectin, we conclude that the 150×l03Mr αsubunit purified by our antibodies is the chicken homolog of the human ag and will be referred to as the chicken α5 subunit.

Fig. 2.

Purification of the 150×l03Mr chicken αsubunit and identification as the homolog of the human α5 subunit. (A) Immunoaffinity purification of the 150×l03Mr chicken α subunit. The β1 integrin heterodimers purified from day 11 chick embryos were denatured and applied to an affinity column constructed from the anti-150 ×l03Mr chicken α subunit mAb (α5A2). The proteins were run on 7% SDS–PAGE gels (non-reduced) and visualized by silver staining. The purified β1 integrins from the day 11 embryo are shown in lane 1; the purified α subunit is shown in lane 2. The 160×l03Mr band in lane 2 is monoclonal antibody released from the affinity column support matrix. The faint 90×l03 M r band in lane 2 represents a breakdown product of the α subunit. (B) Immunoblots of the purified 150×l03Aβ1 chicken α subunit with α subunit specific antibodies. The integrin proteins used for Fig. 2A were run on 7% SDS-PAGE gels, transferred to nitrocellulose and probed with several α subunit-specific antibodies. The polyclonal anti-human α5 serum and anti-chick α subunit mAb (α5A2) show identical immunoreactivity. The control panel was probed with anti-mouse secondary antibody in the absence of primary antibody. The anti-human α3 serum was used as an additional control to show the selective purification of a single α subunit in lane 2.

Fig. 2.

Purification of the 150×l03Mr chicken αsubunit and identification as the homolog of the human α5 subunit. (A) Immunoaffinity purification of the 150×l03Mr chicken α subunit. The β1 integrin heterodimers purified from day 11 chick embryos were denatured and applied to an affinity column constructed from the anti-150 ×l03Mr chicken α subunit mAb (α5A2). The proteins were run on 7% SDS–PAGE gels (non-reduced) and visualized by silver staining. The purified β1 integrins from the day 11 embryo are shown in lane 1; the purified α subunit is shown in lane 2. The 160×l03Mr band in lane 2 is monoclonal antibody released from the affinity column support matrix. The faint 90×l03 M r band in lane 2 represents a breakdown product of the α subunit. (B) Immunoblots of the purified 150×l03Aβ1 chicken α subunit with α subunit specific antibodies. The integrin proteins used for Fig. 2A were run on 7% SDS-PAGE gels, transferred to nitrocellulose and probed with several α subunit-specific antibodies. The polyclonal anti-human α5 serum and anti-chick α subunit mAb (α5A2) show identical immunoreactivity. The control panel was probed with anti-mouse secondary antibody in the absence of primary antibody. The anti-human α3 serum was used as an additional control to show the selective purification of a single α subunit in lane 2.

The α5β1 integrin is widely distributed in embryonic tissues

The integrin purification results described above suggest that the chicken subunit is the major α subunit expressed in the embryo. It follows, therefore, that the α5 subunit is expressed either widely on most cell types or strongly on a few cell types that are prominent in their respective tissues. We have investigated the cell type distribution of the α5β1 integrin in the chick embryo by immunofluorescent staining of tissue cryosections with antibodies specific for the α5 subunit.

Fig. 3 shows a transverse section of a stage 24 (day 4) chick embryo stained with mAbs specific for the α5 subunit. Weak to moderate α5 staining can be seen throughout most of the embryo with the exception of the neural tissues (neural tube, dorsal root ganglia (DRG) and ventral horn) which show only background levels of fluorescence. Bright α5 staining can be seen on cells that outline the neural tube and DRG, and on the walls of the aorta. Moderate staining is seen along the lateral and dorsal side of the embryo, and extremely weak but detectable staining is seen in central portion of the embryo. Fig. 4 shows α5 staining in the limb bud from the same embryo section. Here too αs staining is seen throughout the mesenchyme. The α5 subunit is absent only from the ectodermal cell layer of the limb bud. This distribution of the integrin seen in the day 4 embryo is very similar to the reported distribution of its ligand, fibronectin (Kosher et al. 1982; Rogers et al. 1986).

Fig. 3.

Fluorescent photomicrograph of a cross-section through a stage 24 (day 4) chick embryo showing the localization the α5 subunit. The neural tube (NT), dorsal root ganglia (D) and ventral roots (arrows) do not stain for the α5 subunit. Bright staining is seen on the cells that outline the neural tube and dorsal root ganglia, and on the walls of the aorta (A). In general, α5 staining can be detected throughout the entire mesenchyme.

Fig. 3.

Fluorescent photomicrograph of a cross-section through a stage 24 (day 4) chick embryo showing the localization the α5 subunit. The neural tube (NT), dorsal root ganglia (D) and ventral roots (arrows) do not stain for the α5 subunit. Bright staining is seen on the cells that outline the neural tube and dorsal root ganglia, and on the walls of the aorta (A). In general, α5 staining can be detected throughout the entire mesenchyme.

Fig. 4.

Fluorescent photomicrograph of a cross-section through the limb of a stage 24 (day 4) chick embryo showing the localization of the α5 subunit. The α5 subunit is found throughout the mesenchyme but absent from the ectoderm layer (arrows). The patches of brighter staining consist of endothelial cells.

Fig. 4.

Fluorescent photomicrograph of a cross-section through the limb of a stage 24 (day 4) chick embryo showing the localization of the α5 subunit. The α5 subunit is found throughout the mesenchyme but absent from the ectoderm layer (arrows). The patches of brighter staining consist of endothelial cells.

Later stages of development (days 4-17) show a similar distribution of the α5β1 integrin. Despite extensive cell differentiation and tissue formation, the α5 subunit continues to be expressed on the majority of cell types in the embryo with the exception of neuronal and epithelial cells. Fig. 5 illustrates this widespread α5 staining pattern on a section of the small intestine from a day 16 chick embryo. This section shows α5 staining over the majority of cell types including the smooth muscle layer, but only background staining is seen on the intestinal epithelium. In sections from all muscle types (smooth, cardiac and skeletal muscle) in the day 11 embryo, the α5 subunit is detected on essentially all cells including desmin-positive muscle cells, desminnegative muscle precursors, fibroblasts and endothelial cells (not shown). In the day 14 kidney, α5 staining is observed on all cell types with the exception of the tubules (i.e. epithelial cells) (not shown). The continued absence of the α5 subunit from neurons is illustrated in a section of the day 11 sciatic nerve (Fig. 6). In this section, the α5 staining is limited primarily to the nascent perineurium surrounding the neurons. The neurons themselves do not stain detect-ably for the α5 subunit, although some punctate a5 staining is seen among them. The co-presence of fibronectin in these areas (Fig. 6B) suggests that this staining may correspond to regions of emerging connective tissue and capillaries. Sections of the embryonic retina (days 7–17) and ciliary ganglion (days 8 and 13) show α5 staining on capillaries and connective tissues but do not show detectable staining on the neurons (not shown).

Fig. 5.

Fluorescent photomicrographs of cross-sections through the small intestine of a day 16 chick embryo showing the localization of the α5 subunit and fibronectin. Tissue sections stained with anti-α5 mAbs (A), an anti-fibronectin mAb (B) and a negative control mAb (C). The smooth muscle layer (SM) and epithelial layer (E) are labelled in frame (A). (bar=100 μm).

Fig. 5.

Fluorescent photomicrographs of cross-sections through the small intestine of a day 16 chick embryo showing the localization of the α5 subunit and fibronectin. Tissue sections stained with anti-α5 mAbs (A), an anti-fibronectin mAb (B) and a negative control mAb (C). The smooth muscle layer (SM) and epithelial layer (E) are labelled in frame (A). (bar=100 μm).

Fig. 6.

Fluorescent photomicrographs of cross-sections through the sciatic nerve of an day 11 chick embryo showing the localization of the ‱5 subunit and fibronectin. Tissue sections stained with anti-‱5 mAbs (A), an anti-fibronectin mAb (B), and a negative, control mAb (C). The neurons (N) and perineurium (P) are labelled in frame A. (bar= 100 μm).

Fig. 6.

Fluorescent photomicrographs of cross-sections through the sciatic nerve of an day 11 chick embryo showing the localization of the ‱5 subunit and fibronectin. Tissue sections stained with anti-‱5 mAbs (A), an anti-fibronectin mAb (B), and a negative, control mAb (C). The neurons (N) and perineurium (P) are labelled in frame A. (bar= 100 μm).

As in the day 4 embryo, the distribution of the α5β1 integrin in the later embryo generally parallels that of its ligand, fibronectin. This is illustrated in Figs 5B and 6B where the staining for fibronectin mirrors the staining for α5. This co-distribution of the integrin with fibronectin was observed in all embryonic tissues surveyed.

The α5β1 integrin expression is restricted to a few cell types in the adult

In contrast to the widespread distribution of the α5 subunit in tissues from the embryo, its distribution in the adult is limited to a few cell types. The α5 subunit is found mainly in regions of connective tissue and vasculature. This restricted expression is illustrated in a section of smooth muscle from the adult gizzard (Fig. 7A). In this section, the α5 subunit is detected only on capillaries and in some fibronectin-rich connective tissue regions. The smooth muscle cells, which comprise the major cell type in this tissue, do not stain detectably for the α5 subunit. This restricted expression was seen in all tissues surveyed including all muscle types, the small intestine and the kidney (not shown). The α5 subunit remains absent from the neurons in adult tissues as observed on sections of the adult sciatic nerve and retina (not shown). This restricted expression of the The α5β1 integrin in adult tissues is observed in the same tissues which, in the embryo, expressed this receptor on the majority of cells. This points to a dramatic down-regulation of the The α5β1 integrin on many cell types during the later stages of cell differentiation.

Fig. 7.

Fluorescent photomicrographs of sections through the gizzard of an adult chicken showing localization of the ‱5 subunit and fibronectin. Tissue sections stained with anti-‱5 mAbs (A), an anti-fibronectin mAb (B), and a negative control mAb (C). Some capillaries are designated with arrows in frame A. (bar=50μm).

Fig. 7.

Fluorescent photomicrographs of sections through the gizzard of an adult chicken showing localization of the ‱5 subunit and fibronectin. Tissue sections stained with anti-‱5 mAbs (A), an anti-fibronectin mAb (B), and a negative control mAb (C). Some capillaries are designated with arrows in frame A. (bar=50μm).

The fibronectin distribution in adult tissues differs significantly from that of the The α5β1 integrin, in contrast to the strong co-localization seen in the embryo. Although the The α5β1 integrin always co-localizes with fibronectin in the adult, fibronectin is also found in many areas where the The α5β1 integrin has disappeared. Such differences are apparent in the section of the adult gizzard (Fig. 7) where the α5 subunit is not detectable on the smooth muscle cells, but, bright fibronectin staining surrounds these cells.

The β1 integrins mediate the interactions between cells and molecules in the extracellular matrix (ECM) (Hynes, 1987; Ruoslahti and Pierschbacher, 1987; Buck and Horwitz, 1987). They also mediate linkages between the cell surface and the cytoskeleton (Horwitz et al. 1986, Otey et al. 1990; Argraves et al. 1989). The specificity of the different integrins for ECM ligands is determined by the variable α subunit that associate with the common β1 subunit. Although the ligand specificities for the different αβ heterodimers have not been characterized completely, the following specificities have been assigned: α1β1, Col, LM (Kramer and Marks, 1989; Ignatius and Reichardt, 1988); α2β1, Col, LM (Wayner and Carter, 1987; Staatz et al. 1989; Elices and Hemler, 1989); α3β1, FN, LM, (Col) (Wayner and Carter, 1987; Gehlsen et al. 1988); α4β1, FN (CS-1 region) (Wayner et al. 1989); α5β1, FN (Pytela et al. 1985); α6β1, LN (Sonnenberg et al. 1988); αvβ1, FN (Vogel et al. 1990).

The expression of different α1 integrins confers selective adhesive properties to a cell. Therefore, the differential expression of integrin αβ heterodimers provides a potential mechanism for morphogenetic events during development. Until recently, it has been difficult to address the issue of differential integrin expression during development due to a lack of probes for integrin α subunit in systems amenable to embryologic studies. We have approached the problem of α1 integrin expression during development in two ways. First, we obtained profiles of α subunit expression by affinity purifying the sets of α1 integrins from various embryonic and adult chicken tissues. These profiles offer a glimpse at the diversity and differential expression of the α1 integrins. Second, we generated monoclonal antibodies against chicken α subunit. These antibodies can be used as probes for the expression of the individual α1 integrins during chick development.

In the present work, we describe mAbs raised against a 150×l03Mrα subunit in the prominent ‘band 1’ of the chicken embryo integrins and characterize its expression in embryonic and adult tissues. We find that this 150×l03Mrα subunit is homologous to the human integrin α5 subunit. It is the most prominent α1 integrin αsubunit in the embryo and is found on most cell types in the day 4-17 chick embryo. In the adult, however, the level of α5 expression is dramatically reduced relative to other α subunit. Its distribution in the adult is restricted primarily to connective tissue and vasculature. These observations reveal a down-regulation of α5 on many cell types and also imply an up-regulation of other α subunit during development.

The 150×l03Mrα subunit purified using our antibodies was identified as the avian homologue of the human α5β1 fibronectin receptor by two criteria: (1) its concentration in adhesion plaques of cells adhering to fibronectin, and (2) its cross-reaction with antisera specific for the cytoplasmic domain of the human α5 subunit. During the course of this work, Hofer et al. (1990) described a mAb directed against a 150×l03Mr avian integrin α subunit. They identified this α subunit as a homolog of the human α5 by its localization in adhesion plaques of fibroblasts adhering to fibronectin, and by comparing its amino terminal protein sequence with that of the human o′subunits. Hynes et al. (1989) have also identified a 150 ×l03MT protein in the chicken integrin band 1 as the α subunit of a fibronectin receptor related to the human α5 subunit. This was demonstrated by the cross-reaction of an antiserum raised against the human α5 cytoplasmic domain and by the binding of this receptor to a fibronectin affinity column. The cumulative evidence suggests that we have all identified the same α subunit and that this α subunit is the chicken homolog of the human α5 integrin.

The dominant and widespread expression of the α5subunit in the chick embryo was a striking, unexpected observation. Restricted α5 expression in adult tissues has been reported in human kidney (Korhonen et al. 1990) but widespread expression in embryonic tissues has not been observed. The presence of the α5 subunit on most cells in diverse tissues in day 4-17 chick embryos points to a major role for the α5 subunit during development. The loss of this receptor in adult cell types may be of equal significance. The observed extinction of the α5 subunit seems to accompany the terminal differentiation of many cell types. Loss of the α5β1 integrin upon terminal cell differentiation has been described previously during the terminal differentiation of kératinocytes (Adams and Watt, 1990). In addition, Vuillet-Gaugler et al. (1990) and Cardarelli et al. (1988) report the loss of integrin-dependent binding to fibronectin upon terminal differentiation of erythrocytes and thymocytes, respectively.

The observed changes in the relative distribution of the α5β1 integrin and fibronectin during development may also be significant. Fibronectin and α5β1 integrin distribution are both widespread and nearly identical in the embryo, but in the adult their distributions differ significantly. Several other β1 integrins are reported to serve as receptors for fibronectin in addition to the α5β1 integrin. These include the α3β1, α4β1 and αvβ1 integrins. As seen in the purification profiles, the α5β1 is clearly the predominant β1 integrin fibronectin receptor in the embryo; however, the relative expression of the other fibronectin receptors appears to increase at later stages of development. Intriguingly, it has become apparent that the alternately spiced fibronectin isoforms are more prevalent in adult tissues whereas a single fibronectin isoform is most prevalent in the embryo (ffrench-Constant and Hynes, 1988, 1989). Therefore, the observed pattern of fibronectin receptor expression may be related to the expression of alternate fibronectin isoforms. The α4β1 is one integrin known to serve as the receptor for an alternately spliced domain of fibronectin (Wayner et al. 1989). Other possible functions for the different fibronectin receptor α subunit may reside in their cytoplasmic domains, which appear to be unique (Hemler, 1990; Hynes et al. 1989) and may specify associations with different cytoskeletal components or transduce specific signals.

The question now arises as to the role(s) of the α5β1 integrin in development. A complete answer should address its nearly ubiquitous presence across diverse embryonic cell types, its presence in adult connective tissue, its possible up-regulation during wound healing and inflammation (Clark, 1990; Holers et al. 1989), and its apparent extinction on many terminally differentiated cell types. Possible functions of the α5β1 integrin include: matrix assembly, cytoskeletal organization, cell migration, and transduction of differentiative or proliferative signals. A role as an organizer of the fibronectin ECM (Giancotti and Ruoslahti, 1990; Akiyama et al. 1989; Roman et al. 1989) is consistent with the widespread distribution of this receptor, including its presence in the loose mesenchyme of the embryo and its persistence in connective tissue regions of the adult. The role of α5β1 integrin in cell migration is not clear. Some migratory cells, such as fibroblasts and macrophages, do express this receptor (Holers et al. 1989); however, in some studies, the presence of the α5β1 integrin is reported to inhibit cell migration on fibronectin (Akiyama et al. 1989; Giancotti and Ruoslahti, 1990). The α5β1 integrin might function to mediate signals pertaining to differentiation. The differentiation of both muscle and kératinocytes are inhibited by fibronectin binding (von der Mark and Ocalan, 1989; Adams and Watt, 1989). Erythrocyte differentiation, on the other hand, requires fibronectin binding (Patel and Lodish, 1987). Werb et al. (1989) have shown induction of specific genes as the result of signal transduction through the human α5β1 integrin. Finally, there is a possible relationship between integrin expression and cell proliferation. This receptor appears to be preferentially expressed on proliferating cell types (Cardarelli et al. 1988). Most recently, Giancotti and Ruoslahti (1990) have implicated the α5β1 integrin in the control of cell proliferation and suggest that loss of this receptor contributes to anchorage-independent growth by transformed cells.

     
  • ECM

    extracellular matrix

  •  
  • FN

    fibronectin

  •  
  • LM

    laminin

  •  
  • Col

    collagen

  •  
  • mAb

    monoclonal antibody.

We would like to acknowledge the generous gifts of the anti-human α5 antiserum from Dr L. F. Reichardt, of the antihuman α3 antiserum from Dr C. Buck, and of the stage 24 embryo sections from Dr M. Bronner-Fraser. We would like to thank Steve Miklasz for his assistance in monoclonal antibody production, and also Rhett Miller for his technical assistance throughout the course of this work.

Adams
,
J. C.
and
Watt
,
F. M.
(
1990
).
Changes in kératinocyte adhesion during terminal differentiation: reduction in fibronectin binding precedes axff integrin loss from the cell surface
.
Cell
63
,
425
435
.
Adams
,
J. C.
and
Watt
,
F. M.
(
1989
).
Fibronectin inhibits the terminal differentiation of human kératinocytes
.
Nature
340
,
307
309
.
Akiyama
,
S. K.
,
Yamada
,
S. S.
,
Chen
,
W.-T.
and
Yamada
,
K. M.
(
1989
).
Analysis of fibronectin function with monoclonal antibodies: roles in cell adhesion, migration, matrix assembly, and cytoskeletal organization
.
J. Cell Biol
.
109
,
863
875
.
Albelda
,
S. M.
and
Buck
,
C. A.
(
1990
).
Integrins and other cell adhesion molecules
.
FASEB J
.
4
,
2868
2880
.
Argraves
,
W. S.
,
Dickerson
,
K.
,
Burgess
,
W. H.
and
Ruoslahti
,
E.
(
1989
).
Fibulin, a novel protein that interacts with the fibronectin receptor fl subunit cytoplasmic domain
.
Cell
58
,
623
629
.
Argraves
,
W. S.
,
Suzuki
,
S. H.
,
Arai
,
H.
,
Thompson
,
K.
,
Pierschbacher
,
M. D.
and
Ruoslahti
,
R.
(
1987
).
Amino acid sequence of the human fibronectin receptor
.
J. Cell Biol
.
105
,
1183
1190
.
Bozycko
,
D.
and
Horwitz
,
A. F.
(
1986
).
The participation of a putative cell surface receptor for laminin and fibronectin in peripheral neurite extension
.
J. Neurosci
.
6
,
1241
1251
.
Bronner-Fraser
,
M.
(
1986
).
An antibody to a receptor for fibronectin and laminin perturbs cranial neural crest development in vivo
.
Devi Biol
.
117
,
526
536
.
Buck
,
C. A.
and
Horwitz
,
A. F.
(
1987
).
Cell surface receptors for extracellular matrix molecules
.
Ann. Rev. Cell Biol
.
3
,
179
205
.
Burridge
,
K.
(
1986
).
Substrate adhesions in normal and transformed fibroblasts: organization and regulation of cytoskeletal, membrane and extracellular matrix components at focal contacts
.
Cancer Rev
.
4
,
18
78
.
Cardarelli
,
P. M.
,
Crispe
,
I. N.
and
Pierschbacher
,
M. D.
(
1988
).
Preferential expression of fibronectin receptors on immature thymocytes
.
J. Cell Biol
.
106
,
2183
2190
.
Chen
,
W.-T.
,
Hasegawa
,
E.
,
Hasegawa
,
T.
,
Weinstock
,
C.
and
Yamada
,
K. M.
(
1985
).
Development of surface linkage complexes in cultured fibroblasts
.
J. Cell Biol
.
100
,
1103
1114
.
Clark
,
R. A.
(
1990
).
Fibronectin matrix deposition and fibronectin receptor expression in healing and normal skin
.
J. invest. Derm
.
94
,
128s
134s
.
Elices
,
M. J.
and
Hemler
,
M. E.
(
1989
).
The integrin VLA-2 is a collagen receptor on some cells and a collagen/laminin receptor on others
.
Pro. natn. Acad. Sci. U.S.A
.
86
,
9906
9910
Ffrench-Constant
,
C.
and
Hynes
,
R. O.
(
1988
).
Patterns of fibronectin gene expression and splicing during cell migration in chick embryos
.
Development
104
,
369
382
.
Ffrench-Constant
,
C.
and
Hynes
,
R. O.
(
1989
).
Alternative splicing of fibronectin is temporally and spatially regulated in the chick embryo
.
Development
106
,
375
388
.
Gardner
,
J. M.
and
Fambrough
,
D. M.
(
1983
).
Fibronectin expression during myogenesis
.
J. Cell Biol
.
96
,
474
485
.
Gehlsen
,
K. R.
,
Dillner
,
L.
,
Engwall
,
E.
and
Ruoslahti
,
E.
(
1988
).
The human laminin receptor is a member of the integrin family of cell adhesion receptors
.
Science
24
,
1228
1229
.
Giancotti
,
F. G.
and
Ruoslahti
,
E.
(
1990
).
Elevated levels of the alpha 5 beta 1 fibronectin receptor suppress the transformed phenotype of Chinese hamster ovary cells
.
Cell
60
,
849
859
.
Greve
,
J. M.
and
Gottlieb
,
D. I.
(
1982
).
Monoclonal antibodies which alter the morphology of cultured chick myogenic cells
.
J. cell. Biochem
.
18
,
221
230
.
Hayashi
,
Y.
,
Haimovich
,
B.
,
Reszka
,
A.
,
Boettiger
,
D.
and
Horwitz
,
A.
(
1990
).
Expression and function of chicken integrin β1subunit and its cytoplasmic domain mutants in mouse NIH 3T3 cells
.
J. Cell Biol
.
110
,
175
184
.
Hemler
,
M. E.
(
1990
).
VLA proteins in the integrin family: Structures, functions and their role on leukocytes
.
Ann. Rev. Immunol
.
8
,
365
400
.
Hofer
,
U.
,
Syfrig
,
J.
and
Chiquet-Ehrismann
,
R.
(
1990
).
Identification and characterization of a dimeric chicken fibronectin
.
J. biol. Chem.
265
,
14561
14565
.
Holers
,
V. M.
,
Ruff
,
T. G.
,
Parks
,
D. L.
,
Mcdonald
,
J. A.
,
Ballard
,
L. L.
and
Brown
,
E. J.
(
1989
).
Molecular cloning of the murine fibronectin receptor and its expression during inflammation
.
J. exp. Med
.
169
,
1589
1605
.
Horwitz
,
A.
,
Duggan
,
K.
,
Buck
,
C.
,
Beckerle
,
M. C.
and
Burridge
,
K.
(
1986
).
Interactions of plasma membrane fibronectin receptor with talin: a transmembrane linkage
.
Nature
320
,
531
533
.
Hunkapiller
,
M. W.
,
Lujan
,
E.
,
Ostrander
,
F.
and
Hood
,
L. E.
(
1983
).
Isolation of microgram quantities of proteins from polyacrylamide gels for amino acid analysis
.
Meth. Enzymol
.
91
,
227
236
.
Hynes
.
R. O.
(
1987
).
Integrins, a family of cell surface receptors
.
Cell
48
,
549
555
.
Hynes
,
R. O.
,
Marcantonio
,
E. E.
,
Stepp
,
M. A.
,
Urry
,
L. A.
and
Yee
,
G. H.
(
1989
).
Integrin heterodimer receptor complexity in avian and mammalian cells
.
J. Cell Biol
.
109
,
409
420
.
Ignatius
,
M. J.
and
Reichardt
,
L. F.
(
1988
).
Identification of a neuronal laminin receptor: an Mr 200/120 kD integrin heterodimer that binds laminin in a divalent cation-dependant manner
.
Neuron
1
,
713
725
.
Kennett
,
R. H.
,
Mckearn
,
T. J.
and
Bechtol
,
K. B.
(
1982
).
Monoclonal Antibodies: a New Dimension in Biological Analysis
.
Plenum Press
,
New York
.
Knudsen
,
K. A.
,
Horwitz
,
A. F.
and
Buck
,
C. A.
(
1985
).
A monoclonal antibody identifies a glycoprotein complex involved in cell-substratum adhesion
.
Expl Cell Res
.
157
,
218
228
.
Korhonen
,
M.
,
Ylanne
,
J.
,
Laitinen
,
L.
and
Virtanen
,
I.
(
1990
).
The ol-a6 subunits of integrins are characteristically expressed in distinct segments of developing and adult human nephron
.
J. Cell Biol
.,
1245
1254
.
Kosher
,
R. A.
,
Walker
,
K. H.
and
Ledger
,
P. W.
(
1982
).
Temporal and spatial distribution of fibronectin during development of the chick embryo limb
.
Cell Differentiation
11
,
217
228
.
Kramer
,
R. H.
and
Marks
,
N.
(
1989
).
Identification of integrin collagen receptors on human melanoma cells
.
J. biol. Chem
.
264
,
4684
4688
.
Laemmli
,
U. K.
(
1970
).
Cleavage of structural proteins during the assembly of the head of the bacteriophage T4
.
Nature
227
,
680
685
.
Menko
,
A. S.
and
Boettiger
,
D.
(
1987
).
Occupation of the extracellular matrix receptor, integrin, is a control point for myogenic differentiation
.
Cell
51
,
51
57
.
Neff
,
N. T.
,
Lowrey
,
C.
,
Decker
,
C.
,
Tovar
,
A.
,
Damsky
,
C.
,
Buck
,
C.
and
Horwitz
,
A. F.
(
1982
).
A monoclonal antibody detaches embryonic skeletal muscle from extracellular matrices
.
J. Cell Biol
.
95
,
654
666
.
Otey
,
C. A.
,
Pavalko
,
F. M.
and
Burridge
,
K.
(
1990
).
An interaction between cv-actinin and the ft integrin subunit in vitro
.
J. Cell Biol
.
111
,
721
730
.
Patel
,
V. P.
and
Lodish
,
H. F.
(
1987
).
A fibronectin matrix is required for differentiation of murine erythroleukemia cells into reticulocytes
.
J. Cell Biol
.
105
,
3105
3118
.
Pytela
,
R.
,
Pierschbacher
,
M. D.
and
Ruoslahti
,
E.
(
1985
).
Identification of a 140 kD cell surface glycoprotein with properties of a fibronectin receptor
.
Cell
40
,
191
198
.
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
.
Devi Biol
.
113
,
429
435
.
Roman
,
J.
,
Lachance
,
R. M.
,
Broekelmann
,
T. J.
,
Kennedy
,
C. J.
,
Wayner
,
R.
,
Carter
,
E. A.
and
Mcdonald
,
J. A.
(
1989
).
The fibronectin receptor is organized by extracellular matrix fibronectin: implications for oncogenic transformation and for recognition of fibronectin matrices
.
J. Cell Biol
.
108
,
2529
2543
.
Ruoslahti
,
E.
and
Pierschbacher
,
M. D.
(
1987
).
New perspectives in cell adhesion: RGD and integrins
.
Science
238
.
491
497
.
Singer
,
I. L
,
Scott
,
S.
,
Kawka
,
D. W.
,
Kazazis
,
D. M.
,
Gailit
,
J.
and
Ruoslahti
,
E.
(
1988
).
Cell surface distribution of fibronectin and vitronectin receptors depends on substrate composition and extracellular matrix accumulation
.
J. Cell Biol
.
106
,
2171
2182
.
Sonnenberg
,
A.
,
Modderman
,
P. W.
and
Hogervorst
,
F.
(
1988
).
Laminin receptor on platelets is the VLA-6
.
Nature
360
,
487
489
.
Staatz
,
W.
,
Rajpara
,
S. M.
,
Wayner
,
E. A.
,
Carter
,
W. G.
and
Santoro
,
S. A.
(
1989
).
The membrane glycoprotein la-IIa (VLA-2) complex mediates the Mg++-dependant adhesion of platelets to collagen
.
J. Cell Biol
.
108
,
1917
1924
.
Tamkun
,
J. W.
,
Desimone
,
D. W.
,
Fonda
,
D.
,
Patel
,
R. S.
,
Buck
,
C. A.
,
Horwitz
,
A. F.
and
Hynes
,
R. O.
(
1986
).
Structure of the cysteine-rich subunit of integrin, the fibronectin receptor
.
Cell
46
,
271
282
.
Tomaselli
,
K. J.
,
Damsky
,
C. H.
and
Reichardt
,
L. F.
(
1988
).
Purification and characterization of mammalian integrins expressed by a rat neuronal cell line (PC12): Evidence that they function as alpha/beta heterodimeric receptors for laminin and type IV collagen
.
J. Cell Biol
.
107
,
1241
1252
.
Tomaselli
,
K. J.
,
Reichardt
,
L. F.
and
Bixby
,
J. L.
(
1986
).
Distinct molecular interactions mediate neuronal process outgrowth on extracellular matrices and non-neuronal cell surfaces
.
J. Cell Biol
.
103
,
2659
2672
.
Vogel
,
B.
,
Tarone
,
G.
,
Giancotti
,
F. G.
,
Gailit
,
J.
and
Ruoslahti
,
E.
(
1990
).
A novel fibronectin receptor with an unexpected subunit composition αvβ1
.
J. biol. Chem
.
265
,
5934
5937
.
Von Der Mark
,
K.
and
Ocalan
,
M.
(
1989
).
Antagonistic effects of laminin and fibronectin on the expression of the myogenic phenotype
.
Differentiation
40
,
150
157
.
Vuillet-Gaugler
,
M. H.
,
Breton-Gorius
,
J.
,
Vainchenker
,
W.
,
Guichard
,
J.
,
Leroy
,
C.
,
Tchernia
,
G.
and
Coulombel
,
L.
(
1990
).
Loss of attachment to fibronectin with terminal human erythroid differentiation
.
Blood
75
,
865
873
.
Wayner
,
E. A.
and
Carter
,
W. G.
(
1987
).
Identification of multiple cell adhesion receptors for collagen and fibronectin in human fibrosarcoma cells processing unique a and common fi subunits
.
J. Cell Biol
.
105
,
1873
1884
.
Wayner
,
E. A.
,
Garcia-Pardo
,
A.
,
Humphries
,
M. J.
,
Mcdonald
,
J. A.
and
Carter
,
W. G.
(
1989
).
Identification and characterization of the T lymphocyte adhesion receptor for an alternative cell attachment domain (CS-1) in plasma fibronectin
.
J. Cell Biol
.
109
,
1321
1330
.
Werb
,
Z.
,
Tremble
,
P. M.
,
Behrendtsen
,
O.
,
Crowley
,
E.
and
Damsky
,
C. H.
(
1989
).
Signal transduction through the fibronectin receptor induces collagenase and stromelysin expression
.
J. Cell Biol
.
109
,
877
889
.