Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Dec 1.
Published in final edited form as: J Thromb Haemost. 2016 Nov 19;14(12):2509–2523. doi: 10.1111/jth.13538

S2′-subsite variations between human and mouse enzymes (plasmin, FXIa, kallikrein) elucidate differences in inhibition by TFPI-2 KD1WT-VT, KD1L17R-KT and aprotinin

K Vadivel *, Y Kumar *, G I Ogueli *, S M Ponnuraj *, P Wongkongkathep †, J A Loo †,‡, M S Bajaj §, S P Bajaj *,‡
PMCID: PMC5504414  NIHMSID: NIHMS872810  PMID: 27797450

Summary

Background

Using TFPI-2 Kunitz domain1 (KD1), we obtained a bifunctional antifibrinolytic molecule (KD1L17R-KT) with C-terminal lysine (kringle domain binding) and P2′-residue arginine (improved specificity towards plasmin). KD1L17R-KT strongly inhibited human plasmin (hPm), with no inhibition of human kallikrein (hKLK) or factor XIa (hXIa). Furthermore, KD1L17R-KT reduced blood loss comparable to aprotinin in a mouse liver-laceration model of organ hemorrhage. However, effectiveness of these antifibrinolytic agents in a model of hemorrhage mimicking extremity trauma and their inhibition efficiencies for mouse enzymes (mPm, mKLK or mXIa) remain to be determined.

Objective

Determine potential differences in inhibition constants of various antifibrinolytic agents against mouse and human enzymes and test their effectiveness in a modified mouse tail-amputation hemorrhage model.

Methods/Results

Unexpectedly, mXIa was inhibited with ~17-fold increased affinity by aprotinin (Ki ~20 nM) and with measurable affinity for KD1L17R-KT (Ki ~3 µM); in contrast, KD1WT-VT inhibited hXIa or mXIa with similar affinity. Compared to hPm, mPm had ~3-fold reduced affinity, whereas species specificity for hKLK and mKLK were comparable for each inhibitor. S2′-subsite variations largely accounted for the observed differences. KD1L17R-KT and aprotinin were more effective than KD1WT-VT or tranexamic acid in inhibiting tPA-induced mouse plasma clot lysis. Further, KD1L17R-KT was more effective than KD1WT-VT and was comparable to aprotinin and tranexamic acid in reducing blood loss and D-dimer levels in the mouse tail-amputation model.

Conclusions

Inhibitor potencies differ between antifibrinolytic agents against human and mouse enzymes. KD1L17R-KT is effective in reducing blood loss in a tail-amputation model that mimics extremity injury in trauma.

Keywords: Antifibrinolytic agents, Aprotinin, Fibrinolysis, TFPI-2, Tranexamic acid

Introduction

Trauma or major surgery results in severe tissue damage and rupture of both the small and large vessels in the injured area. Tissue damage leads to local platelet adhesion and aggregation with formation of the primary hemostatic plug. The platelet plug is then strengthened and stabilized by a fibrin network generated through extrinsic and intrinsic coagulation. As the damaged vessels heal, the fibrin clot is lysed to ensure normal blood flow. The breakdown of the clot, fibrinolysis, is due to plasmin (Pm) generated from plasminogen by tissue plasminogen activator (tPA) released from the damaged endothelium [1–3]. However, severe trauma and major surgery frequently evoke hyperfibrinolysis resulting in massive hemorrhage necessitating the use of increased blood products [4–6], which are associated with a high risk of adverse clinical outcomes. These include morbidity and mortality related to transfusion-associated acute lung injury [7,8] as well as transfusion-related immunomodulation and transfusion-associated circulatory overload [8].

Antifibrinolytic agents are useful in preventing blood loss during trauma and major surgical procedures such as cardiopulmonary bypass surgery as well as in reducing the risk of surgical revision due to post-operative bleeding [9,10]. Approximately 4–6% of these patients (~25,000 patients/year in the Unites States) need surgical revision due to bleeding following cardiac surgery [10]. Antifibrinolytic agents when used prophylactically can significantly reduce blood loss and the need for surgical revision as well as the intraoperative and postoperative requirement for allogeneic blood transfusions [9–11]. After the withdrawal of aprotinin (Trasylol®, bovine pancreatic trypsin inhibitor, BPTI) from the world market in year 2008 following the BART study [12], tranexamic acid (TXA) and ε-amino caproic acid (EACA), are the two antifibirinolytics currently in use. However, emerging reports of observational clinical trials (Table S1) reveal that upsurge in the use of TXA [10,12–20] and EACA [10,12,18–22] in patients undergoing cardiac surgery is associated with seizures (1.3–10% TXA; 3.3% EACA) and renal dysfunction (1.5–20% TXA; 12.9–40% EACA) leading to an increase in hospital stay and mortality rate (2.5–7.5% TXA; 2.4–4.0% EACA). Although not in every study [23–27], seizures have also been observed with TXA or EACA in mice [28,29], rats [30,31], and cats [32,33], whereas renal injury has been observed with aprotinin in mice [34] and rats [35]. These observations highlight a need for development of an alternate antifibirnolytic agent that prevents blood loss without the serious adverse effects during surgery and trauma.

TXA and EACA bind to the kringle-lysine binding sites (LBS) of tPA and plasminogen and thereby prevent their binding to the exposed C-terminal lysine residues of fibrin; as a result activation of plasminogen to Pm and fibrinolysis at the clot site is impeded [36,37]. Since currently available active site inhibitors of plasmin or the LBS inhibitors of fibrinolysis are associated with adverse effects, a concerted effort is directed towards development of new safer inhibitors of fibrinolysis [38–47]. In this context, we used Kunitz domain1 (KD1) of human tissue factor pathway inhibitor type-2 (TFPI-2) to obtain a specific inhibitor (KD1L17R-KT) of human Pm (hPm) by targeting the S2′-pocket. Further, KD1L17R-KT also binds to the LBS of tPA or plasminogen with ~300 nM Kd [29]. Unlike KD1 wild-type molecule with C-terminal valine (KD1WT-VT) or aprotinin [34], KD1L17R-KT (or KD1L17R-VT) does not inhibit human factor XIa (hXIa) or plasma kallikrein (hKLK). Further, similar to aprotinin and the lysine analog TXA or EACA, KD1L17R-KT is effective in reducing blood loss in the mouse liver-laceration model [29]. Importantly, it does not appear to cause seizures in mice [29] and renal injury in mice [34] or rats [35]. Although, KD1L17R-KT is effective in reducing blood loss in the liver-laceration model of organ trauma, its antifibrinolytic potential in hemorrhage from extremity trauma has not been evaluated. Moreover, inhibition by KD1WT-VT/KD1L17R-KT or aprotinin of mouse enzymes (mPm, mIXa, mKLK), the animal used for studies, is not known.

Here, we examined reduction of blood loss by KD1L17R-KT in a modified mouse tail-amputation model, which resembles hemorrhage from extremity trauma concerning tissue, bone and large vessel damage. In our model, tail is amputated at 4 cm vs 0.5 cm in previous reports for testing systemic and topical antifibrinolytic agents [48,49]. Since animal studies are performed in mice, we examined specificity of human KD1WT-VT, KD1L17R-KT or bovine aprotinin for mouse enzymes (mPm, mXIa and mKLK). Further, aprotinin strongly inhibits bovine trypsin, a coagulation-type protease; thus, we examined inhibition of trypsin by KD1L17R-KT. We present these findings here and provide rationale for the observed differences between human and mouse enzymes by structural models.

Materials and methods

Materials

Citrated mouse plasma was from Innovative Research (Novi, MI). Aprotinin was from ZymoGenetics (Seattle, WA) and tissue plasminogen activator (tPA) was from Genentech (South San Francisco, CA). Pm substrate S-2251 (H-D-Val-Leu-Lys-p-nitroanilide), and KLK and Factor XIa substrate S-2366 (pyroGlu-Pro-Arg-p-nitroanilide) were from Diapharma Inc. Asserachrom D-Di ELISA kit was from Diagnostica Stago, Inc (Parsippany, NJ). Human α-thrombin (IIa), specific activity ~4493 U/mg (fibrinogen clotting assay), and mPm were from Haematologic Technologies Inc. Human alpha-XIIa was from Enzyme Research Laboratories. mKLK and mXI were from R&D Systems Inc. mXI (100 µg/ml) was activated at 37 °C by factor alpha-XIIa using a 1:20 molar ratio in 50 mM Tris-HCl, pH 7.5 containing 100 mM NaCl (TBS) and 2 mM CaCl2. One µL aliquots were withdrawn at different times, diluted to 100 µL in TBS containing 0.1 mg/ml bovine serum albumin (TBS/BSA, pH 7.5) and 2 mM CaCl2. The formation of mXIa was measured by hydrolysis of synthetic chromogenic substrate S2366 (330 µM). Activation was complete in 4 hrs as measured by mXIa hydrolytic activity and SDS-PAGE [50]. Under reduced conditions, mXI (2 µg) migrated as a single band with MW of ~80,000 Da and mXIa (2 µg) migrated as two bands with MWs of ~45,000 Da and ~37,000 Da. The mXIa was kept on ice and used within 24 hrs.

Expression, purification and characterization of KD1WT-VT and KD1L17R-KT

The procedures to obtain KD1WT-VT and KD1L17R-KT have been described earlier [29,34,51,52]. Briefly, the 73 amino acid KD1WT-VT and KD1L17R-VT were over-expressed as amino-terminal His6-tagged fusion proteins in E. coli strain BL21(DE3) pLysS using the T7 promoter system. To obtain KD1L17R with a C-terminal lysine (KD1L17R-KT), KD1L17R-VT was incubated with IIa at a 1:50 enzyme/substrate ratio for 24 hrs at 37°C [29]. The His6-tag free proteins were purified using Superdex 200 gel filtration chromatography equilibrated with TBS as outlined [29,34,51,52]. The purified KD1WT-VT and KD1L17R-KT were checked for purity (SDS-PAGE) and characterized using N-terminal sequencing and mass spectrometry prior to performing biochemical and animal studies.

Protease inhibition assay

Reactions were carried out in TBS/BSA/2 mM Ca2+, pH 7.5. Each enzyme (mPm, mKLK and mXIa) was incubated with various concentrations of KD1L17R-KT, KD1WT-VT, aprotinin or trypsin for 1 hr at RT in a 96-well microtitration plate (total volume 100 µL). Synthetic substrate (5 µL) specific for each enzyme was then added (final concentration, 1 KM) and residual amidolytic activity was measured in a Molecular Devices Vmax kinetic microplate reader. The inhibition constants, Ki∗, were determined using the nonlinear regression data analysis program GraFit. Data were analyzed with an equation for a tight-binding inhibitor (Equation 1) where vi and v0 are the inhibited and uninhibited rates, respectively, and [I]0 and [E]0 are the total concentrations of inhibitor and enzyme, respectively [53,54].

vi=v0((Ki∗+[I]0+[E]0)2−4[I]0[E]0)12−(Ki∗+[I]0−[E]0))2[E]0 Equation 1

Ki values were obtained by correcting for the effect of substrate according to Beith [53], using Equation 2, where [S] is substrate concentration and KM is specific for each enzyme.

Ki=Ki∗/(1+[S]/KM) Equation 2

Fibrinolysis (clot lysis) assay

Ten µL of each test compound or saline control was added to 240 µL of mouse plasma; 225 µL of this mixture was then added to 25 µL IIa and tPA in TBS/BSA containing 25 mM CaCl2. The final concentration of IIa was 0.15 µg/ml, and of tPA was 1 µg/ml. Clot formation and lysis were monitored (OD405) with a Molecular Devices microplate reader (SPECTRAmax 190) as described earlier [29,34].

Mouse tail-amputation model

A protocol using the tail-amputation mouse model was approved by the UCLA Chancellor’s animal research committee. In this model, tail is amputated at 4 cm from the tip that involves cutting the bone, which we hypothesize will increase local activation of the fibrinolytic system due to enhanced vascular and tissue trauma. Five groups of C57BL/6 mice (~20 gm in weight) of both sexes were used for saline control, positive controls (aprotinin and TXA) and experimental drugs (KD1L17R-KT and KD1WT-VT). Animals were fed standard rodent chow and water, and housed in a vivarium room maintained at 20–24 °C and 12-h light-dark cycles. Dosage of aprotinin was comparable to that used in humans, i.e. 4 µg/g adjusted for mouse weight [55]. A calculated blood level (based on 10 ml volume of distribution) achieved by this dose was ~8 µg/ml (~1 µM). Since KD1WT-VT and KD1L17R-KT are homologs of aprotinin, a similar dose of each KD1 variant was used.

Variable concentrations of TXA have been used in different animal models [23–27]. We used 10 µg/g adjusted for mouse weight (calculated TXA blood level ~20 µg/ml or ~125 µM) based upon the dose used in one regimen to prevent blood loss in a major surgical procedure [55]. Such dose has also been reported to be adequate to prevent systemic fibrinolysis and local fibrinolytic bleedings [56].

All animals were run in a single big randomized experiment and all compounds including saline were tested during each batch. Anesthesia was induced using 5% isoflurane via nosecone. Following induction of anesthesia, mice were placed on a heating pad to maintain rectal temperature at 37 °C. Drugs were injected via tail vein in 100 µL of sterile, balanced salt solution. Care was taken to avoid extravasation and to avoid injecting the drug close to the site of amputation so as to minimize the potential local effect of any extravasated drug. After allowing two minutes for circulation, tail was amputated at 4 cm from the tip using a sharp, sterile, #11 scalpel blade. Anesthesia was maintained with 2% isoflurane via nosecone during the entire experiment. Blood was allowed to flow freely into preweighed eppendorf tubes. Care was taken not to disturb the bleeding tip. Total time of observation was 1 hour, based on the literature and the preliminary experiments during which normal controls demonstrated any significant recurrence of bleeding, potentially from alternate constriction and dilation of the central artery [57].

Bleeding time varied between 8 to 13 min with minimal recurrent bleeding except in the saline group that lasted occasionally up to 30 min. The plasma half-life of TXA in humans, rats and dogs is ~120 min [58]. The half-life of the two KD1 variant homologs (aprotinin and Ecallantide) is also ~120 min [59,60] in humans, whereas of aprotinin in mice, rats or dogs is ~70 min [61]. The half-life of each KD1 variant could also be similar to aprotinin/Ecallantide. Based upon these data, adequate concentration of each drug is presumed to exist up to the cessation time of bleeding. No deaths were observed in any of the five groups. At the end of the experiment, citrated blood was obtained from the lateral saphenous vein for D-dimer assay by venipuncture. After removing the RBC, mouse plasma was obtained by centrifugation for 15 min at 1500g, 4 °C. The data were combined after the completion of all experiments and analyzed. Since all animals were similar in weight and treated alike, potential temporal effects were minimized and no correction was applied. Primary outcome was total blood loss and plasma D-dimer levels.

Statistical methods

We compared mean total blood loss across all five treatment groups using factorial one-way analysis of variance (ANOVA). The p-values for comparing any two means were computed using post-hoc tests and adjusted for multiple comparisons using Tukey’s adjustment. For the D-dimer data, Levene’s F-test revealed that the homogeneity of variance was not met. As such, the Welch’s F-test was used and Games-Howell post-hoc procedure was conducted to determine which pairs of the five mean D-dimer levels differ significantly. All statistical analysis was performed using SPSS V23 (IBM Corp., Armonk, NY, USA).

Molecular modeling

The MODELLER program [62] was employed to model the protease domains of mPm, mXIa and mKLK using PDB entries 1BUI (hPm), 1ZJD (hXIa) and 2ANY (hKLK) as templates [63]. The modeled structures were energy minimized with backbone restraints using CHARMM [64] and were free of Ramachandran outliers. The complexes of KD1WT-VT, KD1L17R-KT and BPTI with mPm, mXIa and mKLK were modeled as described earlier [34,51,52] and energy minimized with backbone restraints using CHARMM prior to analysis. The trypsin-BPTI complex (2PTC) and trypsin-KD1WT-VT complex (1ZR0) were used for structural analysis. The electrostatic potentials were calculated using ABPS module implemented in PyMOL [65,66].

Results and Discussion

Molecular characterization of KD1WT-VT and KD1L17R-KT molecules

The KD1WT-VT preparation is homogeneous and has C-terminal Val73 as determined by SDS-PAGE, mass spectrometry and N-terminal sequence analysis (Fig. 1A). Isolated KD1L17R-KT has C-terminal Lys72 (~92%) with minor proteolysis at the N-terminus between residues Ala3↓Gln4, Gln4↓Glu5, and Thr7↓Gly8 (Fig. 1B). Since Kunitz domain starts at residue 10 (number 1 in BPTI, Fig. 1), the inhibitory domain of KD1L17R-KT is intact and only ~8% of the preparation has C-terminal Val73 that is devoid of binding to the kringle domains of tPA and plasminogen. These data indicate that KD1L17R-KT preparation is suitable for use as a dual reactivity antifibrinolytic agent, which can inhibit the active site of Pm as well as bind to the kringle domains of tPA and plasminogen.

Fig. 1. SDS-PAGE, MALDI-TOF-ESI mass spectrometry and amino acid sequences of KD1WT-VT and KD1L17R-KT.

Fig. 1

Open in a new tab

Mass spectrometry data were obtained using an Applied Biosystems Voyager DE-STR. (A) KD1WT-VT. The mass was 8843 Da and the N-terminal sequence was GSHMDAA at ~80 pmol level. Combined mass spectrometry and sequence data reveal EKVPKV sequence at the C-terminus of KD1WT-VT. For comparison, amino acid sequence alignment of KD1WT-VT with BPTI is presented to indicate that the residue Asn10 in KD1WT-VT represents residue Arg1 in BPTI. Thus, KD1WT-VT has nine extra residues at the N-terminus excluding GSHM. The P1 residue Arg (15 in BPTI numbering, residue 24 in KD1WT-VT) and P2′ residue Arg (17 in BPTI and 26 in KD1WT-VT) are indicated. Inset, SDS-PAGE (15% acrylamide). Lane 1, reduced EZ-Run protein markers (Fisher Bioreagents); lane 2, reduced KD1WT-VT (5 µg). (B) KD1L17R-KT. Mass spectrometry revealed three minor (peaks 1–3, ~8% each) and one major molecular species (peak 4, ~75%). The N-terminal sequence was DAAQEPT at the ~90 pmol level and the other N-terminal sequences (Cycle 1, ~8 pmol; cycle 2, ~6 pmol and cycle 3, ~5 pmol) were QEP, EPT and GNN. Combined mass spectrometry and sequence data revealed that the three molecular species (peaks 1, 3 and 4) each has C-terminal Lys72 and constitutes ~92% of the total mass; only peak 2 (~8%) has C-terminal Val73. The difference in molecular mass results from proteolysis by IIa or a contaminating protease. For comparison, the sequence alignment of KD1L17R-KT with BPTI is given. Inset, SDS-PAGE (15% acrylamide). Lane 1, reduced EZ-Run protein markers; lane 2, reduced KD1L17R-KT (5 µg).

Inhibitory properties of KD1WT-VT, KD1L17R-KT and aprotinin (BPTI) against mPm, mXIa, mKLK and bovine trypsin

KD1L17R-KT inhibits hPm effectively without inhibiting hXIa and hKLK [34]. Since potency of KD1L17R-KT is tested in a mouse bleeding model, we determined its inhibitory proficiency against mPm, mXIa and mKLK, and compared such data obtained with KD1WT-VT and aprotinin. Further, S2′-pocket of trypsin lacks acidic residues, which is a hallmark of Pm [34]; however, aprotinin (P2′ Arg17) inhibits trypsin with pM affinity [67]. We sought an explanation for this observation and examined whether KD1L17R-KT also inhibits trypsin.

Inhibition of mPm by KD1WT-VT, KD1L17R-KT and aprotinin

mPm was inhibited by KD1WT-VT with ~18 nM Ki, by KDL17R-KT with ~4 nM Ki, and by aprotinin with ~1.5 nM Ki (Fig. 2A and Table 1). Thus, as compared to hPm, affinity of each inhibitor is reduced by ~3 to 4-fold for mPm. To understand these uniform differences in affinity of each inhibitor for mPm versus hPm, we obtained molecular models of the protease domains of hPm and mPm complexed with each inhibitor. Although, protease domains of hPm and mPm share ~83% sequence identity [68,69], the notable difference between the S2′-pocket of these two proteases is the presence of Met39 in hPm versus Gln39 in mPm. This is shown in Fig. 3A and Fig. 3B. In hPm, Leu17 of KD1WT-VT or Arg17 side chain hydrocarbons of KD1L17R-KT or BPTI are involved in hydrophobic contacts with Met39; however, such interactions are not possible in mPm due to the presence of Gln39 instead of Met39 (Fig. 3A and 3B). Further, Glu73 and Glu143 of hPm/mPm make salt-bridges with P2′ Arg17 of KD1L17R-KT or BPTI; such interactions are not possible for KD1WT-VT. As compared to KD1WT-VT, these additional interactions provide a reasonable explanation for the higher affinity of KD1L17R-KT or BPTI for hPm/mPm.

Fig. 2. Determination of equilibrium inhibition constants (Ki) of KD1WT-VT, KD1L17R-KT, and aprotinin with mPm, mXIa, mKLK and bovine trypsin.

Fig. 2

Open in a new tab

(A) Percent activity of 3 nM mPm remaining in the presence of various concentrations of KD1L17R-KT (●), KD1WT-VT (ο) and aprotinin (△). (B) Percent activity of 2 nM bovine trypsin remaining in the presence of various concentrations of KD1L17R-KT (●), KD1WT-VT (ο), and aprotinin (△). (C) Percent activity of 1 nM mXIa remaining in the presence of various concentrations of KD1L17R-KT (●), KD1WT-VT (ο), and aprotinin (△). (D) Percent activity of 1 nM mKLK remaining in the presence of various concentrations of KD1L17R-KT (●), KD1WT-VT (ο), and aprotinin (△).

Table 1.

Ki values for the inhibition of mPm, mXIa and mKLK

Enzyme Ki Values (nM)a
KD1WT-VT KD1L17R-KT BPTIc
hPmb 6 ± 0.5 0.9 ± 0.1 0.5 ± 0.1
mPm 18 ± 2 4 ± 0.5 1.5 ± 0.3
hXIab 18 ± 2 NM 346 ± 25
mXIa 21 ± 3 3400 ± 300 20 ± 2
hKLKb 25 ± 3 NM 18 ± 2
mKLK 20 ± 3 NM 10 ± 2
Bovine Trypsin 17 ± 2 7 ± 1 0.07 ± 0.01
Open in a new tab
a

Ki values represent an average of three independent measurements and were determined using the tight binding equilibrium equation [54]. Ki∗ values were corrected [53] for the effect of synthetic substrate used for measuring the active concentration of each enzyme employed. Concentration of mPm used was 3 nM, of bovine trypsin was 2 nM and of mXIa or mKLK was 1 nM.

b

The Ki values for hPm, hXIa and hKLK for each inhibitor are from reference 34. NM, not measurable upto 10 µM;

c

BPTI, Bovine pancreatic trypsin inhibitor (aprotinin).

Fig. 3. Interactions between the protease domains each of hPm, mPm and bovine trypsin with KD1WT-VT, KD1L17R-KT and BPTI.

Fig. 3

Open in a new tab

(A) Interaction between the P2′ residue of KD1WT-VT, KD1L17R-KT or BPTI with S2′-pocket of hPm. Left, hydrophobic interactions between Leu17 of KD1WT-VT with Met39 and Glu143 hydrocarbon side chain of hPm. Middle, salt-bridge interactions of Arg17 of KD1L17R-KT with Glu73 and Glu143 as well as hydrophobic contacts with Met39 of hPm. Right, similar to KD1L17R-KT, interactions of Arg17 of BPTI with Glu73/Glu143 and Met39 of hPm. (B) Interaction between P2′ residue of KD1WT-VT, KD1L17R-KT or BPTI with S2′-pocket of mPm. Left, hydrophobic contact between Leu17 of KD1WT-VT and Glu143 side chain of mPm. Middle, interactions between Arg17 of KD1L17R-KT involving salt-bridge with Glu73 and Glu143 of mPm. Right, similar to KD1L17R-KT, interactions of Arg17 of BPTI with Glu73/Glu143 of mPm. (C) Interaction between the P2′ residue of KD1WT-VT, KD1L17R-KT or BPTI with S2′-pocket of trypsin and additional interactions of trypsin with BPTI. Left, hydrophobic interactions between Leu17 of KD1WT-VT with Tyr39 and Tyr151 of trypsin. Middle, interactions of Arg17 of KD1L17R-KT or BPTI involving π-interactions with Tyr151 and hydrogen bond with His40 main chain of trypsin. Right, additional hydrogen bond interactions between Arg39 of BPTI with Thr98 and Asn97 of trypsin. hPm or mPm 90-loop (green) is short and has no interactions with BPTI. KD1WT-VT is shown in dark green, KD1L17R-KT in sand, BPTI in cyan, hPm in dark yellow, mPm in light blue and trypsin in wheat. The hydrogen bonds are shown in black dashed lines, π-interaction in magenta dashed line and the hydrophobic interactions by arcs with spokes radiating toward the contact atoms.

Inhibition of bovine trypsin by KD1WT-VT, KD1L17R-KT and aprotinin

Trypsin was inhibited by KD1WT-VT with ~17 nM Ki, by KDL17R-KT with ~7 nM Ki, and by aprotinin with ~0.07 nM Ki (Fig. 2B and Table 1). These data are in agreement with a Ki of ~13 nM for KD1WT-VT [51] and a very strong inhibition of trypsin by aprotinin [70]. Similar to the coagulation proteases [34], trypsin has a hydrophobic S2′-pocket that is devoid of acidic residues [52,70]. However, multiple sequence alignment of trypsin with coagulation proteases and plasmin reveal that Tyr39 and Tyr151 are unique to trypsin and are situated at the entrance of the S2′-pocket (Fig. 3C). Further, examination of the X-ray structures of trypsin in complex with KD1WT-VT [52] or aprotinin [70] reveal overall similar interactions except encompassing the S2′-pocket and the 90-loop. As shown in Fig. 3C, the P2′ Leu17 of KD1WT-VT makes hydrophobic contacts with Tyr39 and Tyr151 of trypsin [52]. The P2′ Arg17 of aprotinin makes a hydrogen bond with the main chain carbonyl of His40 as well as π-interactions with the phenyl ring of Tyr151 [70]. Moreover, as experimentally observed with aprotinin [70], P2′ Arg17 of KD1L17R-KT makes similar interactions with His40/Tyr151 of trypsin (Fig. 3C).

Additionally, Arg39 of BPTI is involved in hydrogen bond interactions with the main chain carbonyls of Asn97 and Thr98 in the 90-loop of trypsin [70]. As compared to trypsin, the 90-loop in hPm/mPm has four-residue deletion, which prevents such interactions between aprotinin and Pm (Fig. 3C). These differences could be a reason for stronger inhibition of trypsin versus hPm/mPm by aprotinin (Table 2). Interestingly, residue 39 is Glu in KD1WT-VT or KD1L17R-KT, which precludes its interactions involving the 90-loop of trypsin (or hPm/mPm). Thus, existence of interactions involving the 90-loop could explain the difference in affinity of aprotinin versus KD1WTVT or KD1L17R-KT for trypsin.

Inhibition of mXIa by KD1WT-VT, KD1L17R-KT and aprotinin

KD1WT-VT inhibits mXIa and hXIa with similar affinity with Ki ~20 nM (Fig. 2C and Table 1). Although protease domains of hXIa and mXIa share ~78% sequence identity [71,72], the 39-loop in mXIa has a 3-residue deletion including Arg39D present in the S2′-pocket of hXIa. However, P2′ Leu17 of KD1WT-VT is suitably placed in the S2′-pocket of mXIa (Fig. 4A) or hXIa [34] without interference from Arg39D. In contrast, KD1L17R-KT inhibits mXIa with Ki ~3 µM as compared to no inhibition observed upto 10 µM for hXIa (Fig. 2C and Table 1). A plausible explanation for this observation is the unfavorable interaction between Arg39D of hXIa and P2′ Arg17 of KD1L17R-KT and the absence of Arg39D in mXIa (Fig. 4B). Consequently, KD1WT-VT illustrates no difference in inhibition mXIa or hXIa, whereas KD1L17R-KT inhibits mXIa weakly.

Fig. 4. Modeled interactions between the protease domains each of mXIa and hXIa with KD1WT-VT, KD1L17R-KT and BPTI.

Fig. 4

Open in a new tab

(A) Interaction between P2′ Leu17 of KD1WT-VT with S2′-pocket of mXIa. Similar to hXIa [34], Leu17 of KD1WT-VT is situated in a vastly hydrophobic S2′-pocket environment of mXIa. Pymol generated map of the electrostatic potential at the solvent-accessible surface of the mXIa catalytic domain near residue Leu17 of KD1WT-VT is shown. The mXIa residues within 6 A radius of Leu17 of KD1WT-VT were selected for display. Blue represents positive, red represents negative and white represents hydrophobic surface potential. (B) P2′ Arg17 of BPTI or KD1L17R-KT and S2′-pockets of hXIa and mXIa. Left, the longer 39-loop and the presence of Arg39D in hXIa disfavor the interaction with Arg17 of BPTI or KD1L17R-KT. Right, the short 39-loop and the absence of a basic residue in mXIa relieves the repulsive interaction with Arg17 of BPTI or KD1L17R-KT. The capital letter followed by the residue number, e.g., R39D or G39A indicates insertion. (C) Additional interactions of hXIa (or mXIa) with BPTI and their absence in KD1WT-VT or KD1L17R-KT. Left, Arg39 of BPTI makes salt-bridge with Glu98 and Glu217 of hXIa or mXIa. Right, the presence of Glu39 in KD1WT-VT or KD1L17RKT disfavor these interactions.

Strikingly, aprotinin inhibits mXIa with Ki ~20 nM as compared to Ki ~350 nM for hXIa (Fig. 2C and Table 1). Similar to KD1L17R-KT, P2′ Arg17 of aprotinin is unfavorable to interact with Arg39D of hXIa; however, absence of Arg39D in mXIa abrogates this unfavorable interaction. Further, Arg39 of aprotinin interacts via salt-bridge with Glu98 and Glu217 of hXIa/mXIa, and these interactions are not possible for KD1WT-VT or KD1L17R-KT (Fig. 4C). Thus, as compared to hXIa, a ~17-fold increase in inhibition potential of aprotinin for mXIa primarily stems from the absence of unfavorable interactions between P2′ Arg17 and the S2′-pocket.

Inhibition of mKLK by KD1WT-VT, KD1L17R-KT and aprotinin

No significant differences were observed for inhibition of mKLK or hKLK by KD1WT-VT, KD1L17R-KT and aprotinin (Fig. 2D and Table 1). Although the protease domain of hKLK has Arg39 [73] instead of Thr in mKLK [74], P2′ Leu17 of KD1WT-VT is optimally placed in the S2′-pocket of either enzyme (Fig. 5A). The P2′ Arg17 of KD1L17R-KT or aprotinin is unfavorable to interact with Arg39 of hKLK (Fig. 5B), whereas it is not involved in any interaction with the S2′-pocket of mKLK. Note that unlike mXIa, the 39-loops of mKLK and hKLK have the same number of amino acids. Further, Arg39 and Tyr10 of aprotinin interact with Glu217 of hKLK/mKLK, and such interactions are not possible for KD1L17R-KT (or KD1WT-VT) (Fig. 5C). Thus, the presence of additional interactions allow aprotinin to inhibit hKLK/mKLK and their absence make KD1L17R-KT a non-inhibitor of hKLK/mKLK.

Fig. 5. Modeled interactions between the protease domains each of mKLK and hKLK with KD1WT-VT, KD1L17R-KT and BPTI.

Fig. 5

Open in a new tab

(A) Interaction between P2′ Leu17 of KD1WT-VT with S2′-pocket of mKLK. Similar to hKLK [34], Leu17 of KD1WT-VT is situated in a vastly hydrophobic S2′-pocket environment of mKLK. Pymol generated map of the electrostatic potential at the solvent-accessible surface of the mKLK catalytic domain near residue Leu17 of KD1WT-VT is shown. The mKLK residues within 6 A radius of Leu17 of KD1WT-VT were selected for display. Blue represents positive, red represents negative and white represents hydrophobic surface potential. (B) P2′ Arg17 of BPTI or KD1L17R-KT and S2′-pockets of hKLK and mKLK. Left, Arg39 in hKLK unfavorable for interaction with Arg17 of BPTI or KD1L17R-KT. Right, Short side chain of Thr39 in mKLK can accommodate Arg17 of BPTI or KD1L17R-KT. (C) Additional interactions of hKLK (or mKLK) with BPTI and their absence in KD1WT-VT or KD1L17R-KT. Left, Tyr10 and Arg39 of BPTI make hydrogen bond and a salt-bridge with Glu217 of hKLK or mKLK. Right, the presence of Asp10 and Glu39 in KD1WT-VT or KD1L17R-KT disfavor these interactions.

Fibrinolysis assay

Fibrinolysis data are presented in Fig. 6A–D. Addition of IIa to mouse plasma results in an increase in OD405 reflective of a stable clot formation (curve 1); however, simultaneous addition of tPA results in a biphasic curve with an initial increase in OD405 (fibrin formation), which is followed by a rapid decrease associated with the fibrinolytic activity (curve 2) induced by Pm that is formed by the action of tPA on plasminogen. The data presented in Fig. 6A–D reveal that similar to the human system [34], antifibrinolytic agents inhibit mouse plasma clot fibrinolysis in a dose dependent manner. Complete inhibition of fibrinolysis in this kinetic assay was achieved at ~20 µM for KD1WT-VT, at ~2 µM for KD1L17R-KT, at ~4 µM for aprotinin and at ~500 µM for TXA. Although aprotinin inhibits mPm with higher affinity than KD1L17R-KT (Fig. 2), the inhibition efficiency of each in preventing fibrinolysis is comparable (Fig. 6). These results indicate that the KD1L17R-KT prevents clot lysis by a dual mechanism especially at concentrations ≥1 µM; one, by inhibiting active site of plasmin and two, by inhibiting generation of plasmin at the clot site [29].

Fig. 6. Effect of KD1WT-VT, KD1L17R-KT, aprotinin or TXA on fibrinolysis in mouse plasma.

Fig. 6

Open in a new tab

IIa was added to mouse plasma to initiate clot formation, which is associated with an increase in OD405 (curve 1 in A–D, ο). Simultaneously added tPA converted plasminogen to Pm, which dissolved the fibrin clot completely within ~18 min, as indicated by an initial increase followed by a decrease in OD405 (curve 2 in A–D, ●). Addition of KD1WT-VT, KD1L17R-KT, aprotinin or TXA inhibited fibrinolysis in a dose dependent manner. (A) Effect of KD1WT-VT; 0.5 µM (curve 3△), 1 µM (curve 4▲), 2 µM (curve 5□), 5 µM, (curve 6■), 10 µM (curve 7▽), and 20 µM (curve 8▼). (B) Effect of KD1L17R-KT; 0.2 µM (curve 3△), 0.4 µM (curve 4▲), 0.5 µM (curve 5□), 0.75 µM, (curve 6■), 1 µM (curve 7▽), and 2 µM (curve 8▼). (C) Effect of aprotinin; 0.1 µM (curve 3△), 0.2 µM (curve 4▲), 0.4 µM (curve 5□), 1 µM, (curve 6■), 2 µM (curve 7▽), and 4 µM (curve 8▼). (D) Effect of TXA; 10 µM (curve 3△), 20 µM (curve 4▲), 50 µM (curve 5□), 100 µM, (curve 6■), 250 µM (curve 7▽), and 500 µM (curve 8▼).

Blood loss and D-dimer levels in mouse tail-amputation model

Next, we studied the effect of KD1L17R-KT on blood loss and plasma D-dimer levels in the mouse tail-amputation model. The effects of all four antifibrinolytics on blood loss are depicted in Fig. 7A. There is a significant difference in the mean blood loss between groups as determined by one-way ANOVA (F(4,45)=12.88, p=0.001). A Tukey post-hoc test revealed that compared to saline, the mean blood loss was significantly lower for each antifibrinolytic group. Compared to saline, the mean blood loss was reduced by 190 mg for KD1L17R-KT (95% CI, 104–275; p=0.001), 176 mg for aprotinin (95% CI, 90–261; p=0.001), 112 mg for TXA (95% CI, 26–197; p=0.005), and 87 for KD1WT-VT (95% CI, 15–172; p=0.044). Further, as compared to KD1WT-VT, the mean blood loss was significantly reduced for KD1L17R-KT (95% CI, 17–188, p=0.011) or BPTI (95% CI, 4–175, p=0.037). However, differences in the mean blood loss reduction between KD1L17R-KT and aprotinin (95% CI, -71–99, p=0.991) as well as between KD1L17R-KT and TXA (95% CI, -7–162, p=0.092) or KD1L17R-VT and TXA (95% CI, -110–60, p=0.917) were statistically not significant. Notably, three of the ten animals treated with TXA depicted generalized seizures shortly after injection of the drug. The nature of the seizures were similar to those observed previously with EACA [29]. Further, KD1L17R-KT at the dose tested is comparable to TXA or aprotinin in minimizing the blood loss without observable seizures or renal injury [29,34,35].

Fig. 7. Effect of KD1WT-VT, KD1L17R-KT, aprotinin, and TXA in tail-amputation haemorrhage model.

Fig. 7

Open in a new tab

Animals were treated with saline, KD1WT-VT, KD1L17R-KT, aprotinin or TXA by intravenous bolus injection. The antifibrinolytic agents and saline were free of lipopolysaccharide, as tested using the Limulus amebocyte lysates kit (Bio Whittaker Inc., Walkersville, MD). After allowing two minutes for circulation of the drug, the tail was amputated at 4 cm from the tip using a sharp, sterile, #11 scalpel blade. Blood was allowed to flow freely and drip into preweighed eppendorf tubes. Total blood loss was the difference in the weight of eppendorf tube before and after blood collection. At the end of the experiment, citrated blood samples were obtained from lateral saphenous vein for D-dimer assay by venipuncture. Blood was processed immediately to obtain plasma. The plasma D-dimer concentration was measured using the Asserachrom D-Di kit from Diagnostica Stago. Scatter plot showing blood loss (A) and D-dimer levels (B) in different groups. Black bars represent the mean and the 95% CI (confidence interval). Tukey and Games-Howell post-hoc tests were used to obtain p-values between the different groups for blood loss and the D-dimer levels, respectively. The p-values that are significant between different groups are listed. The D-dimer concentration of unamputated normal mouse plasma was 18 ± 3 ng/ml, which agrees well with the value reported by Castellino and coworkers [75].

All four antifibrinolytics also reduced plasma D-dimer levels (Fig. 7B). Compared to saline, the mean plasma D-dimer levels were reduced by 65 ng/ml for KD1L17R-KT (95% CI, 18–113, p=0.008), 60 ng/ml for aprotinin (95% CI, 11–108, p=0.015) and 55 ng/ml for TXA (95% CI, 6–105, p=0.026). Although the mean D-Dimer levels were reduced by 45 ng/ml for KD1WT-VT, it was statistically not significant (95% CI, -5–95, p=0.091) from saline. Importantly, these data indicate that the fibrinolytic system is activated in this model and KD1L17R-KT is comparable to aprotinin or TXA in inhibiting Pm-induced fibrinolysis.

Aprotinin is more potent (~3-fold) inhibitor of mPm than KD1L17R-KT; however, reduction in blood loss by both inhibitors is similar in the mouse liver-laceration [34] or the tail-amputation model. An explanation for this observation could be that aprotinin also attenuates intrinsic clotting by inhibiting mXIa (KI ~20 nM) and mKLK (Ki ~10 nM), whereas inhibition of these enzymes by KD1L17R-KT is negligible (Table 1). Further, KD1L17R-KT could also inhibit fibrinolysis by impeding generation of Pm by tPA. It is also worth noting that two Pm inhibitors (Ecallantide and MDCO-2010), which inhibit KLK were unsuccessful in phase-3 clinical trials [42,44].

Conclusions

Employing enzyme assays and in silico modeling coupled with the existing crystallographic data [52,70], we provide structural explanations for strong inhibition of mXIa and mKLK by aprotinin and for negligible inhibition by KD1L17R-KT. Structural basis are also provided for inhibition of bovine trypsin by aprotinin and KD1L17R-KT despite the presence of P2′-Arg residue in these inhibitors. The bifunctional KD1L17R-KT is effective in reducing blood loss in mouse tail-amputation and liver-laceration hemorrhage models [29]. Although these mouse models have limitations as compared to the extremity trauma and internal organ hemorrhage in humans, these models are suitable for initial testing the potency of KD1L17R-KT. However, appropriately expressed KD1L17R-KT without the His-tag and with intact C-terminal Lys is essential such that proteolysis is eliminated after purification. We are currently exploring several possibilities for expressing such KD1L17R-KT molecules with one or two additional mutations to perform dose dependent studies in order to test their efficacies.

Supplementary Material

supplement

Essentials.

Current antifibrinolytics—Aminocaproic acid and tranexamic acid can cause seizures or renal injury

KD1L17R-KT, aprotinin and tranexamic acid were tested in a modified mouse tail-amputation model

S2′-subsite variations between human and mouse FXIa result in vastly different inhibition profiles

KD1L17R-KT reduces blood loss and D-dimer levels in mouse with unobserved seizures or renal injury

Acknowledgments

We thank Dr. Philip Liu of Orthopaedic Surgery and Dr. Joni Ricks-Oddie of UCLA IDRE for Statistical Consulting Services. This work was supported, in part, by National Heart, Lung and Blood Institute grant 5R21HL122878.

Footnotes

Addendum

K. Vadivel performed enzyme inhibition, molecular modeling and structural analysis, and wrote the manuscript. Y. Kumar performed enzyme inhibition, fibrinolysis and animal experiments. G.I. Ogueli expressed KD1WT-VT and KD1L17R-KT proteins. S.M. Ponnuraj processed mouse blood samples and performed D-dimer assays. P. Wongkongkathep and J.A. Loo performed mass spectrometry experiments. M.S. Bajaj and S.P. Bajaj conceived, designed, analyzed and interpreted the data, and wrote the manuscript.

Disclosure of conflict of interests

S.P. Bajaj has patents issued and pending on the KD1L17R-KT and related molecules. The view and conclusions drawn are the sole responsibility of the authors.

References

  • 1.Grant MA, Aird WC. In: Molecular evolution of the vertebrate blood coagulation system Thrombosisand Hemostasis: Basic principles & clinical practice. 6. Marder VJ, Aird WC, Bennett JS, Schulman S, White GC II, editors. Lippincott: Williams & Wilkins; 2012. pp. 11–25. [Google Scholar]
  • 2.Urano S, Metzger AR, Castellino FJ. Plasmin-mediated fibrinolysis by variant recombinant tissue plasminogen activators. Proc Natl Acad Sci USA. 1989;86:2568–71. doi: 10.1073/pnas.86.8.2568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Plow EF, Herren T, Redlitz A, Miles LA, Hoover-Plow JL. The cell biology of the plasminogen system. FASEB J. 1995;9:939–45. doi: 10.1096/fasebj.9.10.7615163. [DOI] [PubMed] [Google Scholar]
  • 4.Kashuk JL, Moore EE, Sawyer M, Wohlauer M, Pezold M, Barnett C, Biffl WL, Burlew CC, Johnson JL, Sauaia A. Primary fibrinolysis is integral in the pathogenesis of the acute coagulopathy of trauma. Ann Surg. 2010;252:434–42. doi: 10.1097/SLA.0b013e3181f09191. [DOI] [PubMed] [Google Scholar]
  • 5.Raza I, Davenport R, Rourke C, Platton S, Manson J, Spoors C, Khan S, De'Ath HD, Allard S, Hart DP, Pasi KJ, Hunt BJ, Stanworth S, MacCallum PK, Brohi K. The incidence and magnitude of fibrinolytic activation in trauma patients. J Thromb Haemost. 2013;11:307–14. doi: 10.1111/jth.12078. [DOI] [PubMed] [Google Scholar]
  • 6.Cardenas JC, Wade CE, Holcomb JB. Mechanisms of trauma-induced coagulopathy. Curr Opin Hematol. 2014;21:404–9. doi: 10.1097/MOH.0000000000000063. [DOI] [PubMed] [Google Scholar]
  • 7.Boucher BA, Hannon TJ. Blood management: a primer for clinicians. Pharmacotherapy. 2007;27:1394–411. doi: 10.1592/phco.27.10.1394. [DOI] [PubMed] [Google Scholar]
  • 8.Hendrickson JE, Hillyer CD. Noninfectious serious hazards of transfusion. Anesth Analg. 2009;108:759–69. doi: 10.1213/ane.0b013e3181930a6e. [DOI] [PubMed] [Google Scholar]
  • 9.Ker K, Roberts I, Shakur H, Coats TJ. Antifibrinolytic drugs for acute traumatic injury. Cochrane Database Syst Rev. 2015:CD004896. doi: 10.1002/14651858.CD004896.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Koster A, Faraoni D, Levy JH. Antifibrinolytic Therapy for Cardiac Surgery: An Update. Anesthesiology. 2015;123:214–21. doi: 10.1097/ALN.0000000000000688. [DOI] [PubMed] [Google Scholar]
  • 11.Hartmann M, Sucker C, Boehm O, Koch A, Loer S, Zacharowski K. Effects of cardiac surgery on hemostasis. Transfusion Med Rev. 2006;20:230–41. doi: 10.1016/j.tmrv.2006.03.003. [DOI] [PubMed] [Google Scholar]
  • 12.Fergusson DA, Hebert PC, Mazer CD, Fremes S, MacAdams C, Murkin JM, Teoh K, Duke PC, Arellano R, Blajchman MA, Bussieres JS, Cote D, Karski J, Martineau R, Robblee JA, Rodger M, Wells G, Clinch J, Pretorius R BART Investigators. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery. N Engl J Med. 2008;358:2319–31. doi: 10.1056/NEJMoa0802395. [DOI] [PubMed] [Google Scholar]
  • 13.Sharma V, Katznelson R, Jerath A, Garrido-Olivares L, Carroll J, Rao V, Wasowicz M, Djaiani G. The association between tranexamic acid and convulsive seizures after cardiac surgery: a multivariate analysis in 11529 patients. Anaesthesia. 2014;69:124–30. doi: 10.1111/anae.12516. [DOI] [PubMed] [Google Scholar]
  • 14.Lecker I, Wang DS, Whissell PD, Avramescu S, Mazer CD, Orser BA. Tranexamic acid-associated seizures: Causes and treatment. Ann Neurol. 2016;79:18–26. doi: 10.1002/ana.24558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Martin K, Wiesner G, Breuer T, Lange R, Tassani P. The risks of aprotinin and tranexamic acid in cardiac surgery: a one-year follow-up of 1188 consecutive patients. Anesth Analg. 2008;107:1783–90. doi: 10.1213/ane.0b013e318184bc20. [DOI] [PubMed] [Google Scholar]
  • 16.Breuer T, Martin K, Wilhelm M, Wiesner G, Schreiber C, Hess J, Lange R, Tassani P. The blood sparing effect and the safety of aprotinin compared to tranexamic acid in paediatric cardiac surgery. Eur J Cardiothorac Surg. 2009;35:167–71. doi: 10.1016/j.ejcts.2008.09.038. [DOI] [PubMed] [Google Scholar]
  • 17.Sander M, Spies CD, Martiny V, Rosenthal C, Wernecke KD, von Heymann C. Mortality associated with administration of high-dose tranexamic acid and aprotinin in primary open-heart procedures: a retrospective analysis. Crit Care. 2010;14:R148. doi: 10.1186/cc9216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Martin K, Knorr J, Breuer T, Gertler R, Macguill M, Lange R, Tassani P, Wiesner G. Seizures after open heart surgery: comparison of ε-aminocaproic acid and tranexamic acid. J Cardiothorac Vasc Anesth. 2011;25:20–5. doi: 10.1053/j.jvca.2010.10.007. [DOI] [PubMed] [Google Scholar]
  • 19.Montes FR, Pardo DF, Carreño M, Arciniegas C, Dennis RJ, Umaña JP. Risk factors associated with postoperative seizures in patients undergoing cardiac surgery who received tranexamic acid: A case-control study. Ann Card Anaesth. 2012;15:6–12. doi: 10.4103/0971-9784.91467. [DOI] [PubMed] [Google Scholar]
  • 20.Makhija N, Sarupria A, Kumar Choudhary S, Das S, Lakshmy R, Kiran U. Comparison of epsilon aminocaproic acid and tranexamic Acid in thoracic aortic surgery: clinical efficacy and safety. J Cardiothorac Vasc Anesth. 2013;27:1201–7. doi: 10.1053/j.jvca.2013.04.003. [DOI] [PubMed] [Google Scholar]
  • 21.Shaw AD, Stafford-Smith M, White WD, Phillips-Bute B, Swaminathan M, Milano C, Welsby IJ, Aronson S, Mathew JP, Peterson ED, Newman MF. The effect of aprotinin on outcome after coronary-artery bypass grafting. N Engl J Med. 2008;358:784–93. doi: 10.1056/NEJMoa0707768. [DOI] [PubMed] [Google Scholar]
  • 22.Rabinovici R, Heyman A, Kluger Y, Shinar E. Convulsions induced by aminocaproic acid infusion. DICP. 1989;23:780–1. doi: 10.1177/106002808902301008. [DOI] [PubMed] [Google Scholar]
  • 23.Lenfors S, Gustafsson D. New model for in vivo studies of pharmacological interventions with endogenous fibrinolysis: effects of thrombin inhibitors. Semin Thromb Hemost. 1996;22:335–42. doi: 10.1055/s-2007-999028. [DOI] [PubMed] [Google Scholar]
  • 24.Mao SS, Holahan MA, Bailey C, Wu G, Colussi D, Carroll SS, Cook JJ. Demonstration of enhanced endogenous fibrinolysis in thrombin activatable fibrinolysis inhibitor-deficient mice. Blood Coagul Fibrinolysis. 2005;16:407–15. doi: 10.1097/01.mbc.0000181175.62437.2a. [DOI] [PubMed] [Google Scholar]
  • 25.Paul J, Strickland S, Melchor JP. Fibrin deposition accelerates neurovascular damage and neuroinflammation in mouse models of Alzheimer's disease. J Exp Med. 2007;204:1999–2008. doi: 10.1084/jem.20070304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Bruno NE, Yano Y, Takei Y, Gabazza EC, Qin L, Nagashima M, Morser J, D'Alessandro-Gabazza CN, Taguchi O, Sumida Y. Protective role of thrombin activatable fibrinolysis inhibitor in obstructive nephropathy-associated tubulointerstitial fibrosis. J Thromb Haemost. 2008;6:139–46. doi: 10.1111/j.1538-7836.2007.02826.x. [DOI] [PubMed] [Google Scholar]
  • 27.Kłak M, Anakkala N, Wang W, Lange S, Jonsson IM, Tarkowski A, Jin T. Tranexamic acid, an inhibitor of plasminogen activation, aggravates staphylococcal septic arthritis and sepsis. Scand J Infect Dis. 2010;42:351–8. doi: 10.3109/00365540903510690. [DOI] [PubMed] [Google Scholar]
  • 28.Lecker I, Wang DS, Romaschin AD, Peterson M, Mazer CD, Orser BA. Tranexamic acid concentrations associated with human seizures inhibit glycine receptors. J Clin Invest. 2012;122:4654–66. doi: 10.1172/JCI63375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kumar Y, Vadivel K, Schmidt AE, Ogueli GI, Ponnuraj SM, Rannulu N, Loo JA, Bajaj MS, Bajaj SP. Decoy plasminogen receptor containing a selective Kunitz-inhibitory domain. Biochemistry. 2014;53:505–17. doi: 10.1021/bi401584b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Schlag MG, Hopf R, Redl H. Convulsive seizures following subdural application of fibrin sealant containing tranexamic acid in a rat model. Neurosurgery. 2000;47:1463–7. [PubMed] [Google Scholar]
  • 31.Furtmüller R, Schlag MG, Berger M, Hopf R, Huck S, Sieghart W, Redl H. Tranexamic acid, a widely used antifibrinolytic agent, causes convulsions by a gamma-aminobutyric acidA receptor antagonistic effect. J Pharmacol Exp Ther. 2002;301:168–73. doi: 10.1124/jpet.301.1.168. [DOI] [PubMed] [Google Scholar]
  • 32.Yamaura A, Nakamura T, Makino H, Hagihara Y. Cerebral complication of antifibrinolytic therapy in the treatment of ruptured intracranial aneurysm. Animal experiment and a review of literature. Eur Neurol. 1980;19:77–84. doi: 10.1159/000115131. [DOI] [PubMed] [Google Scholar]
  • 33.Pellegrini A, Giaretta D, Chemello R, Zanotto L, Testa G. Feline generalized epilepsy induced by tranexamic acid (AMCA) Epilepsia. 1982;23:35–45. doi: 10.1111/j.1528-1157.1982.tb05051.x. [DOI] [PubMed] [Google Scholar]
  • 34.Bajaj MS, Ogueli GI, Kumar Y, Vadivel K, Lawson G, Shanker S, Schmidt AE, Bajaj SP. Engineering Kunitz domain 1 (KD1) of human tissue factor pathway inhibitor-2 to selectively inhibit fibrinolysis: properties of KD1-L17R variant. J Biol Chem. 2011;286:4329–40. doi: 10.1074/jbc.M110.191163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Kempaiah P, Danielson LA, Barry M, Kisiel W. Comparative effects of aprotinin and human recombinant R24K KD1 on temporal renal function in Long-Evans rats. J Pharmacol Exp Ther. 2009;331:940–5. doi: 10.1124/jpet.109.161034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Markus G, DePasquale JL, Wissler FC. Quantitative determination of the binding of epsilon-aminocaproic acid to native plasminogen. J Biol Chem. 1978;253:727–32. [PubMed] [Google Scholar]
  • 37.Mathews II, Vanderhoff-Hanaver P, Castellino FJ, Tulinsky A. Crystal structures of the recombinant kringle 1 domain of human plasminogen in complexes with the ligands epsilon-aminocaproic acid and trans-4-(aminomethyl)cyclohexane-1-carboxylic Acid. Biochemistry. 1996;35:2567–76. doi: 10.1021/bi9521351. [DOI] [PubMed] [Google Scholar]
  • 38.Markland W, Ley AC, Lee SW, Ladner RC. Iterative optimization of high-affinity proteases inhibitors using phage display. 1. Plasmin. Biochemistry. 1996;35:8045–57. doi: 10.1021/bi9526286. [DOI] [PubMed] [Google Scholar]
  • 39.Dietrich W, Nicklisch S, Koster A, Spannagl M, Giersiefen H, van de Locht A. CU-2010--a novel small molecule protease inhibitor with antifibrinolytic and anticoagulant properties. Anesthesiology. 2009;110:123–30. doi: 10.1097/ALN.0b013e318191408c. [DOI] [PubMed] [Google Scholar]
  • 40.Flight SM, Johnson LA, Du QS, Warner RL, Trabi M, Gaffney PJ, Lavin MF, de Jersey J, Masci PP. Textilinin-1, an alternative anti-bleeding agent to aprotinin: Importance of plasmin inhibition in controlling blood loss. Br J Haematol. 2009;145:207–11. doi: 10.1111/j.1365-2141.2009.07605.x. [DOI] [PubMed] [Google Scholar]
  • 41.Swedberg JE, Harris JM. Plasmin substrate binding site cooperativity guides the design of potent peptide aldehyde inhibitors. Biochemistry. 2011;50:8454–62. doi: 10.1021/bi201203y. [DOI] [PubMed] [Google Scholar]
  • 42.Bokesch PM, Szabo G, Wojdyga R, Grocott HP, Smith PK, Mazer CD, Vetticaden S, Wheeler A, Levy JH. A phase 2 prospective, randomized, double-blind trial comparing the effects of tranexamic acid with ecallantide on blood loss from high-risk cardiac surgery with cardiopulmonary bypass (CONSERV-2 Trial) J Thorac Cardiovasc Surg. 2012;143:1022–9. doi: 10.1016/j.jtcvs.2011.06.001. [DOI] [PubMed] [Google Scholar]
  • 43.Saupe SM, Leubner S, Betz M, Klebe G, Steinmetzer T. Development of new cyclic plasmin inhibitors with excellent potency and selectivity. J Med Chem. 2013;56:820–31. doi: 10.1021/jm3012917. [DOI] [PubMed] [Google Scholar]
  • 44.Englberger L, Dietrich W, Eberle B, Erdoes G, Keller D, Carrel T. A novel blood-sparing agent in cardiac surgery? First in-patient experience with the synthetic serine protease inhibitor MDCO-2010: A phase II, randomized, double-blind, placebo-controlled study in patients undergoing coronary artery bypass grafting with cardiopulmonary bypass. Anesth Analg. 2014;119:16–25. doi: 10.1213/ANE.0000000000000218. [DOI] [PubMed] [Google Scholar]
  • 45.Cheng L, Pettersen D, Ohlsson B, Schell P, Karle M, Evertsson E, Pahlen S, Jonforsen M, Plowright AT, Bostrom J, Fex T, Thelin A, Hilgendorf C, Xue Y, Wahlund G, Lindberg W, Larsson LO, Gustafsson D. Discovery of the Fibrinolysis Inhibitor AZD6564, Acting via Interference of a Protein-Protein Interaction. ACS Med Chem Lett. 2014;5:538–43. doi: 10.1021/ml400526d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Al-Horani RA, Desai UR. Recent advances on plasmin inhibitors for the treatment of fibrinolysis-related disorders. Med Res Rev. 2014;34:1168–216. doi: 10.1002/med.21315. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.de Veer SJ, Wang CK, Harris JM, Craik DJ, Swedberg JE. Improving the Selectivity of Engineered Protease Inhibitors: Optimizing the P2 Prime Residue Using a Versatile Cyclic Peptide Library. J Med Chem. 2015;58:8257–68. doi: 10.1021/acs.jmedchem.5b01148. [DOI] [PubMed] [Google Scholar]
  • 48.Masci PP, Whitaker AN, Sparrow LG, de Jersey J, Winzor DJ, Watters DJ, Lavin MF, Gaffney PJ. Textilinins from Pseudonaja textilis textilis. Characterization of two plasmin inhibitors that reduce bleeding in an animal model. Blood Coagul Fibrinolysis. 2000;11:385–93. doi: 10.1097/00001721-200006000-00011. [DOI] [PubMed] [Google Scholar]
  • 49.Jankun J, Keck R, Selman SH, Skrzypczak-Jankun E. Systemic or topical application of plasminogen activator inhibitor with extended half-life (VLHL PAI-1) reduces bleeding time and total blood loss. Int J Mol Med. 2010;26:501–4. doi: 10.3892/ijmm_00000491. [DOI] [PubMed] [Google Scholar]
  • 50.Laemmli UK. Cleavage of structural proteins during assembly of head of bacteriophage-t4. Nature. 1970;227:680–5. doi: 10.1038/227680a0. [DOI] [PubMed] [Google Scholar]
  • 51.Chand HS, Schmidt AE, Bajaj SP, Kisiel W. Structure-function analysis of the reactive site in the first Kunitz-type domain of human tissue factor pathway inhibitor-2. J Biol Chem. 2004;279:17500–7. doi: 10.1074/jbc.M400802200. [DOI] [PubMed] [Google Scholar]
  • 52.Schmidt AE, Chand HS, Cascio D, Kisiel W, Bajaj SP. Crystal structure of Kunitz domain 1 (KD1) of tissue factor pathway inhibitor-2 in complex with trypsin. Implications for KD1 specificity of inhibition. J Biol Chem. 2005;280:27832–8. doi: 10.1074/jbc.M504105200. [DOI] [PubMed] [Google Scholar]
  • 53.Beith JG. In vivo significance of kinetic constants of protein proteinase inhibitors. Biochem Med. 1984;32:387–97. doi: 10.1016/0006-2944(84)90046-2. [DOI] [PubMed] [Google Scholar]
  • 54.Morrison JF, Walsh CT. The behavior and significance of slow-binding enzyme inhibitors. Adv Enzymol Relat Areas Mol Biol. 1988;61:201–301. doi: 10.1002/9780470123072.ch5. [DOI] [PubMed] [Google Scholar]
  • 55.Bojar RM. Manual of perioperative care in adult cardiac surgery. 4. Blackwell Publishing, Hoboken; New Jersey: 2005. Cardiac Anesthesia; pp. 129–178. [Google Scholar]
  • 56.Nilsson IM. Clinical pharmacology of aminocaproic and tranexamic acids. J Clin Pathol Suppl (R Coll Pathol) 1980;14:41–7. [PMC free article] [PubMed] [Google Scholar]
  • 57.Broze GJ, Jr, Yin ZF, Lasky N. A tail vein bleeding time model and delayed bleeding in hemophiliac mice. Thromb Haemost. 2001;85:747–8. [PubMed] [Google Scholar]
  • 58. [Accessed 8 September 2016];Australian public assessment report for tranexamic acid. Cyklokapran. https://www.tga.gov.au/sites/default/files/auspar-cyklokapron.pdf.
  • 59. [Accessed 25 July 2016];Trasylol. http://www.accessdata.fda.gov/drugsatfda_docs/label/2006/020304s022lbl.pdf.
  • 60.Farkas H, Varga L. Ecallantide is a novel treatment for attacks of hereditary angioedema due to C1 inhibitor deficiency. Clin Cosmet Investig Dermatol. 2011;4:61–8. doi: 10.2147/CCID.S10322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Markwardt F. Naturally occurring inhibitors of fibrinolysis. In: Markwardt F, editor. Fibrinolytics and Anti-fibrinolytics. Springer-Verlag; 1978. pp. 487–509. [Google Scholar]
  • 62.Eswar N, Webb B, Marti-Renom MA, Madhusudhan MS, Eramian D, Shen MY, Pieper U, Sali A. Comparative Protein Structure Modeling With MODELLER. Current Protocols in Bioinformatics, John Wiley & Sons. 2006;(Supplement 15):5.6.1–30. doi: 10.1002/0471250953.bi0506s15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE. The Protein Data Bank. Nucleic Acids Res. 2000;28:235–42. doi: 10.1093/nar/28.1.235. www.rcsb.org. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Brooks BR, Brooks CL, 3rd, Mackerell AD, Jr, Nilsson L, Petrella RJ, Roux B, Won Y, Archontis G, Bartels C, Boresch S, Caflisch A, Caves L, Cui Q, Dinner AR, Feig M, Fischer S, Gao J, Hodoscek M, Im W, Kuczera K. CHARMM: the biomolecular simulation program. J Comput Chem. 2009;30:1545–614. doi: 10.1002/jcc.21287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Baker NA, Sept D, Joseph S, Holst MJ, McCammon JA. Electrostatics of nanosystems: application to microtubules and the ribosome. Proc Natl Acad Sci USA. 2001;98:10037–41. doi: 10.1073/pnas.181342398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.The PyMOL Molecular Graphics System, Version 1.8 Schr00F6dinger, LLC.
  • 67.Fritz H, Wunderer G. Biochemistry and applications of aprotinin, the kallikrein inhibitor from bovine organs. Arzneimittelforschung/Drug Res. 1983;33:479–94. [PubMed] [Google Scholar]
  • 68.Petersen TE, Martzen MR, Ichinose A, Davie EW. Characterization of the gene for human plasminogen, a key proenzyme in the fibrinolytic system. J Biol Chem. 1990;265:6104–11. [PubMed] [Google Scholar]
  • 69.Degen SJ, Bell SM, Schaefer LA, Elliott RW. Characterization of the cDNA coding for mouse plasminogen and localization of the gene to mouse chromosome 17. Genomics. 1990;8:49–61. doi: 10.1016/0888-7543(90)90225-j. [DOI] [PubMed] [Google Scholar]
  • 70.Marquart M, Walter J, Deisenhofer J, Bode W, Huber R. The geometry of the reactive site and of the peptide groups in trypsin, trypsinogen and its complexes with inhibitors. Acta Cryst. 1983;B39:480–90. [Google Scholar]
  • 71.Fujikawa K, Chung DW, Hendrickson LE, Davie EW. Amino acid sequence of human factor XI, a blood coagulation factor with four tandem repeats that are highly homologous with plasma prekallikrein. Biochemistry. 1986;25:2417–24. doi: 10.1021/bi00357a018. [DOI] [PubMed] [Google Scholar]
  • 72.Gailani D, Sun MF, Sun Y. A comparison of murine and human factor XI. Blood. 1997;90:1055–64. [PubMed] [Google Scholar]
  • 73.Chung DW, Fujikawa K, McMullen BA, Davie EW. Human plasma prekallikrein, a zymogen to a serine protease that contains four tandem repeats. Biochemistry. 1986;25:2410–7. doi: 10.1021/bi00357a017. [DOI] [PubMed] [Google Scholar]
  • 74.Seidah NG, Sawyer N, Hamelin J, Mion P, Beaubien G, Brachpapa L, Rochemont J, Mbikay M, Chrétien M. Mouse plasma kallikrein: cDNA structure, enzyme characterization, and comparison of protein and mRNA levels among species. DNA Cell Biol. 1990;9:737–48. doi: 10.1089/dna.1990.9.737. [DOI] [PubMed] [Google Scholar]
  • 75.Lay AJ, Liang Z, Rosen ED, Castellino FJ. Mice with a severe deficiency in protein C display prothrombotic and proinflammatory phenotypes and compromised maternal reproductive capabilities. J Clin Invest. 2005;115:1552–61. doi: 10.1172/JCI24030. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement
NIHMS872810-supplement-supplement.docx (17KB, docx)

RESOURCES