Identification of Biologically Active Sequences in the Laminin α4 Chain G Domain*

Laminins are a family of trimeric extracellular matrix proteins consisting of α, β, and γ chains. So far five different laminin α chains have been identified. The laminin α4 chain, which is present in laminin-8/9, is expressed in cells of mesenchymal origin, such as endothelial cells and adipocytes. Previously, we identified heparin-binding sites in the C-terminal globular domain (G domain) of the laminin α4 chain. Here we have focused on the biological functions of the laminin α4 chain G domain and screened active sites using a recombinant protein and synthetic peptides. The rec-α4G protein, comprising the entire G domain, promoted cell attachment activity. The cell attachment activity of rec-α4G was completely blocked by heparin and partially inhibited by EDTA. We synthesized 116 overlapping peptides covering the entire G domain and tested their cell attachment activity. Twenty peptides showed cell attachment activity, and 16 bound to heparin. We further tested the effect of the 20 active peptides in competition assays for cell attachment and heparin binding to rec-α4G protein. A4G6 (LAIKNDNLVYVY), A4G20 (DVISLYNFKHIY), A4G82 (TLFLAHGRLVFM), and A4G83 (LVFMFNVGHKKL), which promoted cell attachment and heparin binding, significantly inhibited both cell attachment and heparin binding to rec-α4G. These results suggest that the four active sites are involved in the biological functions of the laminin α4 chain G domain. Furthermore, rec-α4G, A4G6, and A4G20 were found to interact with syndecan-4. These active peptides may be useful for defining of the molecular mechanism laminin-receptor interactions and laminin-mediated cellular signaling pathways.

The laminin ␣ chains are generally large (M r ϭ 400,000) and contain a C-terminal globular domain consisting of five globular modules LG1-LG5. The laminin ␣4 chain lacks the N-terminal short arm and is expressed in cells of mesenchymal origin, such as endothelial cells and adipocytes (23)(24)(25). Laminin ␣4 chain expression is mainly localized to mesenchymal cells present in the lung and cardiac and skeletal muscles fibers (26). The ␣4 chain is also weakly expressed in other adult tissues, such as brain, spleen, liver, kidney, and testis (26). The laminin ␣4 chain may play a critical biological role in the tissues. Proteolytic processing of the laminin ␣4 chain G domain was confirmed with cultured endothelial and Schwannoma cells, where the Cterminal ␣4 LG4 -5 module was released (27). The LG4 -5 fragment was detected in cell cultures but not in tissues by immunostaining. Laminin-8 containing the ␣4 chain bound to ␣ 6 ␤ 1 and ␣ 3 ␤ 1 integrins (28,29). The ␣4 chain G domain bound to heparin, sulfatides, and fibulins but has relatively low affinity for ␣-dystroglycan receptors compared with other laminin ␣ chains (27). By using recombinant proteins, we previously showed that the laminin ␣4 chain G domain bound to heparin, and this affinity was stronger than that of the ␣1 chain G domain (30). Recently, we have identified the heparin-binding sites on the laminin ␣1, ␣3, and ␣5 chains LG4 modules (14,16,18,31). These peptides interacted with syndecan-1 or -2, a membrane-associated proteoglycan, and promoted cell attachment. Heparin binding may be important for the biological activity of the laminin ␣4 chain.
In this paper, we describe the systematic screening for biologically active sequences in the laminin ␣4 chain G domain (mouse laminin ␣4 chain 852-1816) using a recombinant protein and a large set of overlapping peptides. The laminin ␣4 chain G domain recombinant protein promoted cell attachment and heparin binding. For the initial stage of screening for identification of active sequences, we evaluated the cell attachment activities of 104 different peptides using peptide-conjugated Sepharose beads and peptide-coated plates. We also examined the effect of these peptides on heparin binding to the recombinant protein. Four sequences were identified that were active in all of the assays and were also evaluated for additional biological activities. Two were found to interact with syndecan-4.

MATERIALS AND METHODS
Recombinant Protein (rec-␣4G)-A recombinant protein (rec-␣4G), containing the mouse laminin ␣4 chain G domain (residues 852-1816) with the c-Myc sequence at the C terminus, was expressed using Chinese hamster ovary cells as described previously (30). Conditioned medium of rec-␣4G expressing cells was collected, and the rec-␣4G protein was purified using a heparin affinity column (HiTrap, Amersham Biosciences). Purity was confirmed by SDS-PAGE and by Western blotting analysis. Protein concentration was determined with the BCA assay (Pierce) with BSA 1 as standard.
RKRLQVQLSIRT ϩϩϩ ϩϩϩ N/D N/D ϩϩϩ a For cell attachment assays, various amounts of peptides were coated on 96-well plates as described under "Materials and Methods." In all cases, the biological activities of the peptides were quantitated and evaluated relative to those observed with AG73. Cell attachment was evaluated on the following subjective scale: ϩϩϩ, adhesion comparable to those on AG73; ϩϩ, weak adhesion compared with that on AG73; ϩ, very weak adhesion compared with that on AG73; Ϫ, no adhesion. Triplicate experiments gave similar results.
b For rec-␣4G cell attachment inhibition assays, wells were coated with rec-␣4G (1.2 g/well) on 96-well plates, and cells were preincubated with 0.1 mg/ml of each peptide as described under "Materials and Methods." Inhibition of cell attachment was evaluated on the following subjective scale: ϩϩ, completely inhibited rec-␣4G; ϩ, moderate inhibited rec-␣4G; Ϫ, no effect on rec-␣4G. N/D, not determined. Triplicate experiments gave similar results.
c For rec-␣4G heparin binding inhibition assays, heparin binding was evaluated as shown in Fig. 7. ϩϩ, completely inhibited rec-␣4G; ϩ, moderate inhibited rec-␣4G; Ϫ, no inhibition. N/D, not determined. IC 50 values (Fig. 7) are indicated by parentheses. Triplicate experiments gave similar results. d For solid phase heparin binding assay, various amounts of peptides were coated on 96-well plates as described under "Materials and Methods." In all cases, the biological activities of the peptides were quantitated and evaluated relative to those observed with AG73 as shown in Fig. 8. Heparin binding activities was evaluated on the following subjective scale: ϩϩϩ, binding comparable to that on AG73; ϩϩ, weak binding compared with that on AG73; ϩ, very weak binding compared with that on AG73; Ϫ, no binding. Triplicate experiments gave similar results.
Cell Attachment Assay Using Recombinant Protein and Synthetic Peptides-Cell attachment to substrate-coated plates was assayed in 96-well plates (Nunc Inc., Naperville, IL). Plates were coated with various amounts of rec-␣4G in Milli-Q H 2 O (50 l) at 4°C overnight. Peptides were also coated on the plates by drying overnight. The substrate-coated wells were blocked by addition of 1% bovine serum albumin (BSA, Sigma) in DMEM (100 l) for 1 h and then washed twice with DMEM containing 0.1% BSA. Cells, detached by 0.02% EDTA in phosphate-buffered saline and resuspended in DMEM containing 0.1% BSA, were added (20,000 cells/100 l) to each well and incubated at 37°C for 1 h in 5% CO 2 . The attached cells were stained with 0.2% crystal violet aqueous solution in 20% methanol for 10 min. After removal of the unattached cells, 1% SDS (200 l) was used to dissolve the attached cells, and the absorbance at 570 nm was measured in a model 550 Microplate reader (Bio-Rad).
In heparin, EDTA, and peptide inhibition experiments, HT-1080 cells were preincubated with heparin (10 g/ml), EDTA (5 mM), or peptide (0.1 mg/ml) at 37°C for 10 min and then plated. After 30-min, the attached cells were measured as described above. All assays were carried out in triplicate, and each experiment was repeated at least three times.
Cell Attachment Assay Using Peptide-conjugated Sepharose Beads-The 104 soluble peptides were coupled to cyanogen bromide (CNBr)activated Sepharose 4B (Amersham Biosciences) as described previously (33). Peptide solutions (200 g, 1 mg/ml in Milli-Q H 2 O) were mixed with activated Sepharose beads (30 mg). Ethanolamine-coupled beads were prepared as a control. The amount of coupled peptide was determined by amino acid analysis (10 -20 mol peptides per 1 g of Sepharose beads) (33).
Cell attachment to the peptide-Sepharose beads was assayed in 48-well plates (Iwaki, Tokyo, Japan). HT-1080 cells were detached as described above. The cells (100,000 cells/200 l of DMEM containing 0.1% BSA) were incubated with 200 l of the bead solution at 37°C for 1 h in 5% CO 2 . The cells attached to the beads were stained with 0.2% crystal violet aqueous solution in 20% methanol for 10 min. After removal of unattached cells, attached cells were analyzed under a microscope.
Inhibition of Heparin Binding of rec-␣4G by Synthetic Peptides-The effect of peptides on the binding of rec-␣4G to heparin was tested using heparin-Sepharose beads as described previously (30). The rec-␣4G protein (6 g, 0.68 M) was incubated with heparin-Sepharose beads (1 mg, Amersham Biosciences) in 50 mM Tris-HCl buffer (75 l, pH 7.4) in the presence of peptide (20 g, 178 M) at 4°C for 1 h, and then the beads were collected by centrifugation. The supernatant was removed, and the beads were washed twice with 10 mM Tris-HCl (pH 7.4), containing 100 mM NaCl. The rec-␣4G protein bound to the beads was extracted with SDS-PAGE sample buffer. The sample was analyzed by SDS-PAGE in 8% acrylamide gels under reducing conditions.
Solid Phase Heparin Binding Assay on Peptide-coated Plates-Hep-arin binding to peptide-coated plates was tested using biotinylated heparin (Celsus Laboratories Inc, Cincinnati, OH). Various amounts of peptides were coated onto 96-well ELISA plates (Iwaki, Tokyo, Japan) and dried overnight at room temperature. The wells were washed twice with 0.05% Tween 20 in phosphate-buffered saline (buffer B) and then blocked by addition of 150 l of 3% BSA in buffer B for 2 h. After removal of the supernatant and washing twice with buffer B, 10 ng of biotinylated heparin in buffer B was added to the wells. Following a 1-h incubation at 37°C, the supernatant was removed and then the wells were washed three times with buffer B. Streptavidin-conjugated horseradish peroxidase (10 ng, Sigma) in buffer B was added to the wells and then incubated at 37°C for 1 h to detect the biotinylated heparin-bound peptides. After removal of the supernatant and washing three times with buffer B, TMB solution (100 l, Sigma) was added and incubated for 10 min. After quenching with 1 M H 2 SO 4 (100 l), the absorbance at 450 nm was measured in a model 550 Microplate reader. Solid Phase Syndecan-4 Binding Assay to rec-␣4G-and Peptidecoated Plates-Syndecan-4 binding to rec-␣4G-and peptide-coated plates was examined using HT-1080 cell lysate. Cell lysate was prepared as described previously (18). Various amounts of rec-␣4G in buffer A (50 l) were coated on 96-well ELISA plates at 4°C overnight. Various amounts of peptides were also coated by drying overnight. The wells were blocked with 3% BSA at room temperature for 2 h. HT-1080 cell lysate (5 l) in 50 l of 3% BSA was added to the wells and incubated at 4°C overnight. After washing with buffer B, anti-syndecan-4 antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) in 3% BSA (1: 1000) was added and incubated at 37°C for 1 h. The wells were washed with buffer B, and then biotinylated anti-mouse IgG antibody (Vector Laboratories, Inc., Burlingame, CA) in 3% BSA (1: 2500) was added and incubated at 37°C for 1 h. After washing with buffer B, streptavidin-conjugated horseradish peroxidase in buffer B (1:2500) was added and incubated at 37°C for 1 h. After washing with buffer B, TMB solution (50 l) was added and incubated for 10 min. After addition of 1 M H 2 SO 4 (50 l), the absorbance at 450 nm was measured using a model 550 Microplate reader.

Cell Attachment and Heparin Binding Activities of a Recombinant Laminin ␣4
Chain G Domain Protein-A recombinant protein (rec-␣4G) containing the mouse laminin ␣4 chain G domain and the c-Myc sequence at the C terminus was expressed using Chinese hamster ovary cells and purified using a heparin column as described previously (30). First, we examined the ability of the rec-␣4G protein to promote HT-1080 human fibrosarcoma cell attachment. The rec-␣4G protein promoted cell attachment in a dose-dependent manner (Fig. 1A). Cell attachment on rec-␣4G was completely inhibited by heparin and partially inhibited by EDTA (Fig. 1B). These results suggested that cell attachment to rec-␣4G was mediated by heparin and was cation-dependent. Heparin inhibited the G domain-mediated cell attachment to rec-␣4G more effectively when compared with EDTA, suggesting that heparin-like molecules are mainly involved in the cell attachment to the laminin ␣4 chain G domain.
Cell Attachment Activity of Synthetic Peptides Derived from the Laminin ␣4 Chain G Domain-We prepared 116 overlapping synthetic peptides from the laminin ␣4 chain G domain to screen for biologically active sequences (Fig. 2). One hundred and four peptides were dissolved in aqueous solutions, but 12 peptides were insoluble (Fig. 2). First, we tested the soluble peptides for cell attachment activity on peptide-coated plates using HT-1080 cells. Peptide AG73 (RKRLQVQLSIRT), which has the strongest cell attachment activity of the ␣1 chain G domain peptides, and its scrambled peptide AG73T (LQQRRS-VLRTKI) were used as positive and negative controls, respectively (9). Six peptides (A4G6, A4G20, A4G24, A4G31, A4G46, and A4G107) showed strong dose-dependent cell attachment activity (Fig. 3A). These activities were comparable with that of AG73. Five peptides (A4G10, A4G47, A4G82, A4G83, and A4G90) showed moderate cell attachment activity but weaker than that of AG73 (Fig. 3B). Five peptides (A4G4, A4G25, A4G26, A4G78, and A4G102) showed weak cell attachment activity (Table I). None of the other peptides was active.
The 104 soluble peptides were coupled to CNBr-activated Sepharose beads and tested for HT-1080 cell attachment. AG73-conjugated Sepharose bead was used as a positive control (33). HT-1080 cells strongly attached and spread on the AG73-conjugated Sepharose bead, whereas ethanolamine-coupled control beads did not show cell attachment activity (Fig.  4). Five peptides (A4G47, A4G59, A4G69, A4G84, and A4G90) showed strong cell attachment and spreading activity comparable with that of AG73 ( Fig. 4 and Table I). Five additional peptides (A4G6, A4G20, A4G79, A4G82, and A4G83) showed moderate cell attachment activity in the bead assay (Table I). Four peptides (A4G24, A4G31, A4G46, and A4G107), which showed strong activity in the plate assay, did not support cell attachment to the beads. The remainder of the peptide beads did not have significant cell attachment activity. Based on the two separate cell attachment assays, we found 20 potentially active peptides (Table I).
Effects of Peptides on Cell Attachment to rec-␣4G-Next we examined the effect of the 20 cell adhesive peptides on HT-1080 cell attachment to rec-␣4G (Fig. 5). Cell attachment to rec-␣4G was strongly inhibited by A4G6, A4G24, A4G31, and A4G107. A4G20, A4G78, A4G82, A4G83, and A4G90 partially inhibited this attachment, and the remaining 7 peptides were not active. These results suggest that the nine active sequences are important for cell attachment of rec-␣4G.
Effects of EDTA and Heparin on Cell Attachment to Laminin Peptides-The effect of EDTA on HT-1080 cell attachment to peptide-coated plates was examined to determine the role of cations. Eleven peptides, which showed strong or moderate cell attachment activity in the plate assay, were examined (Fig. 6). AG73 was used as a control (9). Attachment to A4G24 was partially inhibited by 5 mM EDTA, but none of the other peptides showed reduced attachment activity in the presence of 5 mM EDTA.
Because cell attachment to rec-␣4G was completely inhibited by 10 g/ml heparin (Fig. 1B), we tested the effects of heparin on HT-1080 cell attachment to 11 peptide-coated plates (Fig. 6). Cell attachment to all of the peptides was inhibited in the presence of 10 g/ml heparin in this assay. These results suggest that cell attachment of these active peptides involves heparin-like binding.
Effects of Peptides on Heparin Binding to rec-␣4G-The rec-␣4G protein strongly bound to the heparin-Sepharose bead. We evaluated the effect of 104 soluble peptides on binding of rec-␣4G to heparin-Sepharose beads (Fig. 7A). A4G6, A4G10, A4G20, A4G82, and A4G83 significantly blocked the binding of the rec-␣4G protein to the heparin-Sepharose beads (Fig. 7A). The reminder of the peptides did not affect heparin binding to rec-␣4G. Next, we examined their inhibitory effects on heparin binding to rec-␣4G using various amounts of the five active peptides (Fig. 7B). These five active peptides showed a dosedependent inhibitory effect (Fig. 7B) with IC 50 values as follows: 10 (A4G6), 150 (A4G10), 5 (A4G20), 125 (A4G82), and 185 M (A4G83). These results suggested that these five sequences are critical for heparin binding in the ␣4 chain G domain.
Heparin Binding Activity of Peptides-Next we tested the heparin binding activity of the 20 peptides that promoted cell attachment and/or inhibited the heparin binding of rec-␣4G using biotinylated heparin. The peptides were coated on the ELISA plates, and biotinylated heparin was added to the peptide-coated plates, and then heparin bound to the peptide was detected by streptavidin-conjugated horseradish peroxidase. As a control, AG73 was used (9). AG73 showed strong heparin binding activity in a dose-dependent manner as described previously (14). Two peptides (A4G20 and A4G82) showed strong heparin binding activity comparable with that of AG73 (Fig. 8). Seven peptides (A4G6, A4G10, A4G24, A4G25, A4G31, A4G90, and A4G107) showed moderate heparin binding activity in this assay (Fig. 8). Seven peptides (A4G4, A4G26, A4G46, A4G47, A4G78, A4G83, and A4G102) weakly bound to heparin (Table   I). Four peptides (A4G59, A4G69, A4G79, and A4G84) did not bind to heparin under these conditions (Table I).
Active Core Sequences of A4G6, A4G20, and A4G82-A4G6, A4G20, and A4G82 promoted strong cell attachment and heparin binding directly, and inhibited cell attachment and heparin binding to rec-␣4G. A4G6S, A4G20S, and A4G82R, which are scrambled peptides of A4G6, A4G20, and A4G82, were also prepared and tested for their cell attachment activity (Table  II). These scrambled peptides did not show cell attachment activity to peptide-coated plates or to peptide-conjugated beads, did not block cell binding to the rec-␣4G protein, and did not bind heparin (data not shown). These results indicate that the activities of these active peptides were due to their sequence and not charge.
We next determined the active core sequences of the most active peptides, A4G6, A4G20, and A4G82, using systematically truncated N-and C-terminal peptides (Table II). A4G6c (KNDNLVYVY), an N-terminal truncated peptide of A4G6, still retained full activity, whereas A4G6d, with a deletion of N-terminal lysine from A4G6c, eliminated its cell binding activity. A4G6h (LAIKNDNLVY), a C-terminal truncated peptide, still retained activity, whereas A4G6i, with a deletion of Cterminal tyrosine from A4G6h, did not show the activity. These results indicate that the seven-amino acid sequence KNDNLVY is critical for A4G6 cell attachment activity in both assays.
Next, A4G20d (LYNFKHIY), an N-terminal truncated peptide of A4G20, still retained activity, whereas a deletion of N-terminal leucine from A4G20d (A4G20e) eliminated its cell binding activity (Table II). A4G20i (DVISLYNFK), a C-terminal truncated peptide, still retained activity, whereas a deletion of C-terminal lysine from A4G20i (A4G20j) eliminated its cell binding activity in both assays. These results indicate that the five-amino acid sequence LYNFK is critical for A4G20 cell attachment activity in both assays.
A4G82 contained a methionine residue at the C terminus. Because the methionine residue was easily oxidized during the synthesis, the methionine residue was replaced with norleucine for the truncation study (Table II). A4G82X (TLFLAH-GRLVFX, where X is Nle) promoted cell attachment as well as that of A4G82. A4G82Xb, with two amino acids deleted from the N-terminal of A4G82X, had similar cell attachment activity to A4G82X, but the phenylalanine residue deleted peptide A4G82Xc had no activity. C-terminal deletion peptide A4G82g (TLFLAHGRLVF) showed no cell attachment activity in both assays. These results indicate that at least the 10-amino acid sequence is important for A4G82 cell attachment activity.
A4G83, overlapped with the A4G82 sequence, and both peptides promoted cell attachment and inhibited the cell attachment and heparin binding to rec-␣4G (Table I). Therefore, we prepared a longer peptide, A4G823, which contained both A4G82 and A4G83. A4G823 showed strong cell attachment activity that was more potent than either A4G82 or A4G83 in both plate and bead assays (Table II). Additionally, A4G823 showed stronger heparin binding activity than that of A4G82 or A4G83. Furthermore, the inhibitory effect of A4G823 on cell attachment to rec-␣4G was much higher than that of either A4G82 or A4G83 (Fig. 5). These results suggest that the active core sequence of A4G823 is a potent active site.
Syndecan-4 Binding to rec-␣4G and Peptides-Next, the expression of syndecans in HT-1080 cells was examined by reverse transcriptase-PCR and Western blotting analysis as described previously (18). Syndecan-4 was detected by reverse transcriptase-PCR as well as Western blotting and found to be mainly expressed in HT-1080 cells (data not shown). Western blotting with anti-syndecan-4 antibody showed an ϳ40-kDa single molecule after heparitinase digestion of HT-1080 cells (data not shown). The binding of rec-␣4G and the most active peptides (A4G6 and A4G20) to syndecan-4 were determined in a solid phase assay using HT-1080 cell lysate. The rec-␣4G protein bound to syndecan-4 in a dose-dependent manner (Fig. 9A). The A4G6 and A4G20 peptides also showed syndecan-4 binding activity, whereas the scrambled peptides A4G6S and A4G20S did not show activity (Fig. 9A). Heparin (10 g/ml) significantly inhibited syndecan-4 binding to the rec-␣4G protein and peptides A4G6 and A4G20, indicating that the substrates bind to synde- RKRLQVQLSIRT ϩϩϩ ϩϩϩ a Sequences of the synthetic peptides are given in the single-letter code. All peptides have C-terminal amides. b Activity was scored on the following subjective scale: ϩϩϩ, adhesion comparable with that on AG73; ϩϩ, weak adhesion compared with that on AG73; ϩ, very weak adhesion compared with that on AG73; Ϫ, no adhesion. Active core sequences are written in boldface.
can-4 with a heparin-dependent manner (Fig. 9B). These results suggest that syndecan-4 plays a critical role for rec-␣4G mediated HT-1080 cell binding and that the A4G6 and A4G20 sites have the potential to be involved in this interaction. DISCUSSION We have found that the laminin ␣4 chain G domain recombinant protein promoted cell attachment and that was inhibited by heparin and partially inhibited by EDTA. These results suggest that heparin-like binding is involved in the cell attachment activity of the laminin ␣4 chain G domain. We identified cell binding and heparin-binding sites in the laminin ␣4 chain G domain by a systematic peptide screening. Twenty cell adhesive peptides were identified using peptide-coated plate and peptide-conjugated bead assays. Fourteen peptides promoted cell attachment activity in either the plate or bead assays, and six peptides were active in both assays. Four of the active peptides were not active when coated on the plates but were active in the bead assay, whereas 10 peptides were not active in the bead assay but were active in the plate assay. These results indicate that both assays should be employed when testing for active peptides. It is likely that the differential activities are due to conformational changes and/or poor coating efficiencies on the plates. For example, A4G59, A4G69, A4G79, and A4G84 showed strong cell attachment activity in the peptide-conjugated Sepharose bead assay, but these peptides were not active in the peptide-coated plate assay. We previously showed that a 12-mer peptide containing the RGD sequence (34) was active in the peptide-conjugated Sepharose bead assay but not active in the peptide-coated plate assay (33). The similarity of A4G59, A4G69, A4G79, and A4G84 and the RGD-containing peptide suggests that the cell binding of these peptides required an active conformation or higher coating efficiency to the plate. In contrast, A4G4, A4G10 A4G24, A4G25, A4G26, A4G31, A4G46, A4G78, A4G102, and A4G107 showed cell attachment activity in the peptide-coated plate assay, but these peptides were not active in the peptide-conjugated Sepharose bead assay. We showed previously (35) that a 12-mer peptide containing the active IKVAV sequence from the laminin ␣1 chain was active in FIG. 9. Binding of syndecan-4 to rec-␣4Gand peptide-coated plates using HT-1080 cell lysate. A, 96-well ELISA plates were coated with various amounts of rec-␣4G and peptides. HT-1080 cell lysate was added after blocking the substrate-coated plates with 3% BSA. Following incubation overnight at 4°C, syndecan-4 bound to the substrate-coated plates was detected by antisyndecan-4 antibody, biotinylated antimouse IgG antibody, and streptavidinconjugated horseradish peroxidase. Triplicate experiments gave similar results. B, effect of heparin on syndecan-4 binding to the rec-␣4Gand peptidecoated plates was examined. The rec-␣4G (0.75 g/well), A4G6 (1 g/well), and A4G20 (2 g/well) were coated, and then HT-1080 lysate and 10 g/ml heparin were added. Each value represents the mean of four separate determinations Ϯ S.D. *, p Ͻ 0.02. the peptide-coated plate assay but was not active in the peptide-conjugated Sepharose bead assay (11). These ␣4 chain G domain peptides may behave similarly to the IKVAV peptide and require a specific conformation for activity.
We sought to partially characterize the cellular receptors for the most active peptides. Previously, we identified several cell binding sequences on the ␣1 chain that recognize integrins (9,11,17). EDTA partially inhibited cell attachment to A4G24 but did not affect the rest of the active peptides, suggesting that it has the potential to interact with integrins. Cell attachment to all of the active ␣4 chain G domain peptides was inhibited by heparin. Cell surface heparin-like molecules are likely important for cell attachment to these peptides. Furthermore, the heparin binding of rec-␣4G is blocked by A4G6, A4G10, A4G20, A4G82, and A4G83 in a dose-dependent manner. In solid phase binding assays, biotinylated heparin also bound to these peptides. These data suggest that heparin-like cell surface molecules are important in ␣4 chain G domain-mediated cell attachment. Recently, we found that AG73, a laminin ␣1 chain LG4 module peptide, interacts with syndecan-1 (14,16), and A3G75, a laminin ␣3 chain LG4 module peptide, binds to syndecan-2 (18). Here we showed the rec-␣4G protein and the most active peptides A4G6 and A4G20 bound to syndecan-4. Taken together, the active peptides in the laminin ␣4 chain G domain have a potential to be involved in syndecan-mediated cell binding.
Active sequences are mainly located on the LG1 and LG4 modules (Fig. 10A). Previously, we identified the AG73 and A3G75 peptides in the laminin ␣1 and ␣3 chains LG4 module. AG73 binds to syndecan-1 and promotes various biological activities including cell adhesion, neurite outgrowth, acinar cell differentiation, and liver metastasis (9,11,16,19,20). A3G75 also promotes cell adhesion and binds to syndecan-2 (18). We have focused on homologous sites of the AG73 and A3G75 sequences in the ␣4 chain G domain (Fig. 10). Active sequences are mapped using a crystal structure-based sequential alignment of the LG modules reported previously (36) (Fig. 10). The AG73 sequence is located on the ␤-strand C region, and the A3G75 sequence is located on the loop region between the ␤-strands E and F in the LG4 module. A4G6 and A4G82-83, which showed the strongest cell attachment and heparin binding activities, are located in the homologous region of A3G75 in the ␣4 chain LG1 and LG4 modules. A4G6 and A4G82-83 would be extruded at the opposite edge of the 14-stranded ␤-sheet sandwich structure where a calcium ion binds. These results suggest that the active sites in the ␣3 and ␣4 chains are conserved at the loop region between the ␤-strands E and F and that the region plays a critical role in the biological functions of the laminin ␣4 chain G domain. In contrast, A4G78 was found to locate on the homologous site of AG73 in the ␣4 chain LG4 module. The cell attachment activity of A4G78 was much lower than that of AG73. The AG73 site was previously found to be chain-specific and to promote cell type-specific activity (14). A4G20 located on the loop region between the ␤-strand L to N in the ␣4 chain LG1 module is not found on any other previously identified active sequences. These chain-specific active sites in the G domain may promote cell type-specific biological activities.
Inhibition of cell attachment on the rec-␣4G protein substrate by A4G6, A4G20, A4G24, A4G31, A4G78, A4G82, A4G83, A4G90, and A4G107 suggests that these active sites are available on the intact molecule. These results also confirm the importance of conformation for cell interactions. However, not all active sites function in the intact molecule. Recently, a proteolytic fragment of laminin-5 generated by matrix metalloproteinase-2 was found to induce cell migration (37). Proteolytic fragments of other laminin chains may also contain activity that is cryptic in the intact molecule but is revealed after proteolysis. Fragments of plasminogen and collagen XVIII, designated angiostatin and endostatin, were found to have important functions in regulating angiogenesis and tumor growth (38,39). It is possible that the active peptide fragments of laminin play a critical role in its biological activity.