Soluble fibrin augments spreading of fibroblasts by providing RGD sequences of fibrinogen in soluble fibrin

Soluble fibrin augments spreading of fibroblasts by providing RGD sequences of fibrinogen in soluble fibrin

Thrombosis Research (2004) 114, 293 — 300 intl.elsevierhealth.com/journals/thre Regular Article Soluble fibrin augments spreading of fibroblasts by...

358KB Sizes 2 Downloads 182 Views

Thrombosis Research (2004) 114, 293 — 300

intl.elsevierhealth.com/journals/thre

Regular Article

Soluble fibrin augments spreading of fibroblasts by providing RGD sequences of fibrinogen in soluble fibrin Wei Yanga,*, Bin Wua, Shinji Asakurab, Isao Kohnoc, Michio Matsudab a

Division of Laboratory of Hematological Research, the 2nd Affiliated Hospital, China Medical University, No. 36, Sanhao Street, Heping District, Shenyang 110004, PR China b Division of Cell and Molecular Medicine, Center for Molecular Medicine, Jichi Medical School, Tochigi 329-0498, Japan c Central Research Laboratories, Iatron Laboratories Inc., Katori-Gun, Chiba 298-2247, Japan Received 23 February 2004; received in revised form 13 May 2004; accepted 15 June 2004 Available online 20 July 2004

KEYWORDS Fibroblasts; RGD sequences; Soluble fibrin

Abstract We previously reported that fibroblasts were found to spread far more avidly on NaBr-solubilized fibrin monomer (FM) monolayers than on immobilized fibrinogen (Fbg), indicating that removal of fibrinopeptides by thrombin is a prerequisite for the fibrin-mediated augmentation of cell spreading [J. Biol. Chem. 272 (1997) 88248829]. Soluble fibrin (SF), a 1:2 complex of fibrin-monomer and fibrinogen, is known to be present in the circulating blood under the pathological condition in which blood coagulation is activated. However, its physiological roles are still incompletely known. Fibroblasts spread on immobilized purified soluble fibrin. Cells spreading on immobilized soluble fibrin were blocked by the exogenous addition of soluble fibrin and glycinearginineglycineaspartic acidserinephenylalanine (GRGDSP)-synthetic peptide but not by the addition of fibrinogen or fibrin monomer. However, cell spreading activity was decreased in the surfaces coated with fragment X, whose Aa-chains lack carboxylterminal segments including arginineglycineaspartic acid (RGD)-2 domain, fibrin monomer complexes. It suggests that the RGD-2 domain of fibrinogen after being complexed with fibrin monomer plays a pivotal role for soluble fibrin-dependent cell spreading.

Abbreviations: TBS, Tris-buffered saline; BSA, bovine serum albumin; SF, soluble fibrin; FM, fibrin monomer; Fbg, fibrinogen; mAb, monoclonal antibody; RGD, arginine–glycine–aspartic acid; GRGDSP, glycine–arginine–glycine–aspartic acid–serine–phenylalanine; GRGESP, glycine–arginine–glycine–glutamic acid–serine–phenylalanine; GPRP, glycine–proline–arginine–proline; DMEM, Dulbecco’s modified eagle’s medium. * Corresponding author. Tel.: +86 24 83956356; fax: +86 24 83955092. E-mail address: [email protected] (W. Yang). 0049-3848/$ - see front matter D 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2004.06.022

294

W. Yang et al. Soluble fibrin in plasma derived from the patients of disseminated intravascular coagulation (DIC) was immuno-purified using the monoclonal antibody (mAb) which specifically recognizes the Ca++-dependent conformer of fibrinogen. The purified soluble fibrin consisted of desAA-fibrin monomer and two fibrinogen molecules and did show the cell spreading activity. Thus, soluble fibrin in plasma plays a role as the modulator of thrombogenic process in vivo. D 2004 Elsevier Ltd. All rights reserved.

Introduction

Materials and methods

Structural analysis data on the molecules and corresponding genes of human fibrinogen (Fbg) have shown that the Fbg molecule has two arginine–glycine–aspartic acid (Arg–Gly–Asp; RGD) segments in each Aa-chain and functions as an adhesion molecule [1]. Either one or both of these RGD segments may modulate cell spreading, and in fact, macrophages and macrophage cell-lines have been shown to adhere to fibrin but not to Fbg [2].The result seems to indicate that a specific configuration induced in the molecules upon conversion to fibrin is critical for supporting the cell adhesion process. Upon conversion of Fbg to fibrin, several peptide segments, which have important physiological functions, have been shown to be exposed on the surface of molecules. For example, the Bh15–42 residue segment, newly exposed at the amino terminus of the fibrin h-chain by thrombin-catalyzed cleavage of fibrinopeptide B has been shown to participate in the spreading of endothelial cells, and a 130–140 kDa polypeptide is thereon involved in the binding with the amino terminal segment of the fibrin h-chain [3]. Structurally, bsoluble fibrin (SF)Q consists of two molecules of Fbg and one molecule of desAABBfibrin monomer (desAABB-FM). Clinically, SF is known to be present in the circulating blood when blood coagulation is activated and thrombin is generated under pathologic conditions such as the disseminated intravascular coagulation (DIC) [4]. Interestingly, a specific monoclonal antibody (mAb), IF-43, failed to recognize acid-FM alone but was able to recognize a neoantigen exposed in the E domain of FM complexed with Fbg or its derivatives [5]. Soe et al. [5] established an aggregation assay of the SF by using latex beads coated with mAb IF-43 and found that SF in plasma was markedly elevated when ischemic states were present. In this communication, for the first time, we describe a new aspect of SF, which augments spreading of fibroblasts by functioning as an adhesion molecule.

Materials All chemical reagents were of the highest analytical grade commercially available and were purchased from the sources shown below. Bovine serum albumin (BSA), soybean trypsin inhibitor type-1 (ST1), trypsin, and sodium dodecyl sulfate (SDS) were from Sigma (St. Louis, MO, USA); microtiter 96-well flat-bottomed plates for cell culture were from Coster (Cambridge, MA, USA).

Synthetic peptides Glycine–arginine–glycine–aspartic acid–serine–phenylalanine (Gly–Arg–Gly–Asp–Ser–Pro; GRGDSP) and glycine–arginine–glycine–glutamic acid–serine–phenylalanine (Gly–Arg–Gly–Glu–Ser–Pro; GRGESP) peptides were purchased from Iwaki Glass (Osaka, Japan). Glycine–proline–arginine–proline (Gly–Pro– Arg–Pro; GPRP) peptide was from Sigma.

Cell Lines The following cells and culture media were used: human fibroblasts (TIG-3) were a generous gift from Dr. Tadashi Shimo-Oka of Iwaki Glass, Chiba, Japan. Rat smooth muscle cells were a gift from Dr. Uichi lkeda of the Department of Cardiology of Jichi Medical School. Dulbecco’s modified eagle’s medium (DMEM) with 10% fetal calf serum was obtained from GIBCO (Grand Island, NY, USA). The cells were cultured in 10 wt.% /vol.% fetal calf serum-containing DMEM at 37 8C in a 5.0% CO2 atmosphere.

Purification of fibrinogen and its derivatives Fbg was purified from citrated or acid citrate dextrose plasma using lysine-Sepharose 4B chromatography, gelatin-Sepharose 4B chromatography, and fractionation by ammonium sulfate as previously described [6]. Fibrinogen fragment X was prepared essentially as described by Nieuwenhui-

Soluble fibrin augments spreading of fibroblasts zen and Gravesen [7]. FM was prepared as described previously [5]. Namely, 2 ml of Fbg solution (1 mg/ml) was clotted for 2 h at 37 8C, then overnight at 4 8C with 1 NIH U of human thrombin (Sigma); the clot was squeezed with a plastic rod and centrifuged for 20 min at 15,000g at 22 8C. The pellet was solubilized with 20 mM acetic acid (acid-FM), brought to 5 mg/ml, and stored at 40 8C.

Preparation and characterization of an mAb that recognizes SF This was described previously [5]. This antibody designated as IF-43 reacts with FM, depleted of both of the two fibrinopeptide As, when the molecule is complexed with two molecules of Fbg or its D domain-containing derivatives via the set of dAT–daT polymerization sites.

Preparation of SF by gel filtration chromatography In order to obtain the SF, acid-FM was mixed with an excess of Fbg or its derivatives, i.e., 1 v of acidFM and 9 v of Fbg or fragment X in Tris-buffered saline (TBS) containing 1 mM CaCl2 at a molar ratio of 1:14.8 and incubated at 37 8C. After 30 min of incubation, the mixture was applied onto a Sephacryl S-300 column (Pharmacia Biotech, Tokyo, Japan; 1100 cm), and the eluted fractions were monitored by A280 for proteins and by an ELISA using IF-43 for the SF.

Preparation of the affinity gels and isolation of SF of plasma derived from patients Ten milligrams of mAb IF-1 [8], which specifically recognizes the Ca++-dependent conformer of Fbg, was conjugated to 2 ml of CNBr-activated Sepharose 4B gels essentially according to the manufacturer’s manual, using 0.1 M NaHCO3, pH 8.3, containing 0.5 M NaCl as the coupling buffer. After coupling, the gels were blocked with 1 M ethanolamine, pH 8.0, and washed successively with the coupling buffer and 0.1 M acetate buffer, pH 4.0, containing 0.5 M NaCl, and finally with the 50 mM Tris–HCl, pH 7.4, containing 0.3 M NaCl and 1 mM CaCl2. The SF was isolated from plasma derived from patients with DIC by gel filtration chromatography using Sephacryl S-300 [5] followed by affinity chromatography utilizing mAb IF-1 that recognizes Ca++-dependent conformation of the Fbg D domain essentially as described [8].

295

Reactivity of IF-43 to the gel-filtered fraction of acid-FM and Fbg mixture Reactivity of IF-43 to the gel-filtered fraction of the acid-FM and Fbg mixture was determined by a direct binding enzyme immunoassay (EIA). Ninetysix polystyrene microtiter wells were coated with various fractions of Sephacryl S-300 filtered fractions of the acid-FM and Fbg mixture. After 3 h of incubation, wells were blocked with 3% BSA in TBS (3% BSA–TBS) for 2 h at 22 8C. Then, wells were washed three times with TBS containing 0.05% Tween 20 (TBS–Tween) and incubated with 10 Ag/ ml IF-43 in TBS–Tween for 2 h at 22 8C. The bound antibody was captured by anti-mouse IgG conjugated with horse radish peroxidase (HRP, Dako, Japan), and plates were incubated at 37 8C with peroxidase substrate and analyzed by kinetic ELISA. Data were expressed by the means of quadruplicate determinations.

Immunoblotting SDS-polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblotting was carried out according to the method of Towbin et al. [9] as described previously. Briefly, after separation of the Fbg on 10% SDS-PAGE slab gels and electroblotting onto Immobilon polyvinylidene difluoride (PVDF) membranes (Milipore, Bedford, MA, USA), blots were soaked in 3% BSA–TBS for 1 h at 37 8C. PVDF membranes were rinsed with TBS–Tween and incubated with a rabbit anti-human Fbg IgG antibody (Dako) in TBS containing 0.1% BSA for 2 h at 22 8C. After washing three times with TBS–Tween, PVDF membranes were incubated for 1 h with peroxidase-conjugated goat anti-rabbit IgG (Dako). After rinsing, blots were immersed in the substrate solution, 4-chloro-1-naphthol/H2O2.

Cell adhesion assay The cell adhesion, i.e., the initial attachment of cells to the substratum, and subsequent spreading thereon were measured by the method of Grinnell et al. [10], with some modifications as described previously [11]. Briefly, 96-well polystyrene plates were coated with various concentrations of Fbg, fragment X, FM, FM–fragment X, or heat-denatured BSA substrata in TBS. After blocking with 3% BSA– TBS for 1 h at 37 8C, plates were washed with 0.2% BSA–TBS three times, 0.1 ml aliquots of suspensions of fibroblasts (5104 cells/ml) were added via a pipet into the coated wells for 60–90 min at 37 8C. Nonadherent cells were removed by washing with

296

W. Yang et al.

TBS, and the plates were examined by phase contrast microscope and photographed. Spread cells were counted, and if necessary, they were analyzed using a computer image processing package (IMAGE 1R, Universal Imaging, West Chester, PA, USA) to determine cell areas and diameters as described previously [11]. The spread cells were defined essentially as described elsewhere [11], namely, being polygonal in shape with a dark surface under phase contrast microscope and a large surface area than round cells+2S.D. The cells were photographed and then analyzed with a computer image processor.

Adhesion inhibition assay Cells were mixed with various concentrations of Fbg, FM, SF, or synthetic peptides in DMEM–0.2% BSA and were added to the wells coated with SF. We applied the same assay conditions as those for the adhesion assay. All experiments were performed at least three times with two independent isolates.

Results The SF was all eluted in the first small protein peak with a relative molecular mass of approximately

Figure 1 Gel filtration chromatography of a mixture of acid-FM and Fbg and fibroblasts spreading to the eluted fractions of gel filtration chromatography. Chromatography and SDS-PAGE followed by immunoblotting performed as described in Materials and methods. Eluted proteins were monitored by A280 (n–n) for proteins and by A415 (o–o) for SF reactive to IF-43 in ELISA. The wells were coated with the fractions, which had been separated by chromatography. Human fibroblasts were added to the wells and spread cells (D–D) were counted as described in Materials and methods. Inset: lane 1, Fbg as reference; lane2, fraction 84 (SF); lane3, acid-FM. Note that the proteins in fraction 84 were found to be a complex of desAABB-FM and Fbg.

Figure 2 Cell adhesion of human fibroblasts cells by fraction 84 (a) and fraction 114 (b). Bar: 20 Am.

9 10 5 (Fig. 1, peak 1). Immunoblot analysis showed that the SF consisted of desAABB-FM and Fbg at a molar ratio of 1:2 on the basis of band intensities for the Bh-chain and the h-chain representing Fbg and acid-FM, respectively (Fig. 1, inset, lane 2). Human fibroblasts were found to spread on SFcoated wells avidly, but not on Fbg-coated wells (Fig. 2). When the cells were replaced with smooth muscle cells, nearly identical augmentation of cell spreading was observed (data not shown). The adhesion of fibroblasts to the immobilized peak 1 fraction was inhibited concentration dependently by the SF but not by either Fbg or FM (Fig. 3). The result thus indicated that a specific conformation elicited in the SF may be responsible for supporting the cell adhesion. Although the precise mechanism has not been fully elucidated, removal of fibrinopeptides by thrombin untethers the closed aC–aC domains [12] and thereby exposes several fibrin-specific regions that may participate in the cell adhesion. The RGD segments residing at

Figure 3 Effect of solubilized Fbg derivatives on SFdependent cell adhesion.

Soluble fibrin augments spreading of fibroblasts

Figure 4 Effect of RGD peptides on SF-dependent adhesion of human fibroblasts.

Figure 5 Isolation of SF of plasma derived from a patient with DIC and fibroblasts spreading of the eluted fractions. (A) The gel filter fractions were monitored by A280 ( – ) for protein and by A415 (o–o) for SF reactivity to IF-43 in ELISA. (B) After affinity chromatography utilizing IF-1, the gel-filtered fractions were monitored by A280 ( – ), and spread cells (o–o) were counted as described. Inset: lane 1, Fbg as reference; lane 2, SF (fraction 7); lane 3, acid-FM.

297 Aa 95–97 and Aa 572–574 may also be categorized into this type of fibrin-specific segments. To see whether or not these RGD segments were exposed in the SF, i.e., a triad complex of one molecule of FM and two molecules of Fbg [5], we examined inhibition by a synthetic RGD peptide about cell adhesion onto the immobilized SF. Indeed, adhesion of fibroblasts was inhibited by the RGD peptide concentration dependently (Fig. 4), indicating that either one or both of the two RGD segments were exposed onto the surface of the complex and involved in the augmentation of the spreading of fibroblasts to the immobilized SF. The SF has been demonstrated in the circulating blood derived from endotoxin-treated rabbits as denoted cryoprofibrin and also from clinical patients with DIC [5]. Their observations prompted us to see whether the SF in plasma from patients with DIC was endowed with the capability to augment cell spreading. We therefore attempted to isolate the SF in plasma derived from patients with DIC (Fig. 5). The isolated SF was found to be a triad complex of one molecule of desAA-FM and two molecules of Fbg as estimated on the densitometric scanning of the h- and Bh-chains on immunoblotting as described in Materials and methods and also elsewhere [5]. The cell adhesion assay indicated that human fibroblasts spread on the immobilized SF fractions (data not shown). To further analyze the molecular mechanism of adhesive activity of SF, we compared with the cell adhesive activity between immobilized FM and SF. The cell adhesive activity to SF substratum is three times higher than that of FM substratum (Fig. 6); however, coating efficiency between FM and SF did not show any difference (data not shown), indicat-

..

..

Figure 6 The cell spreading activities of Fbg, fragment X, FM, FM–fragment X complex and SF substrata.

298 ing that Fbg molecules complexed with FM may prerequisite for the augmentation of cell adhesive activity. Thus, we analyzed the cell adhesive activities of fragment X, devoid of Aa 572–574 (RGD-2 domain) of Fbg. Fig. 6 indicates that immobilized Fbg molecules complexed with Fbg FM play a pivotal role for cell adhesive activity to SF substratum.

Discussion Adhesion and spreading of cells on immobilized thrombin-treated and nontreated fibrinogen have been studied extensively [10,13–17], but there still remain controversies on the mechanism of cell adhesion to the Fbg and fibrin substrata. Although experimental conditions are not necessarily comparable to each other, the controversies may largely arise from inconsistencies regarding the configuration of individual fibrinogen or fibrin molecule immobilized onto tissue culture wells. In two previous reports, we described that human fibroblasts and human glioma cells were able to spread avidly on immobilized FM but not on Fbg [6,11]. In this report, we found the new function of SF, i.e., SF augments cell spreading, and analyzed the molecular mechanism of SF-dependent augmentation of cell spreading.

Analysis of molecular mechanism of SFdependent augmentation of cell spreading The Fbg molecules have two RGD segments at Aa 95–97 (RGD-1) and Aa 572–574 (RGD-2) in each Aachain. We reported that NaBr-solubilized fibrin did show strongly adhesive activity, indicating that the removal of fibrinopeptides by thrombin is a prerequisite for the fibrin-mediated augmentation of cell spreading, NaBr-solubilized fibrin required both RGD-1 and RGD-2 domains for cell adhesion [11]. Structurally, the RGD-2 segment residues are very close to the cluster of negatively charged residues (Aa 586–595), which may interact with a cluster of positively charged residues (Aa 601–608) in the carboxy-terminal segment of the Aa-chain, and form a small globular domain (aC domain) [12]. In the native Fbg molecule , the two aC domains are associated with each other, forming an additional (aC–aC) globule, and then to the central E domain. Upon thrombin-catalyzed cleavage of fibrinopeptides A and B from the central part of the E domain, the additional (aC–aC) globule is released from the E domain and is disconnected into individual aC domains [12]. Because the RGD-2

W. Yang et al. segment resides only three residues apart from the amino-terminus of the cluster of negatively charged residues, the RGD-2 segments may be inaccessible or hidden in the native Fbg molecule. When the (aC–aC) domain is fully dissociated upon conversion of Fbg to fibrin, RGD-2 segments may become available for supporting the cell adhesion. SF also augments cell spreading. However, the molecular mechanism of cell adhesive activity of SF did show quite a difference from that of FM. Cells require the exposure of RGD-1 domain as well as that of the RGD-2 domain after thrombin-catalyzed cleavage of fibrinopeptides and the release of (aC– aC) globule from the E domain on the immobilized NaBr-FM molecules. Data did show decreased cell spreading activity on the surface coated with FM– fragment X Fbg complex (Fig. 6). The results together suggest that RGD-2 segments of Fbg complexed with FM play a pivotal role for SFdependent cell spreading. Previously, we reported that immobilized native Fbg did not show any adhesive activity due to unexposure of RGD domain on the surface of the molecules [11]. Our data suggested that the RGD-2 domain of Fbg molecules connected with FM exposed to the surface of SF even without cleavage of fibrinopeptides of Fbg molecules. Native Fbg is a conformationally labile molecule. For example, immobilized Fbg onto surface induces similar conformational changes without thrombincatalyzed cleavage of fibrinopeptides A and B that are also evoked when Fbg is converted into fibrin [18]. Similar conformational liability of Fbg was also found in platelet–Fbg interaction. That is, the conformational transitions ligand induced binding site (LIBS) epitopes of GPIIb/h3 of platelets do occur after Fbg binds to GPIIb/IIIa; also, Fbg undergoes conformational changes upon binding to the receptor without cleavage of fibrinopeptides A and B [19]. Recently, we also found that when FM polymerized each other or with Fbg, the structure of the FM has a change, in which Aa 20–29 as a neoantigen was exposed on the surface of E domain [20]. Thus, we favor the following hypothesis that the conformational changes induced by the removal of the fibrinopeptides from the amino-terminus of the Aa- and Bh-chains in the E domain of Fbg must be transmitted from the E domain to the D domains of each of the two Fbg molecules complexed with FM, and from the D domain of Fbg to the aC domain of the same Fbg molecule. Even without cleavage of fibrinopeptides A and B, the aC domain is released from the E domain of the Fbg molecule. Then, the RGD-2 domain of Fbg in SF may also become available for supporting the cell adhesion.

Soluble fibrin augments spreading of fibroblasts

Purification and characterization of SF in plasma derived from hypercoagulable states Determination of SF in plasma derived from patients with thrombogenic diseases has been expected to provide useful information on the state and degree of intravascular coagulation as well as determination of FDP also provides this information. The concentration of SF in plasma was not significantly correlated with that of the ddimer, suggesting that these two parameters reflected different aspects and stages of intravascular coagulation. Interestingly, Soe et al. [5] showed that on the basis of clinical manifestations of ischemic diseases in one or more organs and overt bleeding, a high level of SF accompanied by a low level of d-dimers seems to be predominantly associated with thrombosis and a low level of SF accompanied by a high level of d-dimers seems to be predominantly associated with severe DIC-manifesting bleeding. SF derived from the patients with DIC or ischemic states such as cerebral infarction and myocardial infarction contains mainly the desAAFM, as opposed to the desAABB-FM. These data are well correlated with the increased fibrinopeptide A in plasma in coronary thrombosis–ischemic states (variant angina and myocardial infarction) [21]. Our data show that SF, mainly the desAABB-FM– Fbg complex, also has adhesive activity in vivo (Fig. 5), indicating that SF, as well as cytokines such as platelet derived growth factor (PDGF) [22–24] and transforming growth factor-beta (TGF-h) [25–29], might be a key modulator in the proliferation of fibroblasts and smooth muscle cells. Thus, SF might be deeply involved in the ischemic states such as myocardial ischemia, restenosis of coronary artery after percutaneous transluminal coronary angioplasty (PTCA) intervention, or gliosis after cerebral ischemia in vivo.

Acknowledgements This study was supported by the following Japanese sources: The Cell Science Research Foundation, The Sagawa Foundation for Promotion of Cancer Research, Yamanouchi Foundation for Research on Metabolic Disorders, Mochida Memorial Research Foundation, the Ryoichi Naito Foundation for Medical Research Foundation, Nakatomi Health Science Research Foundation, The Foundation for the Development of the Community and Scientific Research, and Scientific Research on Priority Areas (cancer) Grants-in-aid 08671292, 09254257, and

299 09671174 from the Ministry of Education, Science and Culture of the Government of Japan.

References [1] Felding-Habermann B, Ruggeri ZM, Cheresh DA. Distinct biological consequences of integrin alpha v beta 3-mediated melanoma cell adhesion to fibrinogen and its plasmic fragments. J Biol Chem 1992;267:5070 7. [2] Shainoff JR, Stearns DJ, DiBello PM, Hishikawa-Itoh Y. Characterization of a mode of specific binding of fibrin monomer through its amino-terminal domain by macrophages and macrophage cell-lines. Thromb Haemost 1990; 63:193 203. [3] Simon DI, Ezratty AM, Francis SA, Rennke H, Loscalzo J. Fibrin(ogen) is internalized and degraded by activated human monocytoid cells via Mac-1 (CD11b/CD18): a nonplasmin fibrinolytic pathway. Blood 1993;82:2414 22. [4] Wiman B, Ranby M. Determination of soluble fibrin in plasma by a rapid and quantitative spectrophotometric assay. Thromb Haemost 1986;55:189 93. [5] Soe G, Kohno I, Inuzuka K, Itoh Y, Matsuda M. A monoclonal antibody that recognizes a neo-antigen exposed in the E domain of fibrin monomer complexed with fibrinogen or its derivatives: its application to the measurement of soluble fibrin in plasma. Blood 1996;88:2109 17. [6] Yang W, Asakura S, Sakai T, Nakamura M, Fujimura K, Matsuda M. Two-step spreading mode of human glioma cells on fibrin monomer: interaction of alpha(v)beta3 with the substratum followed by interaction of alpha5beta1 with endogenous cellular fibronectin secreted in the extracellular matrix. Thromb Res 1999;93:279 90. [7] Nieuwenhuizen W, Gravesen M. Anticoagulant and calciumbinding properties of high molecular weight derivatives of human fibrinogen, produced by plasmin (fragments X). Biochim Biophys Acta 1981;668:81 8. [8] Takebe M, Soe G, Kohno I, Sugo T, Matsuda M. Calcium ion dependent monoclonal antibody against human fibrinogen; preparation, characterization, and application to fibrinogen purification. Thromb Haemost 1995;73:662 7. [9] Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A 1979;76:4350 4. [10] Grinnel F, Feld M, Minter D. Fibroblast adhesion to fibrinogen and fibrin substrate: requirement for cold-insoluble globulin (plasma fibronectin). Cell 1980;19:517 25. [11] Asakura S, Niwa K, Tomozawa T, Jin YM, Madoiwa S, Sakata Y, et al. Fibroblasts spread on immobilized fibrin monomer by mobilizing a h1-class integrin together with a vitronectin receptor avh3 on their surface. J Biol Chem 1997; 272:8824 9. [12] Vejlich YI, Gorkun OV, Nieuwenhuizen W, Weisel JW. Carboxyl-terminal portions of the chains of fibrinogen and fibrin. Localization by electron microscopy and the effects of isolated aC fragments on polymerization. J Biol Chem 1993;268:13577 85. [13] Hantgan RR, Endenburg SC, Sixma JJ, de Groot PG. Evidence that fibrin alpha-chain RGDX sequences are not required for platelet adhesion in flowing whole blood. Blood 1995;86:1001 9. [14] Farrell DH, Thiagarajan P, Chung DW, Davie EW. Role of fibrinogen alpha and gamma chain sites in platelet aggregation. Proc Natl Acad Sci U S A 1992;89:10729 32.

300 [15] Weisel JW, Nagaswami C, Vilaire G, Bennett JS. Examination of the platelet membrane glycoprotein IIb–IIIa complex and its interaction with fibrinogen and other ligands by electron microscopy. J Biol Chem 1992;267: 16637 43. [16] Dejana E, Colella S, Languino LR, Balconi G, Corbascio GC, Marchisio PC. Fibrinogen induces adhesion, spreading, and microfilament organization of human endothelial cells in vitro. J Cell Biol 1987;104:1403 11. [17] Harley SL, Sturge J, Powell JT. Regulation by fibrinogen and its products of intercellular adhesion molecule-1 expression in human saphenous vein endothelial cells. Arterioscler Thromb Vasc Biol 2000;20:652 8. [18] Zamarron C, Ginsberg M, Plow EF. Monoclonal antibodies specific for a conformationally altered state of fibrinogen. Thromb Haemost 1990;64:41 6. [19] Du X, Gu M, Weisel JW, Nagaswami C, Bennett JS, Bowditch R, et al. Long range propagation of conformational changes in integrin alpha IIb beta 3. J Biol Chem 1993; 268:23087 92. [20] Yang W, Wu B, Sun D. Research on the change of structure on fibrin E domain modulated by E–D binding for fibrin assembly. Chin J Biochem Mol Biol 2004;20:213 8 [In Chinese]. [21] Linder R, Frebelius S, Grip L, Swedenborg J. The influence of direct and antithrombin-dependent thrombin inhibitors on the procoagulant and anticoagulant effects of thrombin. Thromb Res 2003;110:221 6.

W. Yang et al. [22] Koslowski R, Seidel D, Kuhlisch E, Knoch KP. Evidence for the involvement of TGF-beta and PDGF in the regulation of prolyl-a hydroxylase and lysyloxidase in cultured rat lung fibroblasts. Exp Toxicol Pathol 2003;55:257 64. [23] Wells RG, Kruglov E, Dranoff JA. Autocrine release of TGFbeta by portal fibroblasts regulates cell growth. FEBS Lett 2004;559:107 10. [24] Nishishita T, Lin PC. Angiopoietin-1, PDGF-B, and TGF-beta gene regulation in endothelial cell and smooth muscle cell interaction. J Cell Biochem 2004;91:584 93. [25] Moses HL, Yang EY, Pietenpol JA. Regulation of epithelial proliferation by TGF-beta. Ciba Found Symp 1991;157: 66 74. [26] Lopez-Casillas F, Wrana JL, Massague J. Beta glycan presents ligand to the TGF beta signaling receptor. Cell 1993;73:1435 44. [27] Leveen P, Pekny M, Gebre-Medhin S, Swolin B, Laesson E, Betsholtz C. Mice deficient for PDGF B show renal, cardiovascular, and hematological abnormalities. Genes Dev 1994;8:1875 87. [28] Ivanov VO, Rabovsky AB, Ivanova SV, Niedzwiecki A. Transforming growth factor-beta 1 and ascorbate regulate proliferation of cultured smooth muscle cells by independent mechanisms. Atherosclerosis 1998;140:25 34. [29] Bhowmick NA, Chytil A, Plieth D, Gorska AE, Dumont N, Shappell S, et al. TGF-beta signaling in fibroblasts: the oncogenic potential of adjacent epithelia. Science 2004; 303:848 51.