Integrin-mediated Cell Adhesion to Type I Collagen Fibrils
2004; Elsevier BV; Volume: 279; Issue: 30 Linguagem: Inglês
10.1074/jbc.m401409200
ISSN1083-351X
AutoresJohanna Jokinen, Elina Dadu, Petri Nykvist, Jarmo Käpylä, Daniel White, Johanna Ivaska, Piia Vehviläinen, Hilkka Reunanen, Hannu Larjava, Lari Häkkinen, Jyrki Heino,
Tópico(s)Connective tissue disorders research
ResumoIn the integrin family, the collagen receptors form a structurally and functionally distinct subgroup. Two members of this subgroup, α1β1 and α2β1 integrins, are known to bind to monomeric form of type I collagen. However, in tissues type I collagen monomers are organized into large fibrils immediately after they are released from cells. Here, we studied collagen fibril recognition by integrins. By an immunoelectron microscopy method we showed that integrin α2I domain is able to bind to classical D-banded type I collagen fibrils. However, according to the solid phase binding assay, the collagen fibril formation appeared to reduce integrin α1I and α2I domain avidity to collagen and to lower the number of putative αI domain binding sites on it. Respectively, cellular α1β1 integrin was able to mediate cell spreading significantly better on monomeric than on fibrillar type I collagen matrix, whereas α2β1 integrin appeared still to facilitate both cell spreading on fibrillar type I collagen matrix and also the contraction of fibrillar type I collagen gel. Additionally, α2β1 integrin promoted the integrin-mediated formation of long cellular projections typically induced by fibrillar collagen. Thus, these findings suggest that α2β1 integrin is a functional cellular receptor for type I collagen fibrils, whereas α1β1 integrin may only effectively bind type I collagen monomers. Furthermore, when the effect of soluble αI domains on type I collagen fibril formation was tested in vitro, the observations suggest that integrin type collagen receptors might guide or even promote pericellular collagen fibrillogenesis. In the integrin family, the collagen receptors form a structurally and functionally distinct subgroup. Two members of this subgroup, α1β1 and α2β1 integrins, are known to bind to monomeric form of type I collagen. However, in tissues type I collagen monomers are organized into large fibrils immediately after they are released from cells. Here, we studied collagen fibril recognition by integrins. By an immunoelectron microscopy method we showed that integrin α2I domain is able to bind to classical D-banded type I collagen fibrils. However, according to the solid phase binding assay, the collagen fibril formation appeared to reduce integrin α1I and α2I domain avidity to collagen and to lower the number of putative αI domain binding sites on it. Respectively, cellular α1β1 integrin was able to mediate cell spreading significantly better on monomeric than on fibrillar type I collagen matrix, whereas α2β1 integrin appeared still to facilitate both cell spreading on fibrillar type I collagen matrix and also the contraction of fibrillar type I collagen gel. Additionally, α2β1 integrin promoted the integrin-mediated formation of long cellular projections typically induced by fibrillar collagen. Thus, these findings suggest that α2β1 integrin is a functional cellular receptor for type I collagen fibrils, whereas α1β1 integrin may only effectively bind type I collagen monomers. Furthermore, when the effect of soluble αI domains on type I collagen fibril formation was tested in vitro, the observations suggest that integrin type collagen receptors might guide or even promote pericellular collagen fibrillogenesis. A fibril-forming type I collagen, a ubiquitous protein in all vertebrates, is known to provide mechanical stability for tissues and serve as a functional environment for cells (1Prockop D.J. Kivirikko K.I. Annu. Rev. Biochem. 1995; 64: 403-434Google Scholar). Depending on the physical properties of the tissue, type I collagen fibrils are arranged with different suprafibrillar architectures and diameters. Thus, narrow fibrils (∼20 nm) in highly ordered arrangement occur in the cornea, where optical transparency is important, whereas large diameter fibrils (∼500 nm) provide high tensile strength in mature tendon (2Hulmes J. J. Struct. Biol. 2002; 137: 2-10Google Scholar). The mechanism of type I collagen fibril formation has been under extensive research for decades. In tissues, type I collagen is synthesized as a monomeric precursor, which is secreted by exocytosis into the extracellular space. In addition to the triple helical collagenous domain, the precursor contains noncollagenous C- and N-propeptides, which are linked to the triple helical domain by short sequences called telopeptides (3van der Rest M. Garrone R. FASEB J. 1991; 5: 2814-2823Google Scholar). After the enzymatic removal of propeptides, the solubility of collagen monomers decreases, and they spontaneously form fibrils, assisted by remaining nonhelical telopeptides (1Prockop D.J. Kivirikko K.I. Annu. Rev. Biochem. 1995; 64: 403-434Google Scholar, 4Kuznetsova N. Leikin S. J. Biol. Chem. 1999; 274: 36083-36088Google Scholar). Evidently, collagen molecules themselves contain all the information needed for fibril assembly. Therefore, in physiological conditions, acid-solubilized collagen monomers form tissue-type long fibrils with characteristic axial periodic structure also in vitro (5Wood G.C. Biochem. J. 1960; 75: 598-605Google Scholar, 6Wood G.C. Biochem. J. 1960; 75: 605-612Google Scholar, 7Kadler K. Holmes D. Trotter J. Chapman J. Biochem. J. 1996; 316: 1-8Google Scholar). However, self-assembly cannot by itself explain the diverse morphologies of collagen fibrils found in different tissues. In fact, several collagen binding extracellular matrix molecules, decorin, fibromodulin and lumican among others, have been found to influence the size of type I collagen fibrils (8Danielson K.G. Baribault H. Holmes D.F. Graham H. Kadler K.E. Iozzo R.V. J. Cell Biol. 1997; 136: 729-743Google Scholar, 9Chakravati S. Magnuson T. Lass J. Jepsen K. LaMantia C. Carrol H. J. Cell Biol. 1998; 141: 1277-1286Google Scholar, 10Svensson L. Aszodi A. Reinholt F. Fassler R. Heinegard D. Oldberg A. J. Biol. Chem. 1999; 274: 9636-9647Google Scholar). Recently, also cell surface receptors, including integrins, were recognized to contribute to collagen polymerization indicating that the collagen fibril formation is under close cellular control (11Velling T. Risteli J. Wennerberg K. Mosher D.F. Johansson S. J. Biol. Chem. 2002; 277: 37377-37381Google Scholar, 12Li S. Van Den Diepstraten C. D'Souza S.J. Chan B.M. Pickering J.G. Am. J. Pathol. 2003; 163: 1045-1056Google Scholar). The integrins are a large family of cell adhesion receptors involved in cell-cell and cell-matrix interactions. In addition to their importance as anchoring molecules, these heterodimeric receptors transmit bidirectional signals that regulate many important aspects of cell behavior including proliferation, differentiation, and survival. The integrin family contains four collagen receptors that are structurally closely related (13White D.J. Puranen S. Johnson M.S. Heino J. Int. J. Biochem. Cell Biol. 2004; 36: 1405-1410Google Scholar). Two well known collagen receptors are named α1β1 and α2β1 integrins, whereas the α10β1 and α11β1 integrins constitute the latest additions to the integrin family (14Takada Y. Hemler M.E. J. Cell Biol. 1989; 109: 397-407Google Scholar, 15Briesewitz R. Epstein M.R. Marcantonio E.E. J. Biol. Chem. 1993; 268: 2989-2996Google Scholar, 16Camper L. Hellman U. Lundgren-Akerlund E. J. Biol. Chem. 1998; 273: 20383-20389Google Scholar, 17Velling T. Kusche-Gullberg M. Sejersen T. Gullberg D. J. Biol. Chem. 1999; 274: 25735-25742Google Scholar). Whereas all collagen receptor integrins share the common β1 integrin subunit, there are four unique α subunits, resulting in the four heterodimers. Unlike the other matrix receptor integrins, the collagen receptors have an independently folding structural domain, inserted domain (αI domain), which executes ligand recognition (10Svensson L. Aszodi A. Reinholt F. Fassler R. Heinegard D. Oldberg A. J. Biol. Chem. 1999; 274: 9636-9647Google Scholar, 11Velling T. Risteli J. Wennerberg K. Mosher D.F. Johansson S. J. Biol. Chem. 2002; 277: 37377-37381Google Scholar, 12Li S. Van Den Diepstraten C. D'Souza S.J. Chan B.M. Pickering J.G. Am. J. Pathol. 2003; 163: 1045-1056Google Scholar, 13White D.J. Puranen S. Johnson M.S. Heino J. Int. J. Biochem. Cell Biol. 2004; 36: 1405-1410Google Scholar). This ∼200-amino acid domain is located in the integrin α subunit, between the second and third N-terminal blades of the β-propeller domain. The αI domain itself adopts a “Rossman fold” conformation, in which a central β-sheet is surrounded by α-helices forming a divalent metal binding site, referred to as MIDAS (18Lee J.O. Rieu P. Arnaout M.A. Liddington R. Cell. 1995; 80: 631-638Google Scholar). Despite the structural similarity of highly conserved αI domains, the small differences present may provide specific ligand binding characteristics for different integrins (19Kamata T. Takada Y. J. Biol. Chem. 1994; 269: 26006-26010Google Scholar, 20Kern A. Briesewitz R. Bank I. Marcantonio E.E. J. Biol. Chem. 1994; 269: 22811-22816Google Scholar, 21Tulla M. Pentikäinen O. Viitasalo T. Käpylä J. Impola U. Nykvist P. Nissinen L. Johnson M. Heino J. J. Biol. Chem. 2001; 276: 48206-48212Google Scholar, 22Zhang W.M. Käpylä J. Puranen J.S. Knight C.G. Tiger C.F. Pentikäinen O.T. Johnson M.S. Farndale R.W. Heino J. Gullberg D. J. Biol. Chem. 2003; 278: 7270-7277Google Scholar). Although α1β1 integrin is able bind to monomers of fibril-forming collagen types I, II, III, and V, it seems to favor other subtypes such as basement membrane type IV collagen (21Tulla M. Pentikäinen O. Viitasalo T. Käpylä J. Impola U. Nykvist P. Nissinen L. Johnson M. Heino J. J. Biol. Chem. 2001; 276: 48206-48212Google Scholar, 22Zhang W.M. Käpylä J. Puranen J.S. Knight C.G. Tiger C.F. Pentikäinen O.T. Johnson M.S. Farndale R.W. Heino J. Gullberg D. J. Biol. Chem. 2003; 278: 7270-7277Google Scholar, 23Kern A. Eble J. Golbik R. Kühn K. Eur. J. Biochem. 1993; 215: 151-159Google Scholar, 24Calderwood D.A. Tuckwell D.S. Eble J. Kühn K. Humphries M.J. J. Biol. Chem. 1997; 272: 12311-12317Google Scholar, 25Nykvist P. Tu H. Ivaska J. Käpylä J. Pihlajaniemi T. Heino J. J. Biol. Chem. 2000; 275: 8255-8261Google Scholar), beaded filament-forming type VI collagen (21Tulla M. Pentikäinen O. Viitasalo T. Käpylä J. Impola U. Nykvist P. Nissinen L. Johnson M. Heino J. J. Biol. Chem. 2001; 276: 48206-48212Google Scholar), and type XIII transmembrane collagen (25Nykvist P. Tu H. Ivaska J. Käpylä J. Pihlajaniemi T. Heino J. J. Biol. Chem. 2000; 275: 8255-8261Google Scholar). Instead, α2β1 integrin is the major receptor for type I collagen and other fibril-forming collagens (21Tulla M. Pentikäinen O. Viitasalo T. Käpylä J. Impola U. Nykvist P. Nissinen L. Johnson M. Heino J. J. Biol. Chem. 2001; 276: 48206-48212Google Scholar, 22Zhang W.M. Käpylä J. Puranen J.S. Knight C.G. Tiger C.F. Pentikäinen O.T. Johnson M.S. Farndale R.W. Heino J. Gullberg D. J. Biol. Chem. 2003; 278: 7270-7277Google Scholar, 24Calderwood D.A. Tuckwell D.S. Eble J. Kühn K. Humphries M.J. J. Biol. Chem. 1997; 272: 12311-12317Google Scholar, 25Nykvist P. Tu H. Ivaska J. Käpylä J. Pihlajaniemi T. Heino J. J. Biol. Chem. 2000; 275: 8255-8261Google Scholar, 26Gardner H.A. Broberg A. Pozzi A. Laato M. Heino J. J. Cell Sci. 1999; 112: 263-272Google Scholar). Currently, less is known about ligand binding preferences of the novel α10β1 and α11β1 integrins. However, our previous results show that α10I domain resembles α1I rather than α2I in its ligand binding specificity (21Tulla M. Pentikäinen O. Viitasalo T. Käpylä J. Impola U. Nykvist P. Nissinen L. Johnson M. Heino J. J. Biol. Chem. 2001; 276: 48206-48212Google Scholar), suggesting that it may not be a better receptor for fibrillar collagen than α1I domain. Integrin α11I domain seems to bind most strongly to type I collagen, but it is not clear whether it can mediate high affinity interactions (22Zhang W.M. Käpylä J. Puranen J.S. Knight C.G. Tiger C.F. Pentikäinen O.T. Johnson M.S. Farndale R.W. Heino J. Gullberg D. J. Biol. Chem. 2003; 278: 7270-7277Google Scholar). Integrin type collagen receptors do not differ only in their ability to recognize distinct collagen subtypes, but they also transmit divergent signals into the cell. This may explain why many cell types express several collagen receptors concomitantly. Structurally similar α1β1 and α2β1 integrins are connected to distinct signaling pathways, often leading to opposite cellular responses. Whereas α1β1 integrin seems to stimulate cell proliferation (27Pozzi A. Wary K.K. Giancotti F.G. Gardner H.A. J. Cell Biol. 1998; 142: 587-594Google Scholar, 28Ekholm E. Hankenson K.D. Uusitalo H. Hiltunen A. Gardner H. Heino J. Penttinen R. Am. J. Pathol. 2002; 160: 1779-1785Google Scholar) and act as a negative feedback regulator for collagen synthesis (26Gardner H.A. Broberg A. Pozzi A. Laato M. Heino J. J. Cell Sci. 1999; 112: 263-272Google Scholar, 29Riikonen T. Westermarck J. Koivisto L. Broberg A. Kähäri V.-M. Heino J. J. Biol. Chem. 1995; 270: 13548-13552Google Scholar, 30Langholz O. Rockel D. Mauch C. Kozlowska E. Bank I. Krieg T. Eckes B. J. Cell Biol. 1995; 131: 1903-1915Google Scholar), α2β1 integrin inhibits the growth of some cell types (31Koyama H. Raines E.W. Bornfeldt K.E. Roberts J.M. Ross R. Cell. 1996; 87: 1069-1078Google Scholar) as well as increases matrix synthesis and turnover (26Gardner H.A. Broberg A. Pozzi A. Laato M. Heino J. J. Cell Sci. 1999; 112: 263-272Google Scholar, 29Riikonen T. Westermarck J. Koivisto L. Broberg A. Kähäri V.-M. Heino J. J. Biol. Chem. 1995; 270: 13548-13552Google Scholar, 30Langholz O. Rockel D. Mauch C. Kozlowska E. Bank I. Krieg T. Eckes B. J. Cell Biol. 1995; 131: 1903-1915Google Scholar, 32Ivaska J. Reunanen H. Westermarck J. Koivisto L. Kähäri V.-M. Heino J. J. Cell Biol. 1999; 147: 401-416Google Scholar, 33Ravanti L. Heino J. López-Otín C. Kähäri V.M. J. Biol. Chem. 1999; 274: 2446-2455Google Scholar). The properties of the recently identified α10β1 and α11β1 integrins are just beginning to receive attention and still remain to be identified. Specific recognition sites on the triple helical structures of different collagen subtypes can be recognized by a group of integrin type cell adhesion receptors (13White D.J. Puranen S. Johnson M.S. Heino J. Int. J. Biochem. Cell Biol. 2004; 36: 1405-1410Google Scholar, 34Knight C. Morton L. Peachey A. Tuckwell D. Farndale R. Barnes M. J. Biol. Chem. 2000; 275: 35-40Google Scholar, 35Emsley J. Knight C. Farndale R. Barnes M. J. Mol. Biol. 2004; 335: 1019-1028Google Scholar). Although detailed information about integrin binding to the monomeric form of collagen is available, it has not been clear whether αI domains and the corresponding integrins can effectively support cell adhesion to complex collagen fibrils. The objective of this study is to provide further insight into α1β1 and α2β1 integrin interactions with the type I collagen fibril, the primary component of the extracellular matrix. Formation of Type I Collagen Fibrils—Bovine skin type I collagen (Vitrogen®, Cohesion, or Cellon S.A.) was diluted on ice with PBS 1The abbreviations used are: PBS, phosphate-buffered saline; CHO, Chinese hamster ovary; GST, glutathione S-transferase; HGF, human gingival fibroblast(s); MIDAS, metal ion-dependent adhesion site; BSA, bovine serum albumin. to final concentrations of 0.05, 0.1, or 0.5 mg/ml. To initiate collagen fibrillogenesis, samples were adjusted to physiological pH 7.4 with 1 m NaOH, and the temperature was increased from 4 to 37 °C. The progression of fibril formation was followed using a Beckman DU640 spectrophotometer by recording an optical density at 313 nm for 120 h. Finally, fibril quality was verified by transmission electron microscopy (JEM-1200EX, JEOL, Peabody, MA). For the electron microscopy analysis, fibrillogenesis of the samples was allowed to proceed between 60 min and 48 h at 37 °C. The temperature of monomeric samples was maintained below 4 °C. At each time point, collagen samples were spotted onto Formvar/carbon-coated copper electron microscopy grids for 1 h at 37 °C. Specimens were then fixed with 4% paraformaldehyde and 0.1% glutaraldehyde for 30 min at room temperature. Grids were embedded with 1.5% methyl cellulose and stained with 0.4% uranyl acetate for 10 min on ice. The Production of the Human Recombinant Integrin α1I and α2I Domains as GST Fusion Proteins—cDNAs encoding α1I and α2I domains were generated by PCR as described earlier (25Nykvist P. Tu H. Ivaska J. Käpylä J. Pihlajaniemi T. Heino J. J. Biol. Chem. 2000; 275: 8255-8261Google Scholar) using human integrin α1 and α2 cDNAs as templates. Vectors pGEX-4T-3 and pGEX-2T (Amersham Biosciences) were used to generate recombinant N-terminal GST fusion proteins of integrin α1I and α2I domains, respectively. For the protein production, transformed Escherichia coli BL21 cells were grown at 37 °C in LB medium containing 100 μg/ml ampicillin until the A600 of the suspension reached 1–2. Next, the protein overexpression was induced with 0.4 mm isopropyl-β-d-thiogalactopyranoside, and it was allowed to proceed for an additional 4–6 h at room temperature before harvesting the bacterial cells by centrifugation (5000 × g, 10 min, 4 °C). Pelleted cells were resuspended in PBS (pH 7.4) and then lysed by sonication. To permeabilize cell membranes, the disrupted bacterial cells were incubated with 1% Triton X-100 for 30 min on ice, and soluble proteins were isolated from cell debris by centrifugation (34,500 × g, 10 min, 4 °C). Glutathione-Sepharose (Amersham Biosciences) was added to the supernatant, and the mixture was incubated at room temperature for 30 min with gentle agitation. Glutathione-Sepharose beads with specifically bound GST fusion proteins were then collected by low speed centrifugation (500 × g, 5 min, 4 °C), and the slurry was transferred to disposable chromatography columns (Bio-Rad). The columns were washed with PBS, and fusion proteins were eluted with 30 mm reduced glutathione (Sigma). GST-tagged α1I and α2I were then analyzed by native polyacrylamide gel electrophoresis. The recombinant α1I domain contains 227 amino acids, corresponding to amino acids 123–338 of the whole α1 integrin subunit, whereas the recombinant α2I domain is 223 amino acids in length, consisting of amino acids 124–339 of the full-length α2 integrin subunit. The C termini of the α1I and α2I domains contain 10 and 6 nonintegrin amino acids, respectively. Collagen Binding Assay for Recombinant α1I and α2I Domains—To analyze the integrin binding to collagen, universal binding 96-well microtiter plates (Costar) were coated with either the monomeric or fibrillar form of type I collagen (0.5 mg/ml; Cellon S.A.) by 3.5-min exposure to UV light. Fibrillogenesis of type I collagen was allowed to proceed for 1 h at 37 °C before coating, whereas the monomeric state was maintained by keeping the sample on ice. Control wells were treated with Delfia® Diluent II containing BSA (Wallac) diluted 1:2 in PBS. Later the same solution was used for blocking the nonspecific protein absorption sites on all wells for 1 h at 37 °C. Next, GST-α1 and GST-α2 fusion proteins, at concentrations between 10 and 500 nm in Delfia® assay buffer (Wallac), were allowed to bind to collagen for 1 h at 37 °C in the presence of 2 mm MgCl2. All of the solutions used thereafter were supplemented with 2 mm MgCl2. Next, unbound GST-αI domains were removed by repeated PBS wash steps, and europium-labeled Delfia® GST antibody (Wallac), diluted depending on the preparation at 1:800 or 1:2000 in Delfia® assay buffer (Wallac), was added to the wells. After 1 h of incubation at 37 °C, specimens were rinsed with PBS, and finally Delfia® enhancement solution (Wallac) was added to each well in order to dissociate the fluorescent europium label. Europium signal was determined by time-resolved fluorometry (Victor2 multilabel counter; Wallac). Each assay was performed at least in triplicate, and approximate dissociation constants were obtained by fitting the data to a Michaelis-Menten form equation. Statistical analysis was performed using two-way analysis of variance. The significance of difference between αI domain binding to the monomeric and to the fibrillar form of type I collagen was analyzed using two independent variables: the form of the collagen and αI domain concentration. Statistical significance was assigned at p < 0.05. Immunoelectron Microscopy Analysis of α1I and α2I Domain Binding to Collagen Fibrils—Type I collagen fibrillogenesis (0.1 mg/ml Vitrogen®, Cohesion) was initiated as described above. Following 48 h of fibril formation, type I collagen fibrils were spotted onto Formvar/carbon-coated copper electron microscopy grids for 1 h at 37 °C. In order to prevent the nonspecific protein binding, the residual binding sites on grids were blocked with 5% milk powder in PBS for 15 min at room temperature. Then GST-α2I at concentrations of 10, 30, 100, and 300 nm was allowed to bind to collagen fibrils attached to the grids for 1 h at 37 °C in the presence of 2 mm MgCl2. Before fixing the samples with 4% paraformaldehyde for 30 min at room temperature, unbound GST-α2I was removed by washing the samples with PBS containing 2 mm MgCl2. Next, specimens were first rinsed with PBS, and then collagen-bound GST-α2I domains were detected with the antibody against GST (Amersham Biosciences) diluted at 1:400 in 5% milk powder in PBS. After 1 h, the specimens were carefully washed with PBS, and bound primary antibody was localized with protein A-gold conjugate (∼10 nm; a kind gift from Dr. George Posthuma, University Medical Center Utrecht) diluted 1:65 in 5% milk powder in PBS. Following secondary antibody attachment for 30 min at room temperature, samples were washed with PBS and distilled water. Finally, grids were embedded with 1.5% methyl cellulose and stained with 0.4% uranyl acetate on ice for 10 min. Specimens were examined by transmission electron microscopy (JEM-1200EX; JEOL, Peabody, MA), and average numbers of bound GST-α2I or GST control per collagen fibril D-period were counted from representative micrographs. Finally, the dissociation constant for GST-α2I domain binding to fibrillar collagen was estimated by fitting the data to the Michaelis-Menten form equation. GST was used as the background control. Integrin α1I and α2I Domains in Collagen Fibrillogenesis—In order to examine the effects of GST-αI domains on type I collagen fibrillogenesis, fibril assembly was monitored in the presence of GST-α1I or GST-α2I using a Beckman DU 640 spectrophotometer. Changes in turbidity were recorded at 313 nm at intervals of 1 min for 120 min. Fibrillogenesis of 330 μm type I collagen (0.1 mg/ml; Vitrogen) was analyzed in the presence of either 70 μm α1I-GST or α2I-GST at concentrations of 330, 165, 110, 80, 70, 30, and 15 μm. The effect of GST-α1I was analyzed both in the presence of 2 mm MgCl2 and in conditions where Mg2+ was made unavailable for protein binding using 2 mm EDTA as a chelating agent. Cell Cultures—Chinese hamster ovary (CHO) cells, obtained from the ATCC (Manassas, VA), were used as hosts for transfection and expression of wild type integrin α1 or α2 subunits (25Nykvist P. Tu H. Ivaska J. Käpylä J. Pihlajaniemi T. Heino J. J. Biol. Chem. 2000; 275: 8255-8261Google Scholar) as well as a D219N/D292N double mutant α2 subunit. Transfected cells were maintained in α-minimum essential medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mm glutamine, 100 IU/ml penicillin-G, 100 μg/ml streptomycin, and 0.5 mg/ml neomycin analogue G418. The human osteosarcoma cell line, Saos-2, was obtained from the ATCC, and it was also transfected to express integrin α2 subunit. Transfected cells were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (Invitrogen), 2 mm glutamine, 100 IU/ml penicillin G, 100 μg/ml streptomycin, and 0.25 mg/ml G418. Wild type human gingival fibroblasts (HGF) were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 2 mm glutamine, 100 IU/ml penicillin G, and 100 μg/ml streptomycin. Stable Transfections—Integrin α1 cDNA (15Briesewitz R. Epstein M.R. Marcantonio E.E. J. Biol. Chem. 1993; 268: 2989-2996Google Scholar) was a kind gift from Dr. E. Marcantonio (Columbia University, New York, NY), and integrin α2 cDNA (14Takada Y. Hemler M.E. J. Cell Biol. 1989; 109: 397-407Google Scholar) was from Dr. M. Hemler (Dana-Farber Cancer Research Center, Boston, MA). Integrin α1 cDNA was ligated into a modified pLEN expression vector (15Briesewitz R. Epstein M.R. Marcantonio E.E. J. Biol. Chem. 1993; 268: 2989-2996Google Scholar), and integrin α2 cDNA was ligated into the pAWneo2 expression vector (36Ohashi P. Mak T. Van den Elsen P. Yanagi Y. Yoshikai Y. Calman A. Terhorst C. Stobo J. Weiss A. Nature. 1985; 316: 606-609Google Scholar), which carries the spleen focus-forming virus long terminal repeat promoter and a neomycin resistance gene (a kind gift from Dr. A. Weiss, University of California, San Francisco, CA). Two site-specific point mutations, D219N and D292N, were introduced into full-length α2 cDNA, using the Stratagene QuikChange mutagenesis kit, before the construct was transfected into CHO cells (37Käpylä J. Ivaska J. Riikonen R. Nykvist P. Pentikäinen O. Johnson M. Heino J. J. Biol. Chem. 2000; 274: 3348-3354Google Scholar). CHO cells were transfected by electroporation (0.3 kV, 960 microfarads, 0.4-cm cuvette in RPMI plus 1 mm sodium pyruvate, 2 mm l-glutamine, without serum) with 20 μg of pAWneo2 carrying either wild type or mutated α2 cDNA or with 20 μg of pLEN vector carrying α1 cDNA. Construct containing α1 integrin was co-transfected with 1 μg of pAWneo2 null expression vector (25Nykvist P. Tu H. Ivaska J. Käpylä J. Pihlajaniemi T. Heino J. J. Biol. Chem. 2000; 275: 8255-8261Google Scholar). Transfection of Saos-2 cells with the α2-pAWneo2 construct was performed using Lipofectin reagent (Invitrogen) essentially following the manufacturer's recommendations. Control cells were transfected with pAWneo2 null expression vector. Transfected cells were then plated and allowed to recover for 1 day in culture medium. Neomycin analogue G418 (Invitrogen) was added to the culture medium at 0.4–0.5 mg/ml, and G418-resistant clones were selected for 1–3 weeks before they were isolated and analyzed for their expression of α1 or α2 integrin. The cell surface expression levels of the integrins were checked by flow cytometry using antibodies against the integrin α subunit (SR-84 antibody for α1 integrin, a gift from Dr. W. Rettig (Boehringer Ingelheim) and 12F1 antibody for α2 integrin, a gift from Dr. V. Woods (University of California, San Diego) (25Nykvist P. Tu H. Ivaska J. Käpylä J. Pihlajaniemi T. Heino J. J. Biol. Chem. 2000; 275: 8255-8261Google Scholar). Cell Spreading Assays—Universal binding 96-well microtiter plates (Costar) were precoated with either monomeric or fibrillar type I collagen (0.5 mg/ml; Cellon S.A.) by exposing plates to UV light for 3.5 min. Before coating, fibril formation was allowed to proceed for 1 h at 37 °C. Alternatively, wells were coated by allowing collagen fibril formation (0.05 mg/ml) to proceed for 48 h in the wells of a 96-well plate (Nunc) at 37 °C in a humid chamber. To maintain the monomeric state of the collagen solution, samples were kept on ice. Control wells were treated with PBS. The cell lines used were CHO cells transfected to express either α1β1 integrin (CHO-α1β1) or α2β1 integrin (CHO-α2β1), transfected Saos-2 cells, and wild type HGF cells. Semiconfluent cell cultures were detached with 0.01% trypsin and 0.02% EDTA for 3 min at 37 °C. Trypsin activity was inhibited by washing the cells either with 0.2% soybean trypsin inhibitor (Sigma) in a serum-free medium or with serum-supplemented medium. Serum was carefully removed with repeated washing steps when necessary, and cells were suspended in serum-free medium containing 50 μm cycloheximide (Sigma) to prevent de novo protein synthesis. Next, 26,000 CHO cells/cm2, 30,000 Saos-2 cells/cm2, or 15,000 HGF cells/cm2 were added to the collagen-coated microtiter plates. Nonspecific binding sites on all wells were blocked with 0.1% heat-inactivated BSA in PBS for 1 h at 37 °C before the addition of cells. After 2 h of cell spreading at 37 °C, medium containing non-adhered cells was poured out, and cells were fixed with 4% form-aldehyde and 5% sucrose for 30 min at room temperature. Cells in 16 representative fields from four replicate wells were analyzed using phase-contrast microscopy. The total number of cells attached and the percentage of spread cells were counted. A spread cell was characterized as one having a clearly visible ring of cytoplasm around the nucleus. The morphology of cells was further analyzed, and the percentage of cells that formed more than one long projection was determined. The statistical difference between cell spreading on monomeric or fibrillar type I collagen was analyzed using two-way analysis of variance. Three independent experiments were performed, and statistical significance was set at p < 0.05. Immunostaining of Integrin β1 Subunits—Fibrillogenesis of type I collagen (Vitrogen®, Cohesion) was allowed to proceed for 48 h at 37 °C on the surface of glass coverslips. Human gingival fibroblasts were prepared for the spreading experiment as described above. Following 120 min of cell spreading, cells were fixed with 4% formaldehyde and 5% sucrose in PBS containing 1 mm CaCl2 and 0.5 mm MgCl2 (PBS+) for 20 min. After washing, cells were permeabilized with 0.5% Triton X-100 in PBS+ for 4 min. Cells were then washed three times with PBS, and nonspecific binding sites were blocked with 3% BSA and 1% glycine in PBS+ for 30 min. To localize the β1-containing integrins on spread HGF cells, a polyclonal antibody against the cytoplasmic domain of integrin β1 subunit (4080; a generous gift fr
Referência(s)