Results

Distribution of laminin 4 in the skin and its synthesis in the dermal fibroblasts

Laminin 4 chain has a tandem repeat of five LG modules, LG1–5, at the C-terminus. Processing of the 4 LG4–5 module is known to occur in tissues and in cultured cells. Thus, we used two antibodies for immunostaining, anti-4 LG1–3 and anti-4 LG4–5 antibodies, which recognize different modules Figure 1a. Using frozen human skin sections, both antibodies gave similar results and stained the basement membrane zones of blood vessels and fibroblast-like cells in the dermis. Thus, both the unprocessed and processed laminin 4 chains are localized in blood vessel basement membrane and are associated with the fibroblast-like cells. The latter finding was unexpected since the laminin molecules are known to be expressed by epithelial cells and there is no organized basement membrane adjacent to fibroblasts. Strong staining of the entire epidermis was observed only with the anti-4 LG1–3 antibody (Figure 1a(a)) and not with the anti-4 LG4–5 antibody (Figure 1a(d)).

Figure 1.

Laminin 4 localization in the skin and its synthesis by dermal fibroblasts. (A) Immunofluorescence staining of the laminin 4 chain in adult human skin. Antibodies against 4 LG1–3 were used for (a–c). Antibodies against 4 LG4–5 were used for (d–f). In (a) and (d), there was positive staining in the dermis. Positive signals were detected surrounding fibroblast-like cells (b, e) and are located at blood vessel walls (c, f). Original magnification: 200 in (a, d) and 1000 in (b, c, e, f). (B) Conditioned media from human dermal fibroblasts were applied on 5% SDS-PAGE, under reducing (lanes 2, 4, 6) or non-reducing conditions (lanes 1, 3, 5), followed by western blotting. The major band of 175 kDa in lane 2 is a processed laminin 4 chain. The faint band (^*) is an unprocessed 4 chain. In lane E, 4 g of laminin-1 (111) from EHS sarcoma was applied and stained by Coomassie brilliant blue after blotting. 1, 1, and 1 and molecular size markers are on the left side. The antibodies used are: lanes 1, 2, anti-4LG1–3; lanes 3, 4, anti-4LG4–5; lanes 5, 6, anti-laminin 1 chain. 2ME (+) or (-) indicates the presence or absence of 2% mercaptoethanol, respectively.

Full figure and legend (106K)

We next analyzed whether human dermal fibroblasts synthesized the laminin 4 chain. Western blot analysis of conditioned media with anti-4 LG1–3 antibody revealed a major 175 kDa species and a minor 200 kDa species which corresponds to the processed and unprocessed laminin 4 chains, respectively (Figure 1b, lane 2). Anti-4 LG4–5 antibody recognized a 200 kDa species corresponding to the unprocessed laminin 4 chain (Figure 1b, lane 4). The laminin 4 chain is a component chain of both laminin-8 (411) and laminin-9 (421). Therefore, we also used an anti-laminin 1 antibody for western blot of the conditioned media. Anti-laminin 1 antibody recognized a 200 kDa species (Figure 1b, lane 6). Under non-reducing conditions, the bands containing processed and unprocessed laminin 4 chain migrated slower than the laminin 1 chain (about 400 kDa). These observations suggest that dermal fibroblasts secrete a laminin heterotrimer containing 4 chains.

Cell adhesion activity of the laminin 4 LG4 module

Recombinant human laminin 4 chain LG4 module (rec-LG4) was expressed in 293T cells as a fusion protein with an IgG Fc portion at the C-terminus. The purified rec-LG4 was analyzed by SDS-PAGE and western blot and found to be pure band with a predicted molecular weight of 60 kDa Figure 2a. The rec-LG4 proteins were coated as substrates for cell adhesion assays Figure 2b. Dermal fibroblasts attached to rec-LG4 in a concentration-dependent fashion. Cell adhesion activity of rec-LG4 showed the maximum level at 0.2 g per well and reached a plateau at more than 0.2 g per well. We next examined various glycosaminoglycans (GAGs) as competitors for cell adhesion Figure 2c. Fibroblast adhesion to rec-LG4 was inhibited by heparin. HS weakly blocked fibroblast attachment. No significant inhibition however was observed with HA, CSA, DS, and CSC. These results suggested that heparin and HS are involved in cell binding to the laminin 4 LG4 module. As we used the same concentration of GAGs (250 g per mL), the molar concentration of heparin to other GAGs ranged from about 1:1 to 1:10. The inhibition assay with the lower concentration of heparin (10 g per mL) showed complete blockage of cell adhesion to 4 LG4 and to HA4G82 (data not shown). The higher concentration (250 g per mL) was employed to obtain the inhibitory effect of HS in the cell adhesion assay (^{Utani et al, 2001}). These data indicate that the higher potency observed with heparin is based on charge density and not molar concentration.

Figure 2.

Heparin- and HS-dependent cell adhesion activity of the recombinant 4 LG4 module. (A) Expression and purification of recombinant 4 LG4 protein. Purified recombinant proteins were applied on 10% SDS-PAGE. Under reducing condition lane 1, Coomassie brilliant blue staining; lane 2, Western blotting. Molecular size markers are on the left side. (B) Fibroblasts were seeded on different amounts of substrates, including rec-LG4, control recombinant protein (rec-con, human laminin 2 N-terminal 617 amino acids), and collagen I. The bound cells were stained with crystal violet and the attached cells in three randomly selected fields (200) were counted. Each value represents the mean SD of triplicate experiments. (C) Fibroblasts were seeded on rec-LG4 (0.2 g per well) in the presence of 250 g per mL of various GAGs and attached cells were counted. The cell numbers without competitors are taken as 100%. Each value represents the mean SD of four experiments. Hp, heparin; HA, hyaluronic acid; HS, heparan sulfate; CSA, chondroitin sulfate A; DS, dermatan sulfate; CSC, chondroitin sulfate C. Statistical analyses were performed using an unpaired Student's t test and ^*depicts p<0.01.

Full figure and legend (54K)

Identification and characterization of the sequence active for cell adhesion

We prepared 19 synthetic peptides covering the entire human LG4 module Table I. Peptides HA4G82, HA4G83, and HA4G90 strongly inhibited fibroblast adhesion to rec-LG4, whereas the other synthetic peptides showed no inhibition Figure 3a. None of the three peptides competed for fibroblast cell attachment to type I collagen (data not shown). HA4G82, HA4G83, and HA4G90 were also active for cell attachment when they were coated as substrates Figure 3b. As expected, the rest of the peptides did not promote fibroblast attachment (data not shown). The activity of HA4G90 was much weaker than that of HA4G82 and HA4G83. It was reported that the mouse 4 LG4 module has heparin-binding activity (^{Yamaguchi et al, 2000}). The heparin-binding sequence of mouse 4 LG4 corresponds to the HA4G82 sequence of human 4 LG4. Thus, we focused on HA4G82 as a critical sequence within 4 LG4 since fibroblast adhesion to human 4 LG4 was heparin-dependent. Cell adhesion to HA4G82 was inhibited by heparin and HS, and not by other GAGs Figure 3c. Since heparin binds to basic residues, His and Arg residues were substituted in the sequence TLFLAHGRLVFM with Ser (HA4G82S: TLFLASGSLVFM). Dermal fibroblasts attached to the HA4G82 but did not bind to the serine-substituted peptide, HA4G82S Figure 3d. Peptides HA4G82 strongly inhibited fibroblast adhesion to rec-LG4, whereas HA4G82S showed no inhibition activity Figure 3d. These results indicated that both the His and Arg residues within the HA4G82 sequence, which are predicted to be essential for heparin binding, were critical for the cell adhesion activity of laminin 4 LG4.

Figure 3.

Effect of synthetic peptides on cell adhesion to the recombinant 4 LG4 module and cell attachment activity of the peptides. (A) 96-well microplates were coated with 0.2 g per well of rec-LG4. The synthetic peptides (100 g per mL) were added to wells with the cells during the incubation. (B) Fibroblasts were seeded on different amounts of the substrates, HA4G82, HA4G83, and HA4G90. (C) Fibroblasts were seeded on HA4G82 (3.8 g per mL) in the presence of 250 g per mL of various GAGs. (D) Adhesion to HA4G82 and to HA4G82S (5 g per well) was determined (columns 1, 2). The inhibition assay was performed using 0.5 g per well of rec-LG4 (columns 3–5). HA4G82 and HA4G82S (100 g per mL) were added to the wells with the cells during the incubation. The cell numbers without competitors are taken as 100% (A, C, columns 3–5 in D). In columns 1, 2 in (D), the cell numbers attached to the HA4G82 are taken as 100%. The bound cells were stained with crystal violet and the attached cells in randomly selected three fields (200) were counted. Each value represents the mean SD of multiple independent experiments. ^*depicts p<0.01 in (C) and (D).

Full figure and legend (55K)

Table I - List of synthetic peptides.

Full table

293T cells overexpressing syndcan-2 and syndecan-4, but not glypican-1, bind to laminin 4 LG4

The rec-LG4 protein and the HA4G82 peptide promoted the heparin- and HS-dependent fibroblast attachment, suggesting that the cell attachment is mediated by membrane-associated heparan sulfate proteoglycans (HSPGs). Syndecans and glypican are well characterized as cell surface HSPGs. Further, syndecan-2 and -4 are expressed in fibroblasts and are known to be involved in cell adhesion. Therefore, we focused on these molecules as candidates mediating cell adhesion. Because 293T cells showed no adhesion activity to rec-LG4 and HAG82, syndecan-2 and -4 were overexpressed in 293T cells. Further, glypican-1, which is a cell surface HSPG lacking a cytoplasmic domain, was also stably expressed in 293T cells. FACS analysis demonstrated the levels of HS chains in the syndecan-2-, -4-, and glypican-1-overexpressing 293T clones Figure 4a. Syndecan-2- and syndecan-4-overexpressing cells bound to rec-LG4 and HAG4G82. In contrast, glypican-1-expressing cells failed to bind to the substrates Figure 4b. The HS level of the glypican-1 clone was rather higher than that of syndecan-2 or -4 clones, indicating that the HS chain of glypican-1 is not sufficient for cell adhesion to 4 LG4. Various GAGs as competitive inhibitors were tested for their effects on cell adhesion by the syndecan-2- and -4-expressing cells. Attachment of both cell preparations to rec-LG4 and HA4G82 was inhibited by heparin and HS. No significant inhibition however was observed with other GAGs (Figure 4c, d). Syndecan-2- and syndecan-4-expressing cells were able to bind to HA4G82, but were not able to bind to HA4G82S Figure 4e. These results indicate that the rec-LG4 protein and the HA4G82 peptide specifically interact with the HS chains of syndecan-2 and syndecan-4 and promote cell attachment.

Figure 4.

Syndecan-2- and syndecan-4-, but not glypican-1-, overexpressing 293T cells adhere to recombinant 4 LG4 and HA4G82. (A) FACS analysis with 10E4 monoclonal antibody for HS of the parental 293T cells and 293T cells overexpressing syndecan-2 (Syn-2), syndecan-4 (Syn-4), and glypican-1 (GP-1). The solid lines represent the basal level of HS in the parental 293T cells and the gray areas represent HS fluorescence in the 293T cells. (B) Syn-2, Syn-4, and GP-1 cells, and parental 293T cell adhesion to either rec-LG4 (0.6 g per well) or HA4G82 (15 g per well). (C, D) Syn-2 and Syn-4 were seeded on rec-LG4 (C) or HA4G82 (D) in the presence of 250 g per mL of various GAGs. Hp, heparin; HA, hyaluronic acid; HS, heparan sulfate; CSA, chondroitin sulfate A; DS, dermatan sulfate; CSC, chondroitin sulfate C. (E) Syn-2 and Syn-4 adhesion to HA4G82 and its serine-substituted analog HA4G82S (15 g per well). In these experiments, the attached cells in three randomly selected fields (200) were counted. The cell numbers without competitors are taken as 100% (C, D). Each value represents the mean SD of three experiments. ^*depicts p<0.01 in (C) and (D).

Full figure and legend (82K)

Discussion

Here, we have determined the location, function, and cellular receptors of laminin 4 chain in the skin. Our immunofluorescence study demonstrated that processed and unprocessed laminin 4 chains are located in the basement membrane zone of capillary vessels and in an area adjacent to fibroblast-like cells. The numbers of positively stained capillary vessels and fibroblast-like cells are similar to either anti-LG4–5 or anti-LG1–3. In contrast, anti-LG4–5 could not detect the laminin 4 chain in other tissues, such as capillary walls, adipose, and neural tissues (^{Talts et al, 2000}). These results indicate that the degree of processing of the 4 chain is different in the distinct tissues where it is located. It was unexpected that fibroblast-like cells had a positive signal for the laminin 4 chain. An electron microscopic study showed that the laminin 4 chain was not a constituent of endothelial basement membrane but was deposited in adjacent extracellular regions (^{Talts et al, 2000}). The laminin 4 chain surrounding dermal fibroblasts may have a new function. Previously, molecules recognized by anti-EHS laminin antibody (laminin-1, 111) were found in fibroblast conditioned media (^{Woodley et al, 1988}). Furthermore, to verify the presence of a laminin trimer with the laminin 4 chain in fibroblast-like cells, we performed western blotting. By reverse transcriptase-PCR, mRNA expression of laminin 4 and 1, 2, 3, and 1 was also detected (data not shown). Therefore, it is possible that laminin-8 (411) and/or -9 (421) is produced by dermal fibroblasts. Thus, fibroblasts may synthesize and secrete laminin-8 and/or -9, which are present in the adjacent matrix.

Previously, we identified several active sites within the mouse 4 LG1–5 module for cell adhesion (^{Okazaki et al, 2002}). HA4G82 corresponds to the mouse sequence TLFLAHGRLVFM, which is one of the active sites with heparin-dependent cell adhesion ability. The basic residues His and Arg in HA4G82, which were verified to be essential for cell adhesion by serine substitution in this report, are conserved between human and mouse. To identify the cell adhesion receptor for the sequence within HA4G82, we utilized 293T cells overexpressing syndecans for cell adhesion. Our data clearly demonstrate that syndecan-2 and -4 bind to 4 LG4 and function as adhesion receptors. We analyzed the cell surface HSPG by FACS analysis. The level of HSPG in glypican-1-overexpressing cells is higher than that in syndecan-2 or –4 overexpressing cells; however, the glypican-1-overexpressing cells failed to attached to 4 LG4. This observation confirmed that the interaction of the HS chain and laminin 4 LG4 module is necessary but not sufficient for cell adhesion. The protein core sequence is also probably required for this interaction. The well-conserved cytoplasmic domain in syndecans is probably required for mediating the signal from the cell surface LG4–syndecan interaction to the intracellular events.

Recently, analysis with 4-null mouse demonstrated that the 4 chain played a central role in microvessel growth (^{Thyboll et al, 2002}). In the syndecan-4 null mouse, skin wounds showed delayed healing and impaired capillary proliferation compared with that of the wild-type mouse (^{Echtermeyer et al, 2001}). Therefore, our observations suggest that cell adhesion to the laminin 4 LG4 module through syndecans may be important in angiogenesis in wounded skin. A precise understanding of the mechanisms of the signaling pathway from 4 LG4 binding to syndecan may contribute to clinical applications for the treatment of chronic cutaneous wounds.

Materials and Methods

Antibodies

Anti-4 LG1–3 and 4 LG4–5 antibodies were gifts of Dr Timpl (Max-Planck-Institut für Biochemie, Martinsried, Germany) (^{Talts et al, 2000}). Fluorescein isothiocyanate (FITC)-conjugated anti-rabbit IgG antibody was purchased from Jackson ImmunoResearch Laboratories (West Grove, Pennsylvania). Anti laminin -1 polyclonal antibodies (H-190) were obtained from Santa Cruz Biotechnology (Santa Cruz, California). FITC-conjugated anti-heparan sulfate (HS) (10E4) monoclonal antibody (mouse IgM, k) (Seikagaku, Tokyo, Japan) and isotype antibody, FITC, mouse IgM, k, isotype standard (anti-TNP) (PharMingen, San Diego, CA), were used for FACS analysis.

Cultured cells

Human neonatal dermal fibroblasts were obtained from Asahi Techno Glass (Tokyo, Japan). Fibroblasts were maintained in Dulbecco's MEM (DMEM, Invitrogen Life Technologies, Carlsbad, California) supplemented with 10% fetal bovine serum (FBS, Invitrogen Life Technologies), 100 U per mL penicillin, and 50 U per mL streptomycin. 293T cell clones producing syndecan-2, -4, and glypican-1 were maintained with 0.4 g per mL of puromycin (Sigma)/10%FBS/DMEM as previously described (^{Utani et al, 2001}).

Immunofluorescence

Frozen sections of adult human skin, obtained with informed consent, were air dried for 15 min at room temperature, washed with phosphate-buffered saline (PBS) for 15 min, and blocked with 5% normal donkey serum (Chemicon International, Temecula, California)/1% bovine serum albumin (BSA)/0.05% NP-40 (Sigma Chemical, St Louis, Missouri)/PBS for 15 min. Sections were incubated with either anti-4 LG1–3 or 4 LG4–5 antibodies (1:100) overnight at 4°C, and followed by FITC-conjugated anti-rabbit IgG antibody (1:200) for 30 min at 37°C. The specimens were analyzed by a Zeiss immunofluoresence microscope.

Western blotting

For western blotting, conditioned media were collected from fibroblasts in a 75 cm² flask after a 24 h incubation with DMEM. The media were extensively dialyzed against distilled water, lyophilized, and dissolved in 50 L of sample buffer (20% glycerol, 2% SDS, 125 mM Tris-HCl, pH 6.8, 2% mercaptoethanol, and 0.15% bromophenol blue), and 20 L of sample was separated on 5% SDS-PAGE and transferred by electroblotting onto PVDF transfer membranes (Immobilon, Millipore, Bedford, Massachusetts). The membranes were blocked with 10% non-fat milk in Tris-buffered saline (150 mM NaCl, 40 mM Tris-HCl, pH 7.4) with 0.1% Tween-20, probed with the first antibodies (1:200), and visualized by alkaline phosphatase-conjugated anti-rabbit IgG antibody (Jackson ImmunoResearch Laboratories) and 5-bromo-4-chloro-3-indolyl phosphatate/nitro blue tetrazolium (BCIP/NBT) (Moss, Pasadena, Maryland). Mouse laminin-1 from EHS sarcoma was a gift from Dr Kleinman (NIDCR, NIH, Bethesda, Maryland). For western blotting with anti-laminin 1 chain antibody, peroxidase-conjugated anti-rabbit IgG antibody and ECL (Amersham-Pharmacia Biotech, Uppsala, Sweden) were used.

Recombinant 4 LG4 module expression

cDNA was synthesized using mRNA from human fibroblasts prepared using the Superscript Preamplification System (Invitrogen Life Technologies) and amplified using the Expand High Fidelity PCR System (Roche Diagnostics, Basel, Switzerland). After amplification, the PCR product was cut with Avr II, and cDNA was inserted into a MO90 vector, which expresses a signal peptide and a human IgG Fc portion under the EF1- promoter. All constructs were confirmed by DNA sequencing using a thermo-sequence Cy5 dye terminator and an ALF express II DNA sequencer (Amersham Pharmacia Biotech). The primers for RT-PCR were as follows: rec-LG4 (4060–4640 nt of human laminin 4 chain, GenBanK: #11418562), forward 5'-GAGCCTAGGCCACCTTTCCAACAGCCCTAG-3' and backward 5'-GAGCCTAGGCCCACGTATCCTCCTCCTGTT-3'. Recombinant 4 LG4 was expressed as a chimera with a human IgG Fc portion at the C-terminus and purified as previously described with a minor modification (^{Utani et al, 2001}). Briefly, recombinant proteins were expressed in 293T cells by the Ca-P Transfection Kit (Invitrogen Life Technologies). After 24 h, cells were incubated with CHO medium (Invitrogen Life Technologies) for another 2 days, followed by purification with protein A–sepharose (Amersham-Pharmacia Biotech). The protein concentration was calculated with the BCA Protein Assay Kit (Pierce, Rockford, Illinois). Protein purity was determined by reducing 10% SDS-PAGE and Coomassie brilliant blue staining.

Synthesis of human laminin 4 LG4 peptides

Peptides comprising the human 4 LG4 module were manually synthesized by the 9-fluorenylmethoxycarbonyl (Fmoc)-based solid-phase method with a C-terminal amide as previously described (^{Utani et al, 2001}). Peptides were purified by reverse-phase high-performance liquid chromatography (HPLC) using a Mightysil RP-18 column (Kanto Chemical, Tokyo, Japan) with a gradient of water/acetonitrile containing 0.1% trifluoroacetic acid. Purity and identity were confirmed by HPLC and by fast atom bombardment mass spectral analysis at the GC-MS & NMR Laboratory, Graduate School of Agriculture, Hokkaido University.

Cell adhesion assays

Substrates included collagen I (Nitta gelatin, Tokyo, Japan), recombinant 4 LG4, human laminin 2 N-terminal 617 amino acids as a negative control recombinant protein (^{Utani et al, 2001}), and synthetic peptides. 96-well plates (Nalge Nunc International, Rochester, New York) were coated with recombinant proteins and collagen substrates in distilled water overnight at 4°C. Peptide substrates were air dried at room temperature. The wells were rinsed with distilled water, and blocked with 1% BSA (Sigma)/DMEM for 1 h at room temperature. Cells were trypsinized and recovered in 10% FBS/DMEM for 30 min at 37°C in 5% CO₂ in a humidified atmosphere followed by three rinses with 0.1% BSA/DMEM. Then, 100 L of the cell suspension (210⁵ per mL) in 0.1% BSA/DMEM was plated on the wells for 1 h at 37°C in 5% CO₂ in a humidified atmosphere. The wells were rinsed once with PBS and stained with 0.1% crystal violet in 20% methanol (v/v) for 10 min. The attached cells in three randomly selected fields (200) were counted. For competition assays, cells were preincubated for 10 min at room temperature with either synthetic peptides, or 250 g per mL of heparin (Hep), heparan sulfate (HS), chondroitin sulfate A (CSA), dermatan sulfate (DS), chondroitin sulfate C (CSC), or hyaluronic acid (HA) (Seikagaku Kogyo, Tokyo, Japan) and used for the cell adhesion assay as above. Only soluble peptides were assayed in these experiments. The molecular weights of GAGs used in this experiments are 10–20K (Hep), 11K (HS), 25–50K (CSA), 11–25K (DS), 40–80K (CSC), and 100–200K (HA).

FACS analysis

Monolayer cells were harvested by PBS containing 0.53 mM EDTA, re-suspended in 10%FBS/DMEM. Cells (510⁵ per sample) were stained with 20 g per mL FITC-conjugated anti HS (10E4) monoclonal antibody in 1% BSA/PBS. After a 30 min incubation on ice, cells were washed. Flow cytometry analysis was performed on a FACScaliber (Becton Dickinson, Franklin Lakes, NJ), and results were analyzed with CELLQuest software (Becton Dickinson). A 2 g per mL propidium iodide solution was used to determine the population of dead cells. Control staining was performed using isotype-matched antibody (anti-TNP) and there was no significant background staining with 10E4 monoclonal antibody (data not shown).

Statistical analyses

Statistical analyses were performed using an unpaired Student's t test with StatView J., Hulinks, Tokyo, Japan.

References

References

1.	Echtermeyer F, Streit M & Wilcox-Adelman S et al. Delayed wound repair and impaired angiogenesis in mice lacking syndecan-4. J Clin Invest (2001) 107: R9–R14. | PubMed | ISI | ChemPort |
2.	Fujiwara H, Kikkawa Y, Sanzen N & Sekiguchi K. Purification and characterization of human laminin-8. J Biol Chem (2001) 276: 17550–17558. | Article | PubMed | ISI | ChemPort |
3.	Geberhiwot T, Ingerpuu S & Pedraza C et al. Blood platelets contain and secrete laminin-8 (411) and adhere to laminin-8 via 61 integrin. Exp Cell Res (1999) 253: 723–732. | Article | PubMed | ISI | ChemPort |
4.	Iivanainen A, Kortesmaa J & Sahlberg C et al. Primary structure developmental expression, and immunolocalization of the murine laminin 4 chain. J Biol Chem (1997) 272: 27862–27868. | Article | PubMed | ISI | ChemPort |
5.	Kortesmaa J, Yurchenco P & Tryggvason K et al. Recombinant laminin-8 (411). J Biol Chem (2000) 275: 14853–14859. | Article | PubMed | ISI | ChemPort |
6.	Miner JH, Patton BL & Lentz SI et al. The laminin chains: Expression, developmental transitions, and chromosomal locations of 1-5, identification of heterotrimeric laminin 8-11 and cloning of a novel 3 isoform. J Cell Biol (1997) 137: 685–701. | Article | PubMed | ISI | ChemPort |
7.	Nielsen PK, Gho YS & Hoffman MP et al. Identification of a major heparin and cell binding site in the LG4 module of the laminin 5 chain. J Biol Chem (2000) 275: 14517–14523. | Article | PubMed | ISI | ChemPort |
8.	Okazaki I, Suzuki N & Nishi N et al. Identification of biologically active sequences in the laminin 4 chain G domain. J Biol Chem (2002) 277: 37070–37078. | Article | PubMed | ISI | ChemPort |
9.	Patton BL, Miner JH, Chiu AY & Sanes JR. Distribution and function of laminins neuromuscular system of developing, adult, and mutant mice. J Cell Biol (1997) 139: 1507–1521. | Article | PubMed | ISI | ChemPort |
10.	Talts JF, Mann K, Yamada Y & Timpl R et al. Structural analysis and proteolytic processing of recombinant G domain of mouse laminin 2 chain. FEBS Lett (1998) 426: 71–76. | Article | PubMed | ISI | ChemPort |
11.	Talts JF, Sasaki T & Miosge N et al. Structural and functional analysis of the recombinant G domain of the laminin 4 chain and its proteolytic processing in tissues. J Biol Chem (2000) 275: 35192–35199. | Article | PubMed | ISI | ChemPort |
12.	Thyboll J, Kortesmaa J & Cao R et al. Deletion of the laminin 4 chain leads to impaired microvessel maturation. Mol Cell Biol (2002) 22: 1194–1202. | Article | PubMed | ISI | ChemPort |
13.	Timpl R. Macromolecular organization of basement membranes. Curr Opin Cell Biol (1996) 8: 618–624. | Article | PubMed | ISI | ChemPort |
14.	Utani A, Momota Y & Endo H et al. Laminin 3 LG4 module induces matrix metalloproteinase-1 (MMP-1) through MAPK signaling. J Biol Chem (2003) 278: 34483–34490. | Article | PubMed | ISI | ChemPort |
15.	Utani A, Nomizu M & Matsuura H et al. A unique sequence of the laminin 3 G domain binds to heparin and promotes cell adhesion through syndecan-2 and -4. J Biol Chem (2001) 276: 28779–28788. | Article | PubMed | ISI | ChemPort |
16.	Utani A, Nomizu M & Sugiyama S et al. A specific sequence of laminin 2 chain critical for the initiation of heterotrimer assembly. J Biol Chem (1995) 270: 3292–3298. | Article | PubMed | ISI | ChemPort |
17.	Utani A, Nomizu M & Timpl R et al. Laminin chain assembly: Specific sequences at the C-terminus of the long arm are required for the formation of specific double- and triple-stranded coiled-coil structures. J Biol Chem (1994) 269: 19167–19175. | PubMed | ISI | ChemPort |
18.	Woodley DT, Stanley JR, Reese MJ & O'Keefe EJ. Human dermal fibroblasts synthesize laminin. J Invest Dermatol (1988) 90: 679–683. | Article | PubMed | ISI | ChemPort |
19.	Yamaguchi H, Yamashita H & Mori H et al. High and low affinity heparin-binding sites in the G domain of the mouse laminin 4 chain. J Biol Chem (2000) 275: 29458–29465. | Article | PubMed | ISI | ChemPort |

Acknowledgments

We thank Rupert Timpl and Takako Sasaki (Max-Planck-Institut für Biochemie, Martinsried, Germany) for generously providing antibodies and Dr Hynda Kleinman (NIDCR, NIH, Bethesda, Maryland) for critical suggestions. This work was supported by Grant-in-Aid 14570798 for Scientific Research from the Ministry of Education, Science, Culture and Sports of Japan (to A.U.).