Báo cáo khoa học: Functional role of Bb-chain N-terminal fragment in the fibrin polymerization process pdf

Functional role of Bb-chain N-terminal fragment in the fibrin polymerization process E. V. Lugovskoy, P. G. Gritsenko, L. G. Kapustianenko, I. N. Kolesnikova, V. I. Chernishov and S. V. Komisarenko Palladin Institute of Biochemistry, National Academy of Sciences of Ukraine, Kyiv 2 , Ukraine Fibrinogen is a central protein of the blood coagula- tion system. The human fibrinogen molecule (molecu- lar mass  344 kDa) consists of two identical subunits connected by disulfide bonds [1]. Each monomer subunit is formed by three nonidentical polypeptide chains, Aa,Bb and c. The N-terminal ends of all six polypeptide chains are situated in the fibrinogen cen- tral region, which is known as the E-domain. Two Keywords fibrin; monoclonal antibodies; peptides; polymerization sites Correspondence E. Lugovskoy, Palladin Institute of Biochemistry, National Academy of Sciences of Ukraine, 9 Leontovicha Street, 01601, Kyiv, Ukraine Fax: +38 044 2796365 Tel: +38 044 2343354 E-mail: lougovskoy@yahoo.com (Received 15 May 2007, revised 3 July 2007, accepted 10 July 2007) doi:10.1111/j.1742-4658.2007.05983.x Four mAbs of the IgG 1 class to the thrombin-treated N-terminal disulfide knot of fibrin, secreted by various hybridomas, have been selected. Epi- topes for two mAbs, I-3C and III-10d, were situated in human fibrin frag- ment Bb15–26, and those for two other mAbs, I-5G and I-3B, were in fragment Bb26–36. Three of these mAbs, I-5G, I-3B and III-10D, as well as their Fab-fragments, decreased the maximum rate of fibrin desAA and desAABB polymerization up to 90–95% at a molar ratio of mAb (or Fab- fragment) to fibrin of 1 or 2. The fourth mAb, I-3C, did not influence the fibrin desAABB polymerization and inhibited by 50% the maximum rate of fibrin desAA polymerization. These results suggest that these mAb inhibitors block a longitudinal fibrin polymerization site. As the mAbs retard both fibrin desAABB and fibrin desAA polymerization, one can con- clude that the polymerization site does not coincide with polymerization site ‘B’ (B b15–17). To verify this suggestion, the polymerization inhibitory activity of synthetic peptides BbSARGHRPLDKKREEA(12–26), BbLDKKREEA(19–26), BbAPSLRPAPPPI(26–36), BbAPSLRPAPPPIS GGGYRARPA(26–46) and BbGYRARPA(40–46), which imitate the vari- ous sequences in the N-terminal region of the fibrin Bb-chain, have been investigated. Peptides Bb12–26 and Bb26–46, but not Bb40–46, Bb19–26, and Bb26–36, proved to be specific inhibitors of fibrin polymerization. The IC 50 values for Bb12–26 and Bb26–46 were 2.03 · 10 )4 and 2.19 · 10 )4 m, respectively. Turbidity and electron microscopy data showed that peptides Bb12–26 and Bb26–46 inhibited the fibrin protofibril formation stage of fibrin polymerization. The conclusion was drawn that fibrin fragment Bb12–46 took part in fibrin protofibril formation simultaneously with site ‘A’ (Aa17–19) prior to removal of fibrinopeptide B. A model of the inter- molecular connection between fragment Bb12–46 of one fibrin desAA molecule and the D-domain of another has been constructed. Abbreviations 1 Bb12–26, BbSARGHRPLDKKREEA(12–26); Bb19–26, BbLDKKREEA(19–26); Bb26–36, BbAPSLRPAPPPI(26–36); Bb26–46, BbAPSLRPAPPPISGGGYRARPA(26–46); Bb40–46, BbGYRARPA(40–46); NaCl ⁄ P i , 0.02 M potassium phosphate buffer (pH 7.4) with 0.13 M NaCl; t-NDSK, thrombin-treated N-terminal disulfide knot of fibrin; TPBS, NaCl ⁄ P i with 0.05% Tween-20. 4540 FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS peripheral regions of the fibrinogen molecule were his- torically named D-domains. However, it was discov- ered by X-ray analysis that there were two distinct bC- and cC-domains in the peripheral D-region [2]. As a result of the activation of the blood coagulation sys- tem, thrombin is formed; this attacks fibrinogen and splits off two fibrinopeptides A (Aa1–16). Fibrinogen is transformed into fibrin desAA, which is able to polymerize spontaneously, forming two-stranded pro- tofibrils with half-staggered fibrin molecules [3,4]. Pro- tofibrils associate laterally, producing fibrils. The fibrils also associate laterally, branching and forming a three- dimensional network, which is the framework of the whole blood thrombus [5]. It is now widely accepted that the initial step of fibrin polymerization (protofibril formation) is carried out by the intermolecular pairing of ‘A’ and ‘a’ polymerization sites of fibrin monomers. Site ‘A’ (Aa17–19) is exposed in the N-terminal region of the Aa-chain by the splitting off of fibrinopeptide A [6]. Site ‘a’ is formed by amino acid residues cGln329, cAsp330, cHis340 and cAsp364, situated in the cC-domain of the fibrinogen ⁄ fibrin molecule [7]. On protofibril formation, polymerization site ‘B’ (Bb15– 18) is exposed at the N-terminal region of the fibrin b-chain, when fibrinopeptide B (Bb1–14) is split off by thrombin and desAABB is formed [6,8]. Complemen- tary to site ‘B’, a site ‘b’ is formed by amino acid residues BbGlu397, BbAsp398, BbArg406, BbCys407, BbHis408 and BbAsp432, localized in the fibrin bC-domain [7]. The ‘B’–‘b’ pairing leads to structural rearrangements in the D-domain and the strengthening of protofibril lateral contacts [9,10]. Many indirect data show that other polymerization site (or sites) may be situated in fibrin desAA fragment Bb14–54. For example, fibrinogen-325, lacking frag- ment Bb1–42, is clotted by thrombin 180 times slower as compared to intact fibrinogen, and it has been shown that such a fibrin keeps its monomeric form for a long time [11,12]. Pandya et al. [13] found an inhibi- tory effect of peptide Bb15–42 on fibrin polymerization and showed that this effect was not related to polymer- ization site ‘B’ (Bb15–18). Moen et al. [14] showed significantly impaired polymerization of fibrin desAA obtained from recombinant fibrinogen with histidine substituted for arginine at Bb14 [14]. We found earlier that mAb 2d)2a and its Fab-fragment, whose target epitope involves BbArg14, specifically inhibited fibrin desAA polymerization [15]. It was also shown that synthetic peptide Bb40–54 dissociated the (DD)E com- plex at a 5000 molar ratio of peptide to complex [16]. The Bb28–30 and Bb36–44 regions are framed by pro- line brackets, which usually indicates that such pep- tides are involved in protein–protein interactions [17]. Three mAbs to t-NDSK that inhibit fibrin polymeriza- tion are described in this article. The inhibitory effects of synthetic peptides Bb12–26 and Bb26–46 on fibrin polymerization were also studied. The results suggest that an unknown site (not ‘B’), important for fibrin polymerization, is situated at fibrin fragment Bb12–46. This site seems to take part in protofibril formation, and is operational without fibrinopeptide B being removed by thrombin. Results We have obtained 35 hybridomas producing different antibodies of IgG 1 class to human fibrin fragment thrombin-treated N-terminal disulfide knot of fibrin (t-NDSK) with the molecular structure (Aa17–51, Bb15–118, c1–78) 2 . Three of these hybridomas secret- ing mAbs I-5G, I-3B and III-10D were chosen on the basis of the strong and specific inhibition by these mAbs of fibrin desAABB polymerization. Turbidity analysis showed that these mAbs and their Fab-frag- ments specifically inhibited the maximum rate of fibrin desAABB polymerization by up to 90–95% at a molar ratio to fibrin of 1 : 1 or 2 : 1 (Fig. 1). Similar results were obtained with fibrin desAA (where fibrinopep- tides B are conserved) and with fibrin produced in the fibrinogen + thrombin reaction (data not shown). Such a strong and specific inhibition of fibrin polymer- ization by mAbs and their Fab-fragments indicated the specific blockade of a polymerization site situated in this region of the molecule rather than steric hindrance by antibodies. Furthermore, this inhibition could not be explained by the blockade of polymerization site ‘B’ (Bb15–17), as there is no ‘B’ site exposed in fibrin desAA. The fourth mAb, I-3C, did not influence fibrin desAABB polymerization but, surprisingly, it inhibited by 50% the maximum rate of fibrin desAA polymeri- zation. Figure 2 shows that mAb I-5G reacted in ELISA with fibrinogen, fibrin desAABB, t-NDSK, fragment Bb15–118, slightly reacted with c1–78, and did not react with E 3 -fragment (E 3 -fragment includes polypeptide constituent b54–120). mAbs I-3B, III-10D and I-3C reacted with the antigens mentioned above in the same manner. Immunoblot analysis (Fig. 3) showed that mAbs I-5G, I-3B, III-10D and I-3C (not shown) reacted only with the Bb15–118 constituent of t-NDSK. The slight reaction of mAbs with c1–78 in ELISA may be explained by the contamination of the c1–78 preparation with Bb15–118. ELISA and immu- noblot results indicate that epitopes for all four mAbs are situated in the Bb15–53 fibrin fragment. K D values of the binding of mAbs I-5G, I-3B, III-10D and I-3C to t-NDSK were 1.0 · 10 )8 m, 7.4 · 10 )9 m, E. V. Lugovskoy et al. Functional role of Bb-chain N-terminal fragment FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS 4541 2.3 · 10 )8 m, and 3.7 · 10 )9 m, respectively. The com- petitive ELISA showed that mAb I-5G competed with I-3B and mAb III-10D competed with I-3C in fibrinogen binding. mAbs I-5G and I-3B did not com- pete with mAbs III-10D and I-3C. To localize epitopes for these mAbs, the following peptides comprising human fibrinogen sequences in Bb12–46 were synthesized: B bSARGHRPLDKKRE EA(12–26), BbLDKKREEA(19–26), BbAPSLRPAPP PI(26–36), BbAPSLRPAPPPISGGGYRARPA(26–46), and BbGYRARPA(40–46). Competitive ELISA was used to study the binding of mAbs to fibrinogen in the presence of these peptides. It was found that mAbs III-10D and I-3C reacted with peptide Bb12–26 (Fig. 4A), and mAbs I-5G and I-3B with peptides Bb26–36 (Fig. 4B) and Bb26–46 (not shown). There was no reaction between any of these mAbs with pep- tides Bb19–26 and Bb40–46. We also found that mAb I-3C but not other mAbs (III-10D, I-5G and I-3B) competed with mAb 2d)2a obtained in our laboratory earlier with an epitope encompassing amino acid resi- dues BbArg14 and BbGly15 [15]. These results show that epitopes for mAbs I-3C and III-10D are situated in fragment Bb15–26, whereas epitopes for mAbs I-5G and I-3B are situated in Bb26–36 (Fig. 5). We also found that mAb I-3C but not mAb III-10D competed with mAb 2d-2a obtained in our laboratory earlier with an epitope encompassing amino acid resi- dues BbArg14 and BbGly15 [15]. These data show that the epitope for mAb I-3C is situated closer to the N-terminus of fragment Bb15–26 as compared to the epitope for mAb III-10D. The fact that epitopes for these two mAbs do not coincide in the frame of fragment Bb15–26 may explain the difference in their inhibitory effects in relation to fibrin desAABB poly- merization. We suggested that epitopes for these mAbs coin- cided or overlapped with fibrin polymerization site⁄ sites situated in this region of the molecule. To verify this suggestion, we investigated the inhibitory action of Fig. 1. The dependence of the maximum rate of fibrin desAABB polymerization (V max ) on the molar ratios of mAbs I-5G (A), 1-3B (B), and III-10D (C), and their Fab-fragments, to fibrin. Fig. 2. Binding curves of mAb I-5G to the antigens fibrin fragment Bb15–118, t-NDSK, fibrinogen (F), fibrin desAABB, fibrin fragment c1–78 and fibrin fragment E 3 adsorbed to microtiter plates (ELISA). Fig. 3. Immunoblot analysis of reduced t-NDSK using mAbs I-5G, I-3B, and III-10D. Lane 1: molecular mass markers. Lanes 1–3: SDS ⁄ PAGE stained with Coomassie R250. Lanes 4 and 5: immuno- staining with mAb III-10D. Lanes 6 and 7: immunostaining with mAb I-3B. Lanes 8 and 9: immunostaining with mAb I-5G. Lanes 2, 4, 6, and 8: fibrin fragment Bb15–118. Lanes 3, 5, 7, and 9: t-NDSK + b-mercaptoethanol. Functional role of Bb-chain N-terminal fragment E. V. Lugovskoy et al. 4542 FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS synthesized peptides on fibrin polymerization. Turbi- dity analysis showed that peptides Bb12–26 and Bb26– 46 increased lag-time and decreased the maximum rate of polymerization of both fibrin desAABB and fibrin produced in the fibrinogen + thrombin reaction (Fig. 6A,B). The values of IC 50 (the concentrations of the peptides at which 50% of inhibition of the fibrin polymerization maximum rate were observed) for Bb12–26 and Bb26–46 were 2.0 · 10 )4 m and 2.19 · 10 )4 m, respectively. The final turbidity of fibrin clots proved to be decreased only slightly (Fig. 6C). The increase in lag-time may be related to either increasing protofibril critical length or a decrease in the rate of protofibril formation. The combination of three parameters obtained with turbidity analy- sis ) longer lag-time, lower maximum rate of fibrin polymerization, and almost the same value of final tur- bidity ) suggested more the retardation of the first stage of fibrin polymerization, i.e. protofibril forma- tion, than inhibition of protofibril lateral association. The final conclusion on this depended on electron microscopy findings (see below). To compare the inhibitory effect of peptides Bb12– 26 and Bb26–46 with the effect of the known inhibitor of fibrin polymerization, the peptide GPRP, mimicking site ‘A’ (Aa17–19), we performed some experiments with GPRP and found the IC 50 value for the latter to be equal to 2.02 · 10 )5 m. Peptides Bb19–26, Bb26–36 and Bb40–46, at a molar ratio of peptide to fibrin of 1500, did not inhibit polymerization of either fibrin desAABB or fibrin produced in the fibrinogen + thrombin reaction. We found also that the mixture of two peptides Bb26–36 and Bb40–46 at the same molar Fig. 4. Binding curves of mAbs I-5G, I-3B, III-10D and I-3C to the synthetic peptides Bb12–26 (A) and Bb26–36 (B) in competitive ELISA. The peptide concentrations varied from 0.39 lgÆmL )1 to 50 lgÆmL )1 , and the mAb concentration was constant at 1 lgÆmL )1 . Fibrinogen was adsorbed to microtiter plates. Fig. 5. The amino acid sequence of human fibrin fragment Bb12– 46. The epitope localization for mAbs I-5G, I-3B, III-10D and I-3C is indicated by arrows. 16 Fig. 6. The influence of the synthetic peptides Bb12–26 and Bb26–46 on fibrin desAABB polymerization in turbidity analysis. The depen- dence of the lag-time s (A), the maximum rate of fibrin polymerization V max (B) and final turbidity Dh (C) of fibrin clots on the molar ratio of the synthetic peptides to fibrin. E. V. Lugovskoy et al. Functional role of Bb-chain N-terminal fragment FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS 4543 ratio to fibrin had no inhibitory effect on fibrin poly- merization. To determine which stage of fibrin polymerization is affected by these peptides, we performed transmission electron microscopy at different stages of the fibrin polymerization process in the presence of peptides Bb12–26 and Bb26–46. Electron microscopy showed (Fig. 7B,C) that fibrin stayed in monomeric form when peptides Bb12–26 and Bb26–46 were present in poly- merization media at a molar ratio to fibrin of 1500. Without these peptides, fibrin formed protofibrils and their initial lateral associates as usual (Fig. 7A). How- ever, the cross-striated fibrils were formed in both cases with and without the peptides studied (Fig. 7D– F). These data show that peptides Bb12–26 and Bb26– 46 retard the stage of fibrin protofibril formation. Discussion The N-terminal fragment of the fibrin b-chain contains regions related to the polymerization process and to binding of thrombin and endothelial cells [11,12,18]. The known polymerization site ‘B’ is localized in sequence Bb15–18 [6]. Considerable indirect data indi- cate that other polymerization sites may be situated at the fibrin sequence Bb14–54 [11–16]. mAbs may be used as molecular probes to study mechanisms of fibrin polymerization [19–22]. The inhibition of fibrin polymerization by mAbs at a molar ratio to fibrin of about 2 or less most probably indicates that the epi- topes for these mAbs are situated at or near the poly- merization sites of the fibrin molecule. The specific retardation of fibrin polymerization by the correspond- ing smaller Fab-fragments suggests the blocking of a polymerization site by the mAb or its Fab-fragment rather than steric hindrance [23]. We have obtained two mAbs, I-3C and III-10d, with epitopes situated in human fibrin fragment Bb15–26, and two mAbs, I-5G and I-3B, with epitopes in frag- ment Bb26–36. Three of these mAbs (III-10d, I-5G, and I-3B) and their Fab-fragments specifically inhi- bited fibrin desAABB polymerization (Fig. 1). All of them, as well as their Fab-fragments, inhibited poly- merization of fibrin desAA. These results show that mAbs block a previously undescribed polymerization site, which is not a short peptide fragment like site ‘A’ (Aa17–19) or ‘B’ (Bb15–17), but comprises a longitudi- nal amino acid sequence in Bb15–36. As mAbs retard both fibrin desAABB and fibrin desAA polymeriza- tion, one can conclude that the blocked polymerization site does not coincide with polymerization site ‘B’ (Bb15–17). mAb 2d)2a targets an epitope encompassing the peptide bond BbArg14-Gly15, which is cleaved by thrombin [15]. This mAb inhibited the maximum rate of fibrin desAA polymerization by 60%, whereas its Fab-fragment inhibited it by 100%, at molar ratios of antibody to fibrin of 1 and Fab to fibrin of 2, respec- tively. Moen et al. [14] found impaired polymerization of fibrin obtained from recombinant fibrinogen with histidine substituted for arginine at Bb14. Turbidity analysis showed an increase in lag-time and a decrease in maximum polymerization rate of this fibrinogen in the desAA fibrin form. The final turbidity of these fibrin clots proved to be decreased, and electron microscopy showed that fibrin fibrils were thinner than ABC DEF Fig. 7. Electron micrographs of negatively contrasted structures formed during fibrin desAABB polymerization in 80 s (A,B,C) and 180 s (D,E,F) from the start of the process: (A,D) in the absence the synthetic peptides; (B,E) in the presence of Bb12–26 peptide; and (C,F) in the presence of Bb26–46 peptide. Initial protofibril lateral associates are indicated by arrows. The bars represent 100 nm. Functional role of Bb-chain N-terminal fragment E. V. Lugovskoy et al. 4544 FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS normal fibrils. It had been concluded that BbArg14 was involved in lateral association of normal fibrin protofibrils [14]. However, these authors performed electron microscopy of only the final fibrin clots, and not fibrin at each stage of the polymerization process. The question remained unresolved as to when during fibrin polymerization BbArg14 plays a role. Our results with mAbs 2d)2a [15] and the data obtained by Moen et al. [14] with the fibrinogen variant BbArg14His sug- gest that BbArg14 is involved in another fibrin poly- merization site localized at the N-terminal fragment of the fibrin Bb-chain. This site is operational before fibrinopeptide B is split off. The inhibitory activity of synthetic peptides mimick- ing the various fragments of the N-terminal region of the fibrin Bb-chain was investigated. Two peptides, Bb12–26 and Bb26–46, but not Bb19–26, Bb26–36, or Bb40–46, proved to be specific inhibitors of fibrin polymerization (the IC 50 values for Bb12–26 and Bb26–46 were 2.03 · 10 )4 m and 2.19 · 10 )4 m, respec- tively). Turbidity and electron microscopy data showed that peptides Bb12–26 and Bb26–46 inhibited the stage of fibrin protofibril formation. We suggest that two fibrin fragments corresponding to these peptides form the whole longitudinal site Bb12–46 (we have named it ‘C’), which takes part in fibrin intermolecular binding during the construction of two-stranded protofibrils simultaneously with site ‘A’ (Aa17–19). This site ‘C’ is operational without fibrinopeptide B (Bb1–14) splitting off; that is, it does not coincide with polymerization site ‘B’ (Bb15–17). Siebenist et al. found that fibrino- gen lacking fragment Bb1–42 was clotted by thrombin but with an essential delay of protofibril formation [12]. Pandya et al. [13] showed an inhibitory effect of peptide Bb15–42 on fibrin polymerization, and this effect was not determined by polymerization site ‘B’ (Bb15–18). Our results obtained with mAb 2d)2a [15] and with peptide Bb12–26, and the findings of Moen et al. [14], show that the polymerization site situated at the N-terminus of the Bb-chain comprises the amino acid residues localized to the left of the peptide bond Bb14–15. Pandya et al. [13] did not find an inhibitory activity for peptide Bb40–54 up to a molar ratio with fibrin of 1000. However, Moskowitz & Budzynski dis- covered that this peptide dissociated the (DD)E com- plex at a molar ratio of 5000 [16]. The latter result and our data obtained with peptide Bb26–46 show that polymerization site ‘C’ comprises amino acid residues to the right of the peptide bond Bb42–43. Fibrin protofibril formation is carried out by inter- molecular pairing of ‘A’ and ‘a’ polymerization sites localized in fibrin central E- and peripheral D-domains, respectively [3,4]. As polymerization site ‘C’ (Bb12–46) is situated in the E-domain of the fibrin molecule, the complementary site ‘c’ is suggested to be situated in the D-domain. We have tried to model the intermolecular binding between the D-domain of one fibrin desAA molecule and Bb12–46 of another (Fig. 8). The model was prepared with the pymol Molecular Graphics System [24] on the basis of X-ray analysis data of chicken fibrinogen [25] and of human D-dimer bound to the synthetic peptide GPRP [7]. X-ray analysis showed that the N-terminal fragments Bb1–64 and Aa1–27 of the fibrinogen molecule were not visible on electron density maps, as these frag- ments were highly disordered and mobile [25,26]. Human fibrinogen fragments Aa21–27 and Bb12–64 were substituted using computer graphics for chicken fibrinogen fragments Aa1–27 and Bb1–64. In this model, the C-terminal proline of the GPRP peptide, which is bound to the D-dimer molecule at site ‘a’, was manually linked to the N-terminal residue Aa21Val of human fibrinogen fragment Aa21–27. As a result, we have the possibility of finding the mutual space orientation between the fibrin desAA molecules belonging to different strands of protofibril. In the construction of the model, we have taken into consid- eration the fact that site ‘B’ (Bb15–18), being the inac- tive part of the suggested site ‘C’ (Bb12–46), has to be oriented to the complementary site ‘b’ in the bC-domain of another fibrin molecule. This model shows that fragment Bb12–46 of fibrin desAA (site ‘C’) covers a longitudinal complementary site ‘c’ situ- ated in the bC- and cC-domains of a molecule γC γC βC BβArg14 D E D βC ‘‘A’’-‘‘a’’ 17 Fig. 8. The model (in two projections) of the intermolecular connec- tion between the D-domain of one fibrin desAA molecule (blue) and Bb12–46 (magenta) of another. The model was prepared with PYMOL [24] on the basis of the X-ray analysis data of chicken fibrinogen [25] and human D-dimer bound with synthetic peptide GPRP [7]. E. V. Lugovskoy et al. Functional role of Bb-chain N-terminal fragment FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS 4545 belonging to another strand within the protofibril. After fibrinopeptide B is split off the site, ‘B’ is formed, and the latter interacts intermolecularly with the complementary site ‘b’ situated in the bC-domain in a ‘knob’–‘hole’ type of interaction [7,9]. The remain- ing part of site ‘C’ (Bb19–46) probably remains bound to the fibrin D-domain. To determine the localization of a complementary site ‘c’, it is necessary to perform X-ray analysis of the complex between D-dimer and peptide Bb12–46. Recently, Pechik et al. [27] found an interaction of recombinant peptide (Bb1–66) 2 with fibrin D-dimer and D-monomer (K d ¼ 1.3 · 10 )5 m and K d ¼ 1.53 · 10 )4 m, respectively), using the sur- face plasmon resonance method. Thus, our results suggest that the longitudinal N-terminal region of the Bb-chain (Bb12–46) or site ‘C’ is involved in protofibril formation simultaneously with site ‘A’ (Aa17–19) before fibrinopeptide B is split off. Synthetic peptide GPRP (which mimics the polymerization site ‘A’) binds to site ‘a’ of fibrinogen with K a  1 · 10 4 m [28], whereas fibrin desAA inter- action is carried out with a K a  1.56 · 10 7 m [29]. This increase in affinity may be explained by the sug- gested intermolecular pairing of ‘C’–‘c’ sites. It is known that fibrinogen is able to polymerize under special conditions without splitting off fibrinopeptides A and B, forming cross-striated fibrils [30]. This poly- meric interaction of fibrinogen molecules may be explained by the participation of the ‘C’–‘c’ pairing sites, as both sites are probably exposed in the fibrin- ogen molecule. Experimental procedures Preparation of fibrinogen, fibrin desAA, fibrin desAABB and t-NDSK Human fibrinogen was prepared by sodium sulfate precipi- tation from human plasma [31]. DesAABB fibrin monomer was prepared as described by Belitser et al. [32] 3 . DesAA fibrin monomer was prepared by our original method as described previously [33]. t-NDSK was obtained as des- cribed by Timpl & Gollwitzer [34] 4 . Preparation and purification of mAbs Hybridomas were obtained essentially as described by Ko ¨ h- ler & Milstein [35]. mAbs were isolated from hybridoma culture medium by affinity chromatography on fibrinogen Sepharose 4B, as described elsewhere [36]. The determina- tion of IgG class and subclass was performed by ELISA using an Isotyping kit (Clinical Credential; ICN Immunobio- logicals, Lisle, IL, USA) 5 . Preparation of Fab-fragments IgG sample (0.4 mL) (2 mgÆmL )1 ) in 0.02 m potassium phos- phate buffer (pH 7.4) with 0.13 m NaCl (NaCl ⁄ P i ) was added to 1.2 mL of papain Sepharose 4B suspended in digestion buffer (NaCl ⁄ P i with 20 mm cysteine HCl, 10 mm EDTA, pH 7.0) and incubated for 30 min at room temperature with stirring. Fab fragments in supernatant were separated from Fc fragments using 0.4 mL of protein G Sepharose 4B (50% slurry in NaCl ⁄ P i ). Desalting and concentration were done using centrifuge filter units (Ultrafree-15 with Biomax 10K membrane, Millipore, Bedford, MA, USA) 6 . Synthesis of peptides Five peptides comprising human fibrinogen region Bb12–46 (Bb12–26, Bb19–26, Bb26–36, Bb26–46 and Bb40–46) were synthesized by a solid-phase method (Fmoc chemistry). ELISA ELISA was performed in microtiter plates coated with the following antigens: t-NDSK, fibrinogen, fibrin desAABB, Bb15–118 and c1–78 fragments of t-NDSK, and E 3 -frag- ment. Coating was achieved by adding to the wells 110 lL of solutions (10 lgÆmL )1 ) of antigens (fibrinogen in 0.2 m ammonium acetate buffer, pH 8.5; fibrin desAABB, Bb15– 118 and c1–78 fragments of t-NDSK in 0.2 m ammonium acetate buffer, pH 8.5 with 3.0 m urea; t-NDSK and E 3 -fragment in 0.02 m sodium bicarbonate buffer, pH 9.5), with subsequent incubation for 18 h at 4 °C. The plates were washed three times with NaCl ⁄ P i containing 0.05% Tween-20 (TPBS), and 100 lL of mAbs solutions in NaCl ⁄ P i were added to the wells and incubated for 60 min at 37 °C. After washing of the plate, 100 lL aliquots of a 1 : 1000 solution in TPBS of the rabbit anti-(mouse IgG) conjugated with horseradish peroxidase (Sigma 7 -Aldrich, St Louis, MO, USA) were added to each well. After subse- quent incubation (60 min, 37 °C) and washing with NaCl ⁄ P i , 0.03% hydrogen peroxide and 0.04% o-phenylene- diamine were added to each well. The reaction was stopped by adding 50 lL of 2.0 m sulfuric acid. The absorbance at 492 nm was read with an Autoreader RT 2100 C (Rayto, Nanshan, China) 8 . To determine whether mAbs compete with each other for the binding site of the antigen, a competitive ELISA was performed as follows. Microtiter plates were coated with fibrinogen, and the plates were washed with TPBS. The mixtures of competing mAbs at various concentrations and biotinylated mAbs at a constant concentration were added to the wells. After incubation (2 h, 37 °C), the wells were washed, and streptavidin conjugated with horseradish per- oxidase was added to the wells. All subsequent procedures were the same as described above. Functional role of Bb-chain N-terminal fragment E. V. Lugovskoy et al. 4546 FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS To study the binding of mAbs to fibrin synthetic peptides (Bb12–26, Bb26–46 Bb40–46, Bb19–26 and Bb26–36), a competitive ELISA was carried out as follows. Microtiter plates were coated with fibrinogen, and the plates were washed with TPBS. The mixtures of competing peptides at various concentrations and appropriate mAbs at a constant concentration were added to the wells. After incubation (2 h, 37 °C) and washing, a 100 lL aliquot of a 1 : 1000 solution in TPBS of rabbit anti-(mouse IgG) (Sigma- Aldrich) conjugated with horseradish peroxidase was added to each well. All subsequent procedures were the same as described above. Determination of dissociation constants (K D ) K D values were determined by indirect competitive ELISA as described by Friguet et al. [37] 9 . In brief, microtiter wells were coated with either t-NDSK or fibrinogen, and mix- tures of mAbs with relevant antigen (fibrinogen, t-NDSK, etc.) were added to the wells. The concentration of mAb was kept constant, and the concentration of competing antigen was varied. The plates were incubated for 1 h at 37 °C, and washed three times with TPBS. Quantification of the mAbs bound was performed with rabbit anti-(mouse IgG) (Sigma-Aldrich) conjugated with horseradish peroxi- dase as described above. Immunoblot analysis Immunoblot analysis was used to examine the reactivity of mAbs obtained to t-NDSK and to its Bb15–118 peptide. Briefly, 4 lg 10 of t-NDSK reduced by 5% b-mercaptoethanol and 1.3 lg of fragment Bb15–118 were separated by SDS ⁄ PAGE in 15% polyacrylamide gel, and the proteins were electrophoretically transferred to nitrocellulose mem- branes (Hybond ECL; Amersham, Uppsala, Sweden) 11 . Thereafter, the membranes were blocked with dried fatless milk (3.5% in NaCl ⁄ P i ) overnight at 4 °C. Subsequently, the membranes were washed twice in TPBS, and then incu- bated with mAbs (100 lgÆmL )1 ) in TPBS for 2 h at 37 °C. The membranes were washed three times in TPBS, and incubated with Link rabbit anti-(mouse IgG) (Sigma- Aldrich) diluted in TPBS (1 : 100) for 45 min at 37 °C. Then, the membranes were washed twice in TPBS and incubated with peroxidase–antiperoxidase 12;13 complex (Dako, Glostrup, Denmark) 12;13 diluted in TPBS (1 : 800) for 45 min at 37 °C. The membranes were washed twice in TPBS. Finally, the proteins on the membranes were visualized using 4-chloro-a-naphtol and H 2 O 2 as substrates. Turbidity analysis of fibrin polymerization The effects of mAbs or synthetic peptides at various con- centrations on fibrin polymerization were studied spectro- photometrically at 350 nm as described previously [15]. The curve of increasing turbidity during fibrin clotting shows the following parameters: s, the lag-time, which corre- sponds to the time of protofibril formation; V max , maxi- mum rate of fibrin polymerization, which was defined by graphic calculation of the angle of the tangent to the tur- bidity increase curve at the point of maximum steepness; and Dh, final turbidity of fibrin clots. Polymerization of fibrin desAA and desAABB was studied at a final concen- tration 0.1 mgÆmL )1 in the polymerization medium contain- ing 0.05 m ammonium acetate (pH 7.4) with 0.1 m NaCl and 1 · 10 )4 m CaCl 2 . Polymerization of fibrin formed in the fibrinogen + thrombin reaction was investigated at a final concentration of fibrinogen of 0.1 mgÆmL )1 and a final concentration of thrombin of 0.4 NIH unitsÆmL )1 in the same polymerization medium. Electron microscopy The samples of polymerizing fibrin desAABB in the absence or presence of synthetic peptides Bb12–26 or Bb26–46 were taken out of the reaction medium at various times, placed on a carbon-coated grid for 2 min, and then stained with 1% (w ⁄ v) uranyl acetate for 1 min. Transmission electron microscopy was performed in a H-600 electron microscope (Hitachi, Chiyoda, Japan) 14 operated at 75 kV. Electron micrographs were obtained at a magnification of · 50 000 on Kodak SO-163 film. Acknowledgements We are grateful to Professor Ulf Hellman (Ludwig Institute for Cancer Research, Uppsala, Sweden) for the differentiation between Bb15–118 and c1–78 poly- peptide remnants of t-NDSK by MALDI and to Professor Russell Doolittle (Center for Molecular Genetics, University of California, San Diego, CA, USA) for discussion of the results obtained. References 1 Blomback B (1996) Fibrinogen and fibrin ) proteins with complex roles in haemostasis and thrombosis. Thromb Res 83, 1–75. 2 Spraggon G, Everse SJ & Doolittle RF (1997) Crystal structures of fragment D from human fibrinogen and its crosslinked counterpart from fibrin. Nature 389, 455–462. 3 Fowler WE, Hantgan RR, Hermans J & Erickson HP (1981) Structure of the fibrin protofibril. Proc Natl Acad Sci USA 78, 4872–4876. 4 Weisel JW, Veklich Y & Gorkun O (1993) The sequence of cleavage of fibrinopeptides from fibrinogen is impor- E. V. Lugovskoy et al. Functional role of Bb-chain N-terminal fragment FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS 4547 tant for protofibril formation and enhancement of lat- eral agregation in fibrin clots. J Mol Biol 232, 285–297. 5 Weisel JW (1986) Fibrin assembly. Lateral aggregation and the role of the two pairs of fibrinopeptides. Biophys J 50, 1079–1093. 6 Laudano AP & Doolittle RF (1978) Synthetic peptide derivatives that bind to fibrinogen and prevent the poly- merisation of fibrin monomers. Proc Natl Acad Sci USA 75, 3085–3089. 7 Everse SJ, Spraggon G, Veerapandin L, Riley M & Doolittle RF (1998) Crystal structure of fragment dou- ble-D from human fibrin with two different bound ligands. Biochemistry 37, 8637–8642. 8 Mihalyi E (1988) Clotting of bovine fibrinogen. Calcium binding to fibrin during clotting and its dependence on release of fibrinopeptide B. Biochemistry 27, 967–976. 9 Everse SJ, Spraggon G, Veerapandian L & Doolittle RF (1999) Conformational changes in fragments D and double-D from human fibrin(ogen) upon binding the peptide ligand Gly-His-Arg-Pro-amide. Biochemistry 38, 2941–2946. 10 Yang Z, Mochalkin I & Doolittle RF (2000) A model of fibrin formation based on crystal structures of fibrin- ogen and fibrin fragments complexed with synthetic peptides. Proc Natl Acad Sci USA 97, 14156–14161. 11 Pandya BV, Cierniewski CS & Budzynski AZ (1985) Conservation of human fibrinogen conformation after cleavage of the Bb-chain amino terminus. J Biol Chem 260, 2994–3000. 12 Siebenist KR, Di Orio JP, Budzynski AZ & Mosesson MW (1990) The polymerisation and thrombin-binding properties of des-(Bb1–42)-fibrin. J Biol Chem 265, 18650–18655. 13 Pandya BV, Gabriel JL, O’Brien J & Budzynski AZ (1991) Polymerisation site in the beta chain of fibrin: mapping of the Bbeta1–55 sequence. Biochemistry 30, 162–168. 14 Moen JL, Gorkun OV, Weisel JW & Lord ST (2003) Recombinant BbetaArg14His fibrinogen implies partici- pation of N-terminus of Bbeta chain in desAA fibrin polymerisation. Blood 102, 2466–2471. 15 Lugovskoi EV, Makogonenko EM, Chudnovets VS, Derzskaya SG, Gogolinsikaja GK, Kolesnikova IN, Bukhanevich AM & Komisarenko SV (1991) The study of fibrin polymerisation with monoclonal antibodies. Biomed Sci 2, 249–296. 16 Moskowitz KA & Budzynski AZ (1994) The (DD)E complex is maintained by a composite fibrin polymerisa- tion site. Biochemistry 33, 12937–12944. 17 Kini RM & Evans HJ (1996) Prediction of potential protein–protein interaction sites from amino acid sequence. Identification of a fibrin polymerisation site. FEBS Lett 385, 81–86. 18 Gorlatov S & Medved L (2002) Interaction of fibrin(ogen) with the endothelial cell receptor VE-cadherin: mapping of the receptor-binding site in the NH2-terminal portions of the fibrin beta chains. Biochemistry 41 , 4107–4116. 19 Lugovskoi EV & Komisarenko SV (2000) The use of monoclonal antibodies for studying the fibrin polymerisation. Russian J Bioorganic Chem 26, 791– 798. 20 Wisloff F, Michaelsen TE & Godal HC (1984) Inhibi- tion or acceleration of fibrin polymerisation by mono- clonal immunoglobulins and immunoglobulin fragments. Thromb Res 35, 81–90. 21 Lugovskoy EV, Gritsenko PG, Kolesnikova IN, Zolotareva EN, Chernishov VI, Nieuwenhuizen W & Komisarenko SV (2004) Two monoclonal antibodies to D-dimer ) specific inhibitors of fibrin polymerisation. Thromb Res 113, 251–259. 22 Cierniewski CS & Budzynski AZ (1992) Involvement of the a-chain in fibrin clot formation. Effect of monoclo- nal antibodies. Biochemistry 31, 4248–4253. 23 Wasser MN, Welling M, Lamers G, Pauwels EK & Nieu- wenhuizen WA (1990) Effect of an antifibrin monoclo- nal antibody and fragments thereof on some properties of fibrin. Thromb Haemost 63, 39–43. 24 DeLano WL (2002) The PyMOL Molecular Graphics System. DeLano Scientific, San Carlos, CA. 25 Yang Z, Kollman JK, Pandi L & Doolittle RF (2001) Crystal structure of native chicken fibrinogen at 2.7 A ˚ resolution. Biochemistry 40, 12515–12523. 26 Pechik I, Madrazo J, Mosesson MW, Hernandez I, Gilliland GL & Medved L (2004) Crystal structure of the complex between thrombin and the central ‘E’ region of fibrin. Proc Natl Acad Sci USA 101, 2718– 2723. 27 Pechik I, Yakovlev S, Mosesson MW, Gilliland GL & Medved L (2006) Structural basis for sequential cleav- age of fibrinopeptides upon fibrin assembly. Biochemis- try 45, 3588–3597. 28 Laudano AP & Doolittle RF (1980) Studies on synthetic peptides that bind to fibrinogen and prevent fibrin poly- merisation. Structural requirements, number of binding sites, and species differences. Biochemistry 19, 1013– 1019. 29 Lewis SD, Shields PP & Shafer JA (1985) Character- ization of the kinetic pathway for liberation of fibri- nopeptides during assembly of fibrin. J Biol Chem 260, 10192–10199. 30 Williams RC (1983) Band patterns seen by electron microscopy in ordered arrays of bovine and human fibrinogen and fibrin after negative staining. Proc Natl Acad Sci USA 80, 1570–1573. 31 Varetskaya TV (1960) Microheterogeneity of fibrinogen. Cryofibrinogen. Ukr Biokhim Zhurn 32, 13–24. 32 Belitser VA, Varetskaja TV & Malneva GV (1968) Fibrinogen–fibrin interaction. Biochim Biophys Acta 154, 367 15 –375. Functional role of Bb-chain N-terminal fragment E. V. Lugovskoy et al. 4548 FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS 33 Lugovskoi EV & Gogolinskaya GR (1999) Preparation of fibrin desAA by thrombin. Ukr Biokhim Zhurn 71, 107–108. 34 Timpl R & Gollwitzer R (1973) Cyanogen bromide cleavage of bovine fibrinogen. Identification of a dimeric N-terminal peptide and two other disulfide containing fragments. FEBS Lett 29, 92–96. 35 Ko ¨ hler G & Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specifity. Nature 256, 495–497. 36 Heene DL & Matthias FR (1973) Adsorption of fibrino- gen derivates on insolubilized fibrinogen and fibrinmo- nomer. Thromb Res 2, 137–154. 37 Friguet B, Alain F, Chaffte F, Djavadi-Ohaniance L & Goldberg ME (1985) Measurement of the true affinity constant in solution of antigen–antibody complexes by enzyme-linked immunosorbent assay. J Immunol Methods 77, 305–319. E. V. Lugovskoy et al. Functional role of Bb-chain N-terminal fragment FEBS Journal 274 (2007) 4540–4549 ª 2007 The Authors Journal compilation ª 2007 FEBS 4549 . situated in the cC-domain of the fibrinogen ⁄ fibrin molecule [7]. On protofibril formation, polymerization site ‘B’ (Bb15– 18) is exposed at the N-terminal region of the fibrin b-chain, when fibrinopeptide. BbArg14 is involved in another fibrin poly- merization site localized at the N-terminal fragment of the fibrin Bb-chain. This site is operational before fibrinopeptide B is split off. The inhibitory. fibrin desAABB and fibrin produced in the fibrinogen + thrombin reaction (Fig. 6A,B). The values of IC 50 (the concentrations of the peptides at which 50% of inhibition of the fibrin polymerization