scholarly journals The C‐terminal acidic region in the A1 domain of factor VIII facilitates thrombin‐catalyzed activation and cleavage at Arg 372

Author(s):  
Yuto Nakajima ◽  
Keiji Nogami
Keyword(s):  
2004 ◽  
Vol 279 (32) ◽  
pp. 33104-33113 ◽  
Author(s):  
Keiji Nogami ◽  
Jan Freas ◽  
Chandrashekhara Manithody ◽  
Hironao Wakabayashi ◽  
Alireza R. Rezaie ◽  
...  

Author(s):  
Yuto Nakajima ◽  
Hiroaki Minami ◽  
Keiji Nogami

AbstractFactor VIII (FVIII) is activated by thrombin-catalyzed cleavage at Arg372, Arg740, and Arg1689. Our previous studies suggested that thrombin interacted with the FVIII C2 domain specific for cleavage at Arg1689. An alternative report demonstrated, however, that a recombinant (r)FVIII mutant lacking the C2 domain retained >50% cofactor activity, indicating the presence of other thrombin-interactive site(s) associated with cleavage at Arg1689. We have focused, therefore, on the A3 acidic region of FVIII, similar to the hirugen sequence specific for thrombin interaction (54–65 residues). Two synthetic peptides, spanning residues 1659–1669 with sulfated Tyr1664 and residues 1675–1685 with sulfated Try1680, inhibited thrombin-catalyzed FVIII activation and cleavage at Arg1689. Treatment with 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide to cross-link thrombin with either peptide showed possible contributions of both 1664–1666 and 1683–1684 residues for thrombin interaction. Thrombin-catalyzed activation and cleavage at Arg1689 in the alanine-substituted rFVIII mutants within 1663–1666 residues were similar to those of wild type (WT). Similar studies of 1680–1684 residues, however, demonstrated that activation and cleavage by thrombin of the FVIII mutant with Y1680A or D1683A/E1684A, in particular, were severely or moderately reduced to 20 to 30% or 60 to 70% of WT, respectively. Surface plasmon resonance-based analysis revealed that thrombin interacted with both Y1680A and D1683A/E1684A mutants with approximately sixfold weaker affinities of WT. Cleavage at Arg1689 in the isolated light-chain fragments from both mutants was similarly depressed, independently of the heavy-chain subunit. In conclusion, the 1680–1684 residues containing sulfated Tyr1680 in the A3 acidic region also contribute to a thrombin-interactive site responsible for FVIII activation through cleavage at Arg1689.


Blood ◽  
2005 ◽  
Vol 106 (11) ◽  
pp. 1017-1017
Author(s):  
Keiji Nogami ◽  
Midori Shima ◽  
Tomoko Matsumoto ◽  
Katsumi Nishiya ◽  
Masahiro Takeyama ◽  
...  

Abstract Factor VIII (FVIII) functions as a cofactor for factor IXa in the intrinsic tenase complex. This tenase activity is down-regulated by activated protein C (APC) or factor Xa (FXa). Plasmin, the most potent fibrinolytic protease, inactivates FVIII as well as other clotting factors. However, the mechanism of FVIII inactivation by plasmin is poorly understood. FVIII activity reached to the peak value of ~2-fold increase at 3 min after the addition of plasmin in a one-stage clotting assay. Then, the activity was decreased rapidly and was undetectable within 30 min. This time-dependent reaction was not affected in the presence of von Willebrand factor and phospholipid. The activation of FVIII by plasmin was an ~50% level of that by FXa. The rate constant (min-1) of inactivation of FVIIIa by plasmin possessed ~11.3- and ~2.5-folds greater than those by FXa and APC in the presence of protein S, respectively. SDS-PAGE analysis showed that plasmin cleaved the 90~210-kDa heavy chain of FVIII to 50, 48,45, 40, and 38-kDa fragments via 90-kDa fragment. Using western blot and N-terminal sequence analyses, these fragments derived from the heavy chain were identified as A11-372, A1337-372-A2, A11-336, A2, and A137-336, respectively, by cleavages at Arg372, Arg740, Lys36 and Arg336 in the A1 domain. On the other hand, the 80-kDa light chain was cleaved to 67-kDa fragment via 70-kDa fragment by cleavages at Arg1721 and Arg1689, respectively, consistent with the pattern of cleavage by FXa. However, the cleavage at Arg336 by plasmin was much quicker than that at Arg372, contrast with that by FXa. Furthermore, this cleavage was faster than that by APC, consistent with rapid inactivation of FVIII. In addition, the cleavage at Arg336 of FVIIIa by plasmin was faster than that of isolated A1 or A1/A3-C1–C2 dimer, different with that by FXa. These results demonstrate the importance of cleavage at Arg336 for the mechanism of plasmin-catalyzed FVIII inactivation. Furthermore, this cleavage appears to be selectively modulated by the A2 domain that may interact with plasmin.


Blood ◽  
2006 ◽  
Vol 108 (11) ◽  
pp. 1716-1716
Author(s):  
Andrew J. Gale ◽  
Diana Rozenshteyn ◽  
Justin Riceberg

Abstract Neutrophils and monocytes express cathepsin G and elastase and also can bind to activated platelets, thus they can be localized to the site of active coagulation. Early studies suggested that cathepsin G and elastase inactivated factor VIII (FVIII) and were thus anticoagulant. But other studies have suggested procoagulant functions for cathepsin G and elastase in activation of factor V or activation of platelets among other possible mechanisms. Therefore, we investigated the effects of human neutrophil elastase and human neutrophil cathepsin G on FVIII/VIIIa. Elastase does inactivate both FVIII and FVIIIa but cathepsin G activates FVIII while having very little effect on FVIIIa. Cathepsin G activation of FVIII is enhanced by phospholipid vesicles, apparently due to enhanced rate of cleavage and stabilization of the resulting molecule. The maximum level of activation is less than that of thrombin, but it is still four-fold as measured in an APTT assay. Cleavage sites for both proteases in FVIII were identified by Edman degradation and gel analysis. FVIII cleavages are limited to a few specific sites that are mostly located near known activating and inactivating cleavage sites. A notable exception is a cleavage site for elastase after valine 26 in the A1 domain. Cathepsin G cleavage sites near to thrombin cleavage sites likely contribute to the partial activation of FVIII. The unique elastase cleavage site at valine 26 likely contributes to the inactivation of FVIII and FVIIIa. Therefore, it is possible that neutrophils and monocytes may provide some pro-coagulant effect by activating FVIII and may also provide negative feedback by inactivating FVIIIa as well.


Blood ◽  
2008 ◽  
Vol 112 (11) ◽  
pp. 1036-1036
Author(s):  
Tetsuhiro Soeda ◽  
Keiji Nogami ◽  
Tomoko Matsumoto ◽  
Kenichi Ogiwara ◽  
Katsumi Nishiya ◽  
...  

Abstract Factor VIIa (FVIIa), complexed with tissue factor (TF), is a trigger of blood coagulation through activation of factor X in the initiation phase. FVIIa can catalyze intrinsic clotting factors such as not only factor IX, but also factor VIII (FVIII). However the role and the mechanisms of the FVIIa-catalyzed FVIII are poorly understood. We first examined FVIIa-catalyzed FVIII activation in the presence of phospholipid (PL) using a one-stage clotting assay. The levels of FVIII activity elevated rapidly by ~4-fold within 30 sec after the addition of FVIIa (1 nM)-TF (1 nM)complex, and subsequently decreased to the initial level within 20 min. This time-dependent reaction was enhanced by the presence of TF and PL in dose-dependent manners, but was moderately inhibited (~50%) in the presence of von Willebrand factor at physiological concentration of 10 μg/mL. FVIII cleavage was evaluated using western blotting immediately after the addition of FVIIa-TF complex. The heavy chain of FVIII was proteolyzed more rapidly (at 15 sec) by cleavages at Arg740 (A2-B junction) and Arg372 (A1-A2 junction) by FVIIa-TF complex, whilst the cleavage at Arg336 in the A1 domain was appeared at ~2.5 min. However little cleavage of the light chain of FVIII was observed, supporting that cleavages at Arg740/Arg372 and Arg336 by FVIIa-TF complex contribute to the up- and down-regulation of FVIII(a) activity, respectively. Of interest, no proteolysis of isolated intact heavy chain was observed, indicating that the proteolysis of the heavy chain was governed by the presence of the light chain. Compared to FVIII activation by thrombin (0.1–1 nM), the activation by FVIIa (0.1–1 nM)-TF (1 nM) complex was observed more rapidly. The activation rate observed by the addition of FVIIa-TF complex was ~50-fold greater than that by thrombin. Kinetics by the chromogenic Xa generation assay showed the catalytic efficiency (kcat/Km; 8.9 min−1/32.8 nM) on FVIIa-TF complex-catalyzed FVIII activation showed ~4-fold greater than that on thrombin-catalyzed activation (kcat/Km; 7.5 min−1/86.4 nM). Furthermore, the catalytic efficiencies on cleavages at Arg740 and Arg372 of FVIII by FVIIa-TF complex were ~3- and ~20-fold greater compared to those by thrombin, respectively. These findings suggested that FVIIa-TF complex was a greater FVIII activator than thrombin in very early phase. In order to localize the binding region for FVIIa, we evaluated the interactions between FVIII subunit and Glu-Gly-Arg-active site modified FVIIa, lacking enzymatic activity, in a surface plasmon resonance-based assay. The heavy chain of FVIII bound to EGR-FVIIa with higher affinity than the light chain (Kd; 2.1 and 45 nM, respectively). Binding was particularly evident with the A2, A3, and C2 domains (Kd; 34, 37, and 44 nM, respectively), whilst the A1 domain failed to bind. In conclusion, we demonstrated that FVIIa-TF complex rapidly activated FVIII by proteolysis of the heavy chain and the activation was governed by the presence of the light chain. Furthermore, present results suggested the role of TF-dependent FVIII activation by FVIIa which is responsible for the initiation phase of blood coagulation.


Blood ◽  
2009 ◽  
Vol 114 (22) ◽  
pp. 3176-3176
Author(s):  
Hironao Wakabayashi ◽  
Amy E Griffiths ◽  
Philip Fay

Abstract Abstract 3176 Poster Board III-100 Factor VIII (FVIII) consists of a heavy chain (A1A2B domains) and light chain (A3C1C2 domains), while the contiguous A1A2 domains are separate subunits in the cofactor, FVIIIa. Previously we have generated FVIII mutants with enhanced stability by mutating residues located at A1-A2 or A2-A3 interfaces (Wakabayashi et al, Blood, 112, 2761-9, 2008, Wakabayashi et al, J. Thromb. Haemost. 7, 438-44, 2009). FVIII X-ray structures show close contacts between the Ca2+ binding site contained within the A1 domain and the C2 domain of LC. In this study we mutated residues located at this interface to examine the effects on FVIII(a) stability. Studies assessing FVIII thermal and chemical stability involved monitoring the rates of loss of FVIII activity by FXa generation assay following incubation of FVIII (4 nM) at 57°C or in various concentrations of guanidinium (0-1.2 M). The rate of decay of FVIIIa was monitored over time at 23°C using FXa generation assays following activation of FVIII (1.5 nM) with thrombin. Data were fitted to single exponential decay equations and rates of decay were compared. In one variant, a disulfide bond was introduced between the two domains by a double mutation at Arg121 in A1 and Leu2302 in the C2 domain to Cys (R121C/L2302C). In addition, based on the finding that there is a gap between the methyl groups of Ala108 (A1 domain) and Ala2328 (C2 domain) we mutated Ala108 to Val, Ile, or Leu to examine whether these mutants increase the stability of FVIII by an improved hydrophobic interaction at this site. Significant increases in FVIII thermal stability, up to 4-fold compared with WT, were observed in R121C/L2302C, Ala108Ile, and Ala108Leu. R121C/L2302C and Ala108Ile retained ∼80% FVIII activity as measured by FXa generation assay compared to WT value, however, that of Ala108Leu was ∼25% the WT value. Only Ala108Ile showed an improvement in chemical stability (10% increase in IC50 value as compared with WT FVIII) and FVIIIa decay due to A2 subunit dissociation was similar to WT FVIII (20-40% reduction in FVIIIa decay rate compared to WT). Ca2+ is necessary for FVIII function and EGTA (2 mM) reduced WT FVIII activity by ∼70%. However, EGTA-treated R121C/L2302C FVIII retained ∼100% activity, suggesting that the Ca2+ requirement for FVIII function may be substituted by covalent bonding between the Ca2+ binding region in A1 and C2 subunit. Furthermore, the Ala108Ile variant showed ∼60% activity remaining after EGTA treatment suggesting partial relief of this Ca2+ dependency for stability of the A1-C2 interaction. Next, we tested whether the mutations at the A1-C2 interface can be combined with mutations at A1-A2 or A2-A3 interfaces to generate a FVIII with further improved stability. Previously characterized FVIII variants, designated A domain mutants, showing up to 2-fold increases in thermal stability compared with WT FVIII included Asp519Ala, Asp519Val, Glu665Ala, Glu665Val, Glu1984Ala, and Glu1984Val. In combining those mutations with either R121C/L2302C or Ala108Ile, we obtained variants with >5-fold increases in thermal stability (9/12 mutants), with the Ala108Ile/Glu665Val variant showing the greatest increase (∼10-fold). Most of the mutants (9/12) showed normal FVIII activity values by FXa generation assay (>60%) and 15-30% increases in IC50 values for chemical stability as compared with WT. In addition, the high FVIIIa stability of the A domain mutants was largely preserved in the combined mutations. Collectively, these results suggest that alterations at this A1-C2 contact region by covalent modification or increasing hydrophobic interaction yields improved FVIII stability that can be combined with other high stability mutations to produce additive effects. Disclosures No relevant conflicts of interest to declare.


Blood ◽  
1989 ◽  
Vol 74 (5) ◽  
pp. 1618-1626 ◽  
Author(s):  
D Scandella ◽  
M Mattingly ◽  
S de Graaf ◽  
CA Fulcher

Human factor VIII(FVIII) inhibitors are pathologic, circulating antibodies that inactivate FVIII. We have examined the location of epitopes on the FVIII protein for inhibitors from hemophilia A and nonhemophilic individuals. The inhibitors were of type I or type II in the kinetics of their inactivation of FVIII. A cDNA clone of human FVIII was used to express defined FVIII protein fragments in Escherichia coli for immunoblotting with inhibitor plasma. An epitope for 18 heavy-chain inhibitors was localized to the aminoterminal 18.3 Kd of the A2 domain. Two of these inhibitors also recognized an epitope located between A1 and A2 domains. Similarly, an epitope for 23 light- chain inhibitors was localized to the C2 domain. Weaker epitopes for 13 of the same inhibitors within the C1 and C2 domains were also observed. Four of the 23 inhibitors in addition bound strongly to the A3 domain. Most inhibitors (22 of 23) were neutralized in vitro only by the FVIII fragments to which they bound on immunoblots; however, one inhibitor that was neutralized by a fragment containing the A1 domain did not bind to it on immunoblots. Conversely, 3 of 3 inhibitors that bound to the A3 domain and 5 of 15 that bound to the A2 domain were not neutralized by the corresponding fragments. The epitope specificity of an inhibitor did not depend on its source or type. Our results show that FVIII inhibitors bind to limited areas within the heavy and light chains of FVIII. Some inhibitor plasmas contain additional antibodies that may not be inhibitory.


2002 ◽  
Vol 118 (2) ◽  
pp. 589-594 ◽  
Author(s):  
Andrew D. Mumford ◽  
Michael Laffan ◽  
James O'Donnell ◽  
John H. McVey ◽  
Daniel J. D. Johnson ◽  
...  
Keyword(s):  

1996 ◽  
Vol 75 (01) ◽  
pp. 036-044 ◽  
Author(s):  
Thomas L Ortel ◽  
Karen D Moore ◽  
Mirella Ezban ◽  
William H Kane

SummaryFactor VIII and factor V share a repetitive domain structure of A1-A2-B-A3-C1-C2. To define the region(s) within the factor VIII heavy chain that result in inefficient expression of the recombinant protein, we expressed a series of factor VIH/factor V chimeras that contained heterologous sequences from the A1 and/or A2 domains. Substitution of the factor VIIIA1 domain dramatically reduced secretion of factor V ~ 500-fold, whereas substitution of the factor VIII A2 domain had minimal effect on secretion. Conversely, substitution of the factor V A1 domain increased secretion of factor VIII ~3-fold, whereas substitution of the factor V A2 domain actually reduced secretion ~4-fold. Pulse chase experiments confirmed that reduced expression levels were due to decreased secretion rather than instability of secreted protein. Smaller substitutions did not further localize within the A1 domain the regions responsible for inefficient secretion.


Sign in / Sign up

Export Citation Format

Share Document