QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 17 August 2004
doi: 10.1242/jcs.01310

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sigle, R. O. Articles by Carter, W. G. Search for Related Content PubMed PubMed Citation Articles by Sigle, R. O. Articles by Carter, W. G.

Globular domains 4/5 of the laminin 3 chain mediate deposition of precursor laminin 5

Randy O. Sigle¹, Susana G. Gil¹, Mallar Bhattacharya², Maureen C. Ryan^1,*, Tai-Mei Yang¹, Tod A. Brown¹, Ariel Boutaud⁴, Yuko Miyashita², John Olerud² and William G. Carter^1,3,

¹ Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue, Seattle, WA 98109, USA
² Department of Medicince (Dermatology), University of Washington, 1959 N.E. Pacific St, Seattle, WA 98195, USA
³ Department of Pathobiology, University of Washington, 1959 N.E. Pacific St, Seattle, WA 98195, USA
⁴ BioStratum Incorporated, 4620 Creekstone Drive, Suite 200, Durham, NC 27703, USA

View larger version (46K): [in a new window]   Fig. 4. Thrombin cleavage of G4/5 inhibits deposition of pre-lam5 by 3–/– MKs. 3–/– MKs attached to glass coverslips were incubated (overnight) with either culture media alone (A; a-c), media with heparin-purified lam5 (A; d-f) or heparin-purified lam5 that has been digested with thrombin (A; g-i). The coverslips were processed for immunofluorescence to visualize ECM deposits containing the human 2 chain (a,d,g; B4-6), 3 chain (b,e,h; C2-5) or the G4/5 domains of 3200 (c,f,i; D2-1). Deposition occurred only where lam5 containing 3200 (C; lane 1) was added, and was prevented by thrombin digestion of the 3200 to 3165 (C, lane 2). Bar in (f), 20 µm. (B) Quantitation of deposit area/cell for six random fields for each of the conditions in A. (C) Western blot of lam5 preparations used in A and B. Lane 1, heparin-purified lam5; lane 2, heparin-purified lam5 from lane 1, digested with thrombin.

View larger version (75K): [in a new window]   Fig. 5. Deposition of lam5 into the PBM is inhibited by thrombin removal of G4/5. (A) HFKs were seeded onto confluent cultures of MF on coverslips, cultured for 18-20 hours, then fixed and processed for immunofluorescence with Mabs against the 3 chain (a-c; P5H10) or precursor 3200 (d-f; D2-1). Arrowheads indicate nuclei of HFKs. Thrombin digestion after the culture period removed G4/5 staining in 3200 (e), but 3 was still present in the PBM (b). Thrombin digestion throughout the culture period removed G4/5 from 3200 detected with anti-G4/5 (f) and suppressed deposition lam5 detected with anti-3 (c). Bar in (f), 20 µm. (B) HFKs grown on MFs were double stained with (a) anti-laminin 3 chain (P5H10, green) and (b) anti-integrin ß4 subunit (P4G11, red) and the images overlaid (c, yellow) Bar in (c), 20 µm. (C) ELISA quantitation of lam5 deposits. After the designated time periods (5, 10, 20 hours), the HFK/MF co-cultures were processed for ELISA using anti-2 Mab (B4-6). The amount of lam5 detected was significantly lower for all time points when thrombin was included during the co-culture (thrombin). Hirudin, a thrombin inhibitor, prevented the effects of thrombin.

View larger version (25K): [in a new window]   Fig. 6. Mutations in recombinant human laminin 3 chain. Wild-type and mutated human laminin 3 chains. Numbering of amino acid residues is from Ryan et al. (Ryan et al., 1994). Noncleavable 3 contains a 47 amino acid internal deletion from residue S1308 to residue C1354, in the spacer region between the G3 and G4 domains, producing a sequence in which C1307 is followed directly by S1355. This modification prevents cleavage of the G4/5 domains in culture. Precleaved 3 is a truncated 3 chain in which the G4 and G5 domains are not included, mimicking the cleavage that removes the G4/5 domains. The sequence stops at residue H1364, and five additional histidine residues have been added as an affinity tag at the C-terminus.

View larger version (81K): [in a new window]   Fig. 8. Expression of recombinant lam5 rescues adhesion and survival. 3+/+, 3–/–, 3–/– expressing noncleavable or precleaved human 3 chain were plated at 104 cells/well in 24-well Petri plates. Growth and deposition of the human 3 chain were monitored over a 7 day period. (A) Cell numbers were compared by counting 12 separate fields for a 10x objective. (B) ELISA for human 3 chain deposition onto the Petri dishes. (C) Phase-contrast images of cells after 7 days on Petri dishes. Bar, 25 µm. (D) Quantitation of cells in S-phase of the cell cycle after 48 hours on a Petri dish as determined by FACs analysis.

View larger version (65K): [in a new window]   Fig. 10. Deposits of recombinant lam5 direct assembly of integrin ß4 in detergent-resistant SACs. (A) 3+/+ MKs (a-b), or 3–/– MKs expressing noncleavable human laminin 3 chain (e-f and i-j), or precleaved 3 chain (g-h and k-l). Cells were cultured on BSA-coated glass coverslips for 24 hours. 3–/– MKs (c-d), which cannot attach to BSA-coated surfaces, were cultured on collagen-coated coverslips. Cultures were extracted with 0.25% Triton X-100 in PBS, fixed and examined by immunofluorescence microscopy. Double staining with anti-integrin ß4 (Mab P4G11; panels b, d, f, h, j and l) and Mabs against: mouse/human laminin 3 (D3-4; a and c), human laminin 3 chain (P5H10; e and g), or against G4/5 in human laminin 3-200 (D2-1;i and k). a-b, c-d, e-f, g-h, i-j, and k-l are each the same field. Bar, 10 µm. (B) Quantitation by ELISA of human laminin 3 chain deposition (Mab P5H10) and mouse laminin 3 chain (Mab p3C8), and Triton X-100 resistant mouse integrin ß4 in Triton extracted cultures of 3–/– MKs, 3+/+ MKs or 3–/– MKs expressing noncleavable (non-cl.) or precleaved (pre-cl.) human laminin 3 chain.

View larger version (135K): [in a new window]   Fig. 1. Transient expression of G4/5 in pre-lam5 in the PBM of epidermal wounds. Excisional wounds in mouse skin were healing for 7 days (a and c), or about 14 days (b and d). Cryostat sections of the wounds were probed with rabbit Mab R5808-f1 against mouse G5 in laminin 3 (a and b), or rat Mab against mouse lam5 (P3C8, c and d). Migrating leading cells in the partially healed wound expressed G4/5 (a) and lam5 (c), whereas closed wounds downregulated G4/5 (b) but maintained lam5 (d) (leading cells in outgrowth indicated with arrow; provisional BM indicated by arrowheads).

View larger version (39K): [in a new window]   Fig. 2. Assembly and secretion of lam5 and 6. 35S-labeled proteins were immunoprecipitated from the conditioned media of HFK or JEBG keratinocytes (JEBG) using the indicated Mabs (D2-1, C2-5, etc.). Before incubation with antibodies, one half of the collected media was digested with or without thrombin (+/– thrombin; 1 NIH unit/ml for 2 hours). The precipitates were separated by 6% SDS-PAGE. Lane 1, Mab D2-1 specific for the human 3 G4/5 domains; lane 2, Mab C2-5 specific for the human 3 chain; lane 3, Mab B4-6 specific for the human 2 chain; lane 4, Mab G2'-1 specific for the 170 kDa thrombospondin (TSP).

View larger version (58K): [in a new window]   Fig. 3. Binding and deposition of soluble pre-lam5, but not lam6, promotes spreading of 3–/– MKs. (A) Soluble lam5 and lam6 in conditioned media of HFKs and JEBG keratinocytes were trapped onto plastic surfaces using immobilized Mabs, and analyzed for adhesion and spreading of 3–/– MKs. Control IgG does not trap any adhesive activity. Anti-lam 3 Mab P5H10 traps both lam5 and lam6; anti-lam 2, Mab B4-6 traps only lam5. Immobilized lam5 and 6 promoted adhesion, establishing that both are inherently adhesive if deposited on the substratum. (B) 3–/– MKs were attached to glass coverslips as round cells. The glass surface was blocked with BSA, and the cells were cultured with soluble lam5 in conditioned media from HFKs (a,c) or soluble lam6 in JEBG keratinocytes-conditioned media (b,d). After 24 hours the cells were fixed, permeabilized and immunostained for the human laminin 3 chain (P5H10, c and d). Representative phase images are shown in (a) and (b). Bars, 25 µm.

View larger version (47K): [in a new window]   Fig. 7. Immunoblotting of recombinant chimeric human/mouse lam5 deposited by 3–/– MKs. ECM from confluent cultures of 3–/– MKs (lane 3), 3+/+ MKs (lane 4) and 3–/– MKs expressing noncleavable (lane 5) or precleaved (lane 6) human 3 chains. Control samples of HFK lam5 containing 3200 chain (lane 2) or 3165 chain (lane 1). Samples were fractionated by SDS-PAGE (A and B 5%, C 10%) and immunoblotted. (A) Mab P5H10 specific for the human 3 chain; (B) goat polyclonal antibody anti-ß3 chain; (C) Rabbit Mab R5808-F1 specific for G4/5 domain of mouse laminin 3200.

View larger version (50K): [in a new window]   Fig. 9. HFK adhesion to recombinant lam5 and effects of anti-integrin inhibitory antibodies. Trypsin suspended HFKs were plated into 24-well plates that are coated with type I collagen (COL.) or ECM from HFKs, 3+/+ MKs, 3–/– MKs, or 3–/– MKs expressing noncleavable and precleaved lam5. HFKs were allowed to adhere to the surfaces for 30 minutes at 37°C with or without the indicated Mabs: anti-LAM5 C2-9 against human laminin 3 chain; integrin ß1, P4C10; integrin 3, P1B5; integrin 2, P1H5; integrin 5, P1D6. Adhesion is reported as the proportion of fluorescently labeled cells remaining adhered after washing. Each condition was assayed in triplicate. This is a single adhesion assay, the results of which are representative of results obtained on at least three occasions.

View larger version (60K): [in a new window]   Fig. 11. Noncleavable and precleaved lam5 promote phosphorylation of focal adhesion kinase. As indicated, HFKs or 3–/– MKs were suspended with trypsin then either left in suspension or re-adhered to the indicated surfaces (collagen, HFK, 3+/+, etc.) for 30 minutes. Triton X-100 extracts were prepared containing 100 µM Na3VO4, 10 mM NaF, PMSF and NEM from the adherent cells and suspended cells. Extracts were assayed for phosphorylated focal adhesion kinase (P-FAK) by immunoblotting with antibody FAK P-Y397.

View larger version (34K): [in a new window]   Fig. 12. Non-cleavable lam5 is deposited while precleaved lam5 is secreted. 3–/– MKs expressing the noncleavable or precleaved human laminin 3 chain were cultured on collagen-coated plates in the presence of 35S-methionine for 3 days. In controls, HFKs were also labeled. Labeled proteins, either secreted into the conditioned culture media (CCM; B) or deposited into the ECM (A) were precipitated with antibody P5H10, against human laminin 3 chain (3), or antibody R5922 against mouse/human laminin ß1 and 1 chains (ß11). The labeled precipitates were fractionated by SDS-PAGE followed by fluorography. (C) Cultures of 3–/– MKs expressing the noncleavable or precleaved human laminin 3 chain were pulse labeled (2 hours) with 35S-methionine. Labeled cells were chased for 0, 3, 8 or 24 hours in media without label. Conditioned culture media (CCM) was immunoprecipitated with antibody P5H10, specific for the human 3 chain, and fractionated by SDS-PAGE (6%, reducing). (D) Cultures of 3–/– MKs or 3–/– MKs expressing noncleavable or precleaved human laminin 3 chain (2.5x106 cells well) were cultured (24 hours) in KGM (1.0 ml). The adherent cells were extracted with Triton X-100 detergent. Total human 3 chain in the ECM and culture medium was quantitated by ELISA with anti-human 3 chain Mab (P5H10). As a control, total laminin ß1 and 1 chains in culture medium were quantitated with anti-ß1/1 chain antibody (R5922). Each analysis was performed on three separate wells.