Portal de Periódicos da CAPES

Generation of a novel factor IX with augmented clotting activities in vitro and in vivo

2010; Elsevier BV; Volume: 8; Issue: 8 Linguagem: Inglês

10.1111/j.1538-7836.2010.03913.x

1538-7933

Chia-Ni Lin, C Kao, Carol H. Miao, Nobuko Hamaguchi, Hua‐Lin Wu, Guey-Yueh Shi, Y.L. LIU, Katherine A. High, Shu‐Wha Lin,

Blood Coagulation and Thrombosis Mechanisms

Journal of Thrombosis and HaemostasisVolume 8, Issue 8 p. 1773-1783 ORIGINAL ARTICLEFree Access Generation of a novel factor IX with augmented clotting activities in vitro and in vivo C. N. LIN, C. N. LIN Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, Taipei, TaiwanSearch for more papers by this authorC. Y. KAO, C. Y. KAO Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, Taipei, TaiwanSearch for more papers by this authorC. H. MIAO, C. H. MIAO Seattle Children's Hospital Research Institute Department of Pediatrics, University of Washington, Seattle, WASearch for more papers by this authorN. HAMAGUCHI, N. HAMAGUCHI Department of Biochemistry, Brandeis University, Waltham, MA, USASearch for more papers by this authorH. L. WU, H. L. WU Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University Cardiovascular Research Center, National Cheng Kung University, Tainan, TaiwanSearch for more papers by this authorG. Y. SHI, G. Y. SHI Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University Cardiovascular Research Center, National Cheng Kung University, Tainan, TaiwanSearch for more papers by this authorY. L. LIU, Y. L. LIU Division of Hematology and Center for Cellular and Molecular Therapeutics, The Children's Hospital of PhiladelphiaSearch for more papers by this authorK. A. HIGH, K. A. HIGH Division of Hematology and Center for Cellular and Molecular Therapeutics, The Children's Hospital of Philadelphia Howard Hughes Medical Institute Department of Pediatrics, School of Medicine, University of Pennsylvania, Philadelphia, PA, USASearch for more papers by this authorS. W. LIN, S. W. LIN Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, Taipei, Taiwan Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, TaiwanSearch for more papers by this author C. N. LIN, C. N. LIN Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, Taipei, TaiwanSearch for more papers by this authorC. Y. KAO, C. Y. KAO Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, Taipei, TaiwanSearch for more papers by this authorC. H. MIAO, C. H. MIAO Seattle Children's Hospital Research Institute Department of Pediatrics, University of Washington, Seattle, WASearch for more papers by this authorN. HAMAGUCHI, N. HAMAGUCHI Department of Biochemistry, Brandeis University, Waltham, MA, USASearch for more papers by this authorH. L. WU, H. L. WU Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University Cardiovascular Research Center, National Cheng Kung University, Tainan, TaiwanSearch for more papers by this authorG. Y. SHI, G. Y. SHI Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University Cardiovascular Research Center, National Cheng Kung University, Tainan, TaiwanSearch for more papers by this authorY. L. LIU, Y. L. LIU Division of Hematology and Center for Cellular and Molecular Therapeutics, The Children's Hospital of PhiladelphiaSearch for more papers by this authorK. A. HIGH, K. A. HIGH Division of Hematology and Center for Cellular and Molecular Therapeutics, The Children's Hospital of Philadelphia Howard Hughes Medical Institute Department of Pediatrics, School of Medicine, University of Pennsylvania, Philadelphia, PA, USASearch for more papers by this authorS. W. LIN, S. W. LIN Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, Taipei, Taiwan Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, TaiwanSearch for more papers by this author First published: 03 August 2010 https://doi.org/10.1111/j.1538-7836.2010.03913.xCitations: 18 Shu-Wha Lin, Department of Clinical Laboratory Sciences and Medical Biotechnology, National Taiwan University Hospital, No.7, Chung-San S. Rd, Taipei 100, Taiwan.Tel.: +886 2 23123456 Ext 66907; fax: +886 2 23817083.E-mail: mtshuwha@ntu.edu.twKatherine A. High, The Children's Hospital of Philadelphia, 3615 Civic Center Blvd., 302 Abramson Research Ctr., Philadelphia, PA 19104-4318, USA.Tel.:+1 215 590 4521; fax: +1 215 590 3660.E-mail: high@email.chop.edu AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Summary. Background: Hemophilia B is an X-linked inherited disorder caused by the lack of functional factor IX (FIX). Currently, treatment of hemophilia B is performed by intravenous infusion of plasma-derived or recombinant FIX. Objective: In an effort to reduce factor usage and cost, we investigated the potential use of FIX variants with enhanced specific clotting activity. Methods: Seven recombinant FIX variants using alanine replacement were generated and assayed for their activity in vitro and in vivo. Results: One variant containing three substitutions (V86A/E277A/R338A, FIX-Triple) exhibited 13-fold higher specific clotting activity and a 10-fold increased affinity for human FVIIIa compared with FIX-wild-type (FIX-WT) and was thus investigated systematically in vivo. Liver-specific FIX-Triple gene expression following hydrodynamic plasmid delivery revealed a 3.5-fold higher specific clotting activity compared with FIX-WT. Human FIX-Triple and FIX-WT knock-in mice were generated and it was confirmed that FIX-Triple has 7-fold higher specific clotting activity than FIX-WT under normal physiological conditions. Protein infusion of FIX-Triple into hemophilia B mice resulted in greater improvement of hemostasis than that achieved with FIX-WT. Moreover, tail-vein administration of a serotype 8 recombinant Adeno-associated vector (AAV8) expressing either FIX-WT or FIX-Triple in hemophilia B mice demonstrated a 7-fold higher specific clotting activity of FIX-Triple than FIX-WT. Conclusions: Our results indicate that the FIX-Triple variant exhibits significantly enhanced clotting activity relative to FIX-WT due to tighter binding to FVIIIa, as demonstrated both in vitro and in vivo. Therefore, FIX-Triple is a good candidate for further evaluation in protein replacement therapy as well as gene-based therapeutic strategies. Introduction Hemophilia B is an X-linked inherited disorder caused by the lack of functional coagulation Factor IX (FIX) and characterized clinically by a bleeding diathesis of variable severity. Currently, treatment of hemophilia B is performed by intravenous infusion of plasma-derived or recombinant FIX. However, the high cost associated with the protein concentrates adds a financial burden to protein replacement therapy. An attractive approach is gene replacement therapy and there has been progress in several clinical trials for genetic treatment of hemophilia [1]. An alternative approach is to develop a FIX molecule with higher specific clotting activity that will reduce the amount of the clotting factor consumed in treatment. Activated FIX (FIXa) itself exhibits extremely low intrinsic activity toward natural and synthetic substrates. Complex formation with FVIIIa overcomes this limitation and results in a marked increase in FX activation by FIXa [2]. Using alanine-scanning mutagenesis in the second epidermal growth factor (EGF-2) domain, we have identified various residues that are critical for FVIIIa binding [3]. A previous report also indicated that the sequence and length of the junction (residues Leu84-Thr87) between both EGF domains contribute to the enhancement of FIXa enzymatic activity when assembled with FVIIIa [4]. The protease domain of FIXa also interacts with FVIIIa. Mutations in the α-helix (residues 330–338) of FIXa may interfere with the interaction between FIXa and FVIIIa [5] and the regions within the catalytic domain (residues 301–303 and 333–339) may confer FVIIIa binding [6]. In addition, we have performed alanine-scanning mutagenesis on the protease domain of FIX and have identified various residues that enhance clotting function or contribute to enhanced clotting activity [7]. The goal of this study is to develop a FIX molecule with higher clotting activity, which may provide an alternative agent to decrease the amount of protein or vector required for replacement therapy or for gene therapy. We hypothesized that an alanine replacement at the residues responsible for FVIIIa binding and for FIX clotting activity might generate a FIX molecule with significantly higher clotting activity, as demonstrated by R338A-FIX [8]. We report a FIX variant with simultaneous alanine replacements at positions 86, 277 and 338, which exhibited increased clotting activities both in vitro and in vivo. Materials and methods In vitro mutagenesis, expression and purification of FIX Site-specific mutagenesis was performed on the human FIX cDNA (accession, M11309, GI:180552) by a PCR-based method (QuikChange; Stratagene, Boston, MA, USA) using primer pairs that replaced the codons at residues 86, 277 or 338 [numbering is relative to the amino-terminal tyrosine (position 1) of the circulating zymogen] with those coding for alanine. The sequences of the primers are: IX86 (5′-GTGAATTA GATGCAACATGTA-3′) paired with IX86T7 (5′-TGTTACATGTTGCATCTA ATTCAC3′) for replacing residue 86; IX277 (5 -GAACTGGACGCACCCTTAGTGC-3′) paired with IX277-2 (5′-CTA AGGGTGCGTCCAGTTCCAG-3′) for replacing residue 277; and IX338 (5′-CATGTCTTGCATCTACAAAG-3′) paired with IX338D (5′-CTTTGTAGATGCAA GACATG-3′) for residue 338. With these primer pairs, seven different FIX cDNAs were generated, fully sequenced, and together with wild-type FIX (FIX-WT) subcloned into pCR3™-Uni (Invitrogen, Carlsbad, CA, USA) for expression in human embryonic kidney (HEK) 293 cells. Of the seven recombinant FIX variants, three were single alanine replacements at residues 86, 277 or 338 (FIX-86, FIX-277 or FIX-338), three were double alanine replacements at 86 and 277 (FIX-86/277), 86 and 338 (FIX-86/338) or 277 and 338 (FIX-277/338), and one was a triple replacement with 86, 277 and 338 simultaneously replaced by alanine (FIX-Triple). Expansion of the selected cell clones into large quantities in serum-free media and purification of the recombinant proteins were performed as described previously [3]. The purified recombinant FIX proteins were either verified by SDS-PAGE followed by staining with Coomassie blue or subjected to immunoblotting. The quantity of the purified FIX was determined by a BCA™ protein assay kit (Pierce, Rockford, IL, USA). Enzyme-linked immunosorbent assay (ELISA) and activated partial thromboplastin time (aPTT) FIX assay The protein quantity of FIX was determined by ELISA using a polyclonal antibody [Gafix-AP160; Enzyme Research Laboratories Inc. (ERL), South Bend, IN, USA] as coating antibody and a peroxidase-conjugated polyclonal antibody (Gafix-HRP; ERL) as the detecting antibody. These antibodies do not cross-react with the endogenous mouse FIX. One-stage aPTT is used as instructed by the manufacturer (Dade®Actin®FSL; Dade Behring Marburg Inc., Marburg, Germany) to measure clotting activity of recombinant proteins. Normal pooled human plasma was prepared from at least 20 healthy individuals. Serial dilutions of the pooled plasma were prepared to derive standard curves for calculation of FIX protein quantity. The specific clotting activity of FIX was calculated by dividing the clotting activity by its protein level. Interaction of FIXa with FX and FIXa with FVIIIa Preparation of FIXa, phosphatidyl-choline-phosphatidyl-serine (75%/25%) and the titration of the activated FIXa by antithrombin III were as described previously [3, 9]. Interaction of FIXa with FX was analyzed by kinetically monitoring the hydrolysis of Spectrozyme FXa by activated FX generated by FIXa in the absence and presence of FVIIIa as described previously [3, 8]. In the incubation mixtures that contained FVIIIa, we modified the activation time of FVIII in this study. In our hands FVIII activity reached highest levels when it was activated by thrombin at 2 min based on our titration, and this result is consistent with the result of Mathur et al. [10]. The concentration of thrombin was 20 nm, and activation of 16 nm FVIII was carried out for 2 min, at which time thrombin was inhibited by 25 nm recombinant hirudin. A Vmax microtiter plate reader (SpectraMax340; MDS, Sunnyvale, CA, USA) was used for all spectrophotometric assays. Data were analyzed according to the referenced equations by nonlinear least squares regression analysis using the Marquardt algorithm [11]. The qualities of the fits were assessed by the criteria described [12]. Initial velocity measurements of FX cleavage by Xase were analyzed by fitting the data to the Henri-Michaelis-Menten equation [13]. Results are derived from two or more separate experiments. Binding of FIXa and FVIIIa was conducted by monitoring the intrinsic Xase activity at limiting concentrations of FVIIIa as described previously [8]. The apparent Kd (EC50) of FIXa and FVIIIa in the Tenase complex was derived from calculations as described previously [3]. Factor VIIIa binding to FIXa studied by surface plasmon resonance (SPR) The kinetics of FVIIIa and FIXa interaction were determined with an SPR-based assay using a BIAcore 3000 biosensor system (BIAcore AB, Uppsala, Sweden). FIX-WT or FIX-Triple was activated by FXIa and were covalently immobilized onto a CM5 sensor chip using amine coupling chemistry [14]. A flow cell to assess non-specific background binding was prepared using FVIIa instead of FIXa. FVIII was activated by thrombin for 2 min and the reaction was inhibited by recombinant hirudin. FVIIIa was injected immediately across the flow cell at various concentrations (1.88–90 nm). Binding (association) of the FVIIIa was monitored in 10 mm HEPES, pH7.4, 150 mm NaCl, 2 mm CaCl2 and 0.005% surfactant P20, at a flow rate of 20 μL min−1 for 3 min at 25 °C. The dissociation of bound FVIIIa was monitored for a 3-min period by replacing the FVIIIa containing buffer with buffer alone. Removal of bound analyte to regenerate the flow cell was accomplished by infusing HEPES-buffered saline containing 3 mm EDTA. After subtraction of the non-specific binding curve, the rate constants for association (kon) and dissociation (koff) were derived by fitting the association/dissociation curves using a one to one Langmuir model on Bioevaluation 3.0 software (Biacore Inc.). Equilibrium dissociation constants (KD) were calculated as koff/kon. Data represent mean ± SD of three experiments. Animal experiments All animal experiments described below were performed following standard procedures. Animals were treated according to the guidelines of the National Taiwan University in Taiwan, the National Institutes of Health, the Seattle Children's Hospital and Regional Medical Center, and the Children's Hospital of Philadelphia. Two different hemophilia B mouse models, FIX knock-out (FIXKO) mice (B6.129P2-F9tm1Dws) [15] and R333Q-hFIX mice (B6.129S6-TgH-F9R333QhF9Dws) [16], were obtained from D. Stafford at the University of North Carolina at Chapel Hill. For hydrodynamic injection and AAV injection, FIXKO mice were used. For protein injection, R333Q-hFIX mice were used. These mice were backcrossed to C57BL/6 background for more than six generations. Only male mice were used in this study. Hydrodynamic injection and measurement of FIX expression The cDNAs coding for FIX-WT and the alanine replacement variants were individually subcloned into pBS-HCRHPI-A [17]. Fifty micrograms of the plasmids were dissolved in 2 mL PBS (phosphate-buffered saline) and hydrodynamically injected through the tail vein into 17–24 g FIXKO mice over a period of 6–8 s. Blood samples were taken from the inferior vena cava 24 h after injection. Plasma samples were prepared and measured for antigen level and clotting activity. Generation of knock-in mice A positive-negative selection-based targeting vector that contains the human FIX cDNA was a generous gift from D. Stafford. The vector was modified by site-directed mutagenesis (Quikchange XL mutagenesis kit; Stratagene, Boston, MA, USA) to generate FIX-Triple or FIX-WT. Targeting strategy was described previously [16]. Identification of the ES cell clones, subsequent injection of the clone into a blastocyst, chimera production and breeding to C57BL/6 background were as described previously [18]. Identification of the knock-in mice and genotyping of the knock-in allele were as described previously [16]. Mice with the knock-in FIX-Triple or FIX-WT alleles were backcrossed into C57BL/6 for two generations to produce hemizygous male mice. Mice were sacrificed at 6–8 weeks old. Plasma samples and liver tissues were collected for FIX protein and mRNA measurement. Human FIX protein levels and clotting activity in the mouse plasma were measured by ELISA and aPTT. The same experiments were carried out on knock-in mice backcrossed to C57BL/6 for more than six generations and the results are similar. Protein infusion into hemophilia B mice R333Q-hFIX mice were injected intravenously with 10 μg per 20 g body weight of recombinant proteins (FIX-WT, FIX-338 and FIX-Triple) to reach a targeted circulating level of 5 μg mL−1. Analyses of Gla content of these proteins were kindly performed by R. Camire at the Children's Hospital of Philadelphia. These proteins were fully carboxylated. At 5, 15, 60, 120 and 180 min after injection, blood samples were collected from the inferior vena cava for FIX measurement. To preferentially detect the injected Thr148 polymorphism isoform of human FIX compared with the Ala148 isoform in the endogenous R333Q protein, we used the anti-human FIX monoclonal antibody A-1 [19], conjugated with horseradish peroxidase as the detecting antibody. Based on this system, we estimate the half-life of recombinant FIX by using the equation published by Tranholm et al. [20]. Gene transfer experiments using a recombinant single-stranded AAV8 vector The expression cassettes of FIX-WT, FIX-338 and FIX-Triple in pBS-HCRHPI-A were subcloned into pAAV-MCS (Stratagene, La Jolla, CA, USA). The resultant plasmids were used to produce serotype 8 AAV vectors by the triple transfection method [21]. Tail vein administration of AAV8-FIX (FIX-WT, FIX-338 and FIX-Triple) vectors into 8-week-old FIXKO male mice was performed with vector doses of 4 × 1012, 4 × 1011 and 8 × 1010 vector genome (vg) kg−1 as described [22]. Plasma samples were collected from the retro-orbital plexus at 2 and 4 weeks after AAV injection. Tail clip assay [23] was performed at 8 weeks after injection with vector dose of 4 × 1011 vg kg−1. Mice were anesthetized and the distal tail (3 mm of diameter) was cut and immediately immersed in 37 °C saline for 10 min. Blood loss was determined by measuring the absorbance of hemoglobin (A575 nm). Mice were sacrificed 8 weeks after AAV injection and blood samples were collected from the inferior vena cava. FIX expression was detected by ELISA. Statistical analysis An unpaired two-tail Student's t-test was used for statistical analysis with P < 0.05 indicating statistical significance. Results Engineered recombinant FIX variants exhibited higher clotting activity than FIX-WT A total of seven alanine-replaced FIX variants were expressed individually in HEK 293 cells. The expression levels of the FIX variants were comparable to that of the FIX-WT at 0.40–0.53 μg per 24 h per 106 cells. After purification, all the recombinant FIX-WT and variant proteins appeared as a single band in SDS-PAGE with similar mobility (Fig. 1A). All recombinant proteins were identified by Western blot analysis (Fig. 1B). The sequence of the first five residues of each recombinant protein was confirmed by amino acid sequence analysis. Biochemical analysis, including antibody recognition of FIX-WT and the seven alanine variants by ELISA and immunoblotting, activation by FXIa and FVIIa·TF (data not shown), and binding to antithrombin in the presence of heparin (Fig. S1), showed no difference between FIX-WT and all variants. The specific clotting activity is shown in Table 1. Recombinant FIX-WT was fully functional and has a specific clotting activity of 188.8 ± 32.4 U mg−1, close to that of plasma-derived FIX (200 U mg−1). All of the variants were also functional and had 1.1–13 times higher clotting activity than FIX-WT. FIX-Triple was the most active, and had 13 times higher clotting activity than recombinant FIX-WT. Figure 1Open in figure viewerPowerPoint SDS-PAGE analysis of purified recombinant FIX. Purified proteins from FIX-wild-type (WT) and FIX variants were subjected to SDS-PAGE. The samples were resolved under reducing conditions. (A) FIX protein (4 μg) bands visualized by Coomassie blue staining. (B) Western blotting performed with a peroxidase-conjugated goat anti-human FIX antibody (0.5 μg per lane). Table 1. Specific clotting activity of purified factor (F) IX Specific clotting activity (U mg−1) FIX-WT 188.8 ± 32.4 FIX-86 214.4 ± 20.0 FIX-277 238.4 ± 30.8 FIX-338 723.2 ± 5.6 FIX-86/277 253.6 ± 13.6 FIX-86/338 340.4 ± 82.4 FIX-277/338 375.2 ± 132.0 FIX-Triple 2641.6 ± 113.6 The concentration of purified recombinant wild-type and alanine-replaced FIX was determined by the BCA™ protein assay kit. The purified FIX protein preparations were diluted to 5 μg mL−1 and the clotting activity was determined by aPTT. Standard curves for clotting activity were derived in parallel experiments from serial dilutions of normal plasma pooled from 20 healthy donors. Samples were diluted in sodium barbital buffer containing 0.1% of BSA and were measured in at least two different dilutions. The experiments were repeated two times. The data were presented as mean ± SD. The increased clotting activity of FIX-Triple correlated with its binding affinity to FVIIIa To further evaluate how these FIX variants function with FX and FVIIIa, kinetic parameters of the activated FIX variants were measured using purified protein. In the absence of FVIIIa, the steady state kinetic constants for each of the recombinant variants are shown in Table 2, with kcat/KM = 186–794 M−1s−1, exhibiting a 106-fold reduction as compared with those in the presence of FVIIIa. As shown in Table 2, the apparent Kd of the three single alanine-replacement variants and one double-replacement variant (FIX-86/277) in complex with FVIIIa approached that of FIX-WT (variants vs. FIX-WT, 1.20–1.95 nm vs. 2.44 nm). Surprisingly, the apparent Kd of the two double-replacement variants (FIX-86/338 and FIX-277/338) were significantly lower than that of FIX-WT. There was a 10-fold difference in apparent Kd between FIX-Triple and FIX-WT in binding FVIIIa. Moreover, the Vmax of FIX-Triple is 3.6-fold higher compared with FIX-WT. These results can account, partially if not fully, for the increased clotting activity of FIX-Triple. Table 2. Kinetic parameters K m (nm) V max (nm Fxa/min) Enzyme*(nm) K cat† (s−1) K cat/Km (m−1 × s−1) K d, app (nm) Without FVIIIa FIX-WT 514.94 ± 78.44 0.16 ± 0.01 10 2.60E-04 505 ND FIX-86 299.94 ± 64.91 0.14 ± 0.01 10 2.36E-04 786 ND FIX-277 965.05 ± 64.28 0.11 ± 0.01 10 1.79E-04 186 ND FIX-338 696.07 ± 61.40 0.26 ± 0.01 10 4.32E-04 620 ND FIX-86/277 304.64 ± 57.81 0.11 ± 0.01 10 1.77E-04 581 ND FIX-86/338 545.45 ± 101.90 0.18 ± 0.02 10 2.95E-04 542 ND FIX-277/338 544.22 ± 79.40 0.12 ± 0.01 10 2.05E-04 377 ND FIX-Triple 329.70 ± 94.55 0.16 ± 0.02 10 2.62E-04 794 ND With 0.4 nm FVIIIa‡ FIX-WT 23.10 ± 4.87 24.33 ± 1.53 0.033 12.40 5.37E±08 2.44 FIX-86 22.20 ± 4.99 12.51 ± 0.83 0.049 4.21 1.90E±08 1.42 FIX-277 23.91 ± 6.18 26.08 ± 2.03 0.056 7.80 3.26E±08 1.20 FIX-338 25.78 ± 4.82 48.06 ± 2.75 0.052 15.36 5.96E±08 1.32 FIX-86/277 31.60 ± 4.03 38.92 ± 2.00 0.039 16.61 5.26E±08 1.95 FIX-86/338 28.62 ± 3.83 31.31 ± 1.33 0.106 4.93 1.72E±08 0.40 FIX-277/338 32.40 ± 7.60 57.58 ± 4.72 0.114 8.40 2.59E±08 0.34 FIX-Triple 80.97 ± 8.00 88.31 ± 3.89 0.144 10.25 1.27E±08 0.19 Interaction of factor (F) IXa with FX and FVIIIa was analyzed by monitoring kinetically the hydrolysis of Spectrozyme FXa as described in 'Materials and methods'. The errors in the fitted constants represent ± 2 SD. Data are calculated from two to three experiments. *The concentration of FIXa-VIIIa complex was derived from experimental conditions and observed Kd. †kcat = Vmax/[enzyme], the units are m FXa, m−1FIXa (or FIXa-FVIIIa), s−1. ‡The reaction was incubated with 0. 4 nm FVIIIa and 0.25 nm FIXa to activate 0–200 nm FX in the presence of 0.5 mm Spectrozyme FXa, 40 μm PCPS and 5 mm Ca2+. ND, not determined. SPR analysis of FVIIIa binding to immobilized FIXa The direct binding of FVIIIa to FIXa was further investigated by SPR analysis. The signal of the 1900 resonance unit (RU) of FVIIa was immobilized on the reference channel. Approximately 1630 RU of FIXa-WT or 1590 RU of FIXa-Triple was immobilized. No binding response of FVIII, thrombin or hirudin to immobilized IXa could be detected. The response curve of the FIXa-Triple immobilized chip was different from the FIXa-WT chip when we injected various concentrations of FVIIIa into the flow cell (Fig. 2). Association (kon) and dissociation (koff) rate constants between FVIIIa and FIXa were investigated by passing different concentrations of FVIIIa over the FIXa-coated channel. For the FIXa-WT chip, we injected five different concentrations of FVIIIa (from 7.5–90 nm). There was no detectable response when we injected 3.75 nm of FVIIIa into the FIXa-WT chip. For the FIXa-Triple chip, five concentrations (from 1.88–30 nm) of FVIIIa were used. The kon was (7.6 ± 2.3) × 104 m−1 s−1 for FIX-WT and (3.6 ± 1.2) × 105 m−1 s−1 for FIX-Triple. The koff was (2.6 ± 0.7) × 10−3 s−1 for FIX-WT and (1.2 ± 0.4) × 10−3 s−1 for FIX-Triple. The equilibrium dissociation constant KD is (35 ± 0.1) × 10−9 m for IXa-WT and (3.5 ± 0.9) × 10−9 m for FIXa-Triple. The affinity of FVIIIa to IXa-Triple is 10-fold higher than that of FIX-WT. Figure 2Open in figure viewerPowerPoint SPR analysis of factor (F) VIIIa binding to immobilized FIXa. IXa-wild-type (WT) or IXa-Triple was immobilized onto a CM5 sensorchip as described under 'Materials and methods'. FVIIa was immobilized on a reference channel to assess non-specific background binding. Factor VIIIa at various concentrations in HEPES buffer containing 2 mm CaCl2 and 0.005% P20 was injected across the flow cell using 3-min association and 3-min dissociation times at 25 °C at a flow rate of 20 μL min−1. (A) Response curves from bottom to top, representing FVIIIa at concentrations of 7.5, 15, 30, 60 and 90 nm binding to FIXa-WT immobilized chip. (B) Response curves from bottom to top, representing FVIIIa at concentrations of 1.88, 3.75, 7.5, 15 and 30 nm binding to FIXa-Triple immobilized chip. The data shown in the figure are representative runs for each experiment. All experiments were performed in triplicate. Hydrodynamic-based expression of human FIX in FIXKO mice To characterize FIX-Triple and the other variants in vivo, the expression plasmids were injected hydrodynamically into FIXKO mice. As shown in Fig. 3, mice (n = 4/group) treated with FIX-WT plasmid DNA and all alanine replacement variants, except for FIX-Triple, expressed 0.72–1.17 μg mL−1, with clotting activities of 0.09–0.53 U mL−1 in plasma. The FIX-Triple DNA-treated mice expressed 10 times higher FIX clotting activities than did those treated with FIX-WT DNA. Approximately 3.5-fold higher specific clotting activity was observed in mice injected with FIX-Triple DNA than in those injected with FIX-WT, and a three times greater FIX protein level was observed. This increase in transgene expression levels was not observed when FIX-WT and FIX-Triple were expressed in knock-in mice or AAV-treated mice, which both have a long-term expression system and better characterize the stable transgene levels. Figure 3Open in figure viewerPowerPoint Human factor (F) IX levels in FIX knock-out mice 24 h after hydrodynamic injection with pBS-HCRHPI-FIX-A. The cDNAs coding for FIX-wild-type (WT) and the alanine replacement variants were individually subcloned into pBS-HCRHPI-A. Male FIX knock-out mice were subjected to hydrodynamic shock by tail vein injection of 2 mL of 50 μg DNA in 6–8 s (n = 4 per group). The mice were recovered and sacrificed 24 h after injection for collection of blood plasma for clotting activity and FIX protein level determination. Data represent mean ± SD. Each bar represents the data of an individual mouse. The black bar indicates the FIX expression level and the scale is indicated in lower x-axis. The blank bar including the black area indicates the clotting activity of FIX and the scale is indicated in the upper x-axis. *Specific clotting activity. Generation of FIX-knock-in mice To evaluate an endogenous and steady-state production of FIX-Triple, we generated knock-in mice, in an identical fashion to R333Q-hFIX mice, carrying either human FIX-WT (FIX