Murine Thrombin Lacks Na+ Activation but Retains High Catalytic Activity*

Human thrombin utilizes Na+ as a driving force for the cleavage of substrates mediating its procoagulant, prothrombotic, and signaling functions. Murine thrombin has Asp-222 in the Na+ binding site of the human enzyme replaced by Lys. The charge reversal substitution abrogates Na+ activation, which is partially restored with the K222D mutation, and ensures high activity even in the absence of Na+. This property makes the murine enzyme more resistant to the effect of mutations that destabilize Na+ binding and shift thrombin to its anticoagulant slow form. Compared with the human enzyme, murine thrombin cleaves fibrinogen and protein C with similar kcat/Km values but activates PAR1 and PAR4 with kcat/Km values 4- and 26-fold higher, respectively. The significantly higher specificity constant toward PAR4 accounts for the dominant role of this receptor in platelet activation in the mouse. Murine thrombin can also cleave substrates carrying Phe at P1, which potentially broadens the repertoire of molecular targets available to the enzyme in vivo.

Thrombin is a Na ϩ -activated serine protease (1) that is responsible for the progression and regulation of blood coagulation (2). As in other monovalent cation-activated enzymes (3), the role of Na ϩ in thrombin function is to lower energy barriers for substrate binding in the ground and transition states so that physiologic substrates like PAR1 and fibrinogen can be hydrolyzed efficiently during platelet activation and formation of a blood clot. Other members of the vitamin K-dependent family of clotting proteases to which thrombin belongs are endowed with Na ϩ activation (4). In fact, the activity of activated protein C (5) and coagulation factors VIIa (6), IXa (7), and Xa (8) is influenced significantly by the presence of Na ϩ .
Details of the molecular mechanism of Na ϩ activation in thrombin have emerged recently from mutagenesis and structural analysis (9). Na ϩ binding is severely compromised (Ͼ30-fold increase in K d ) upon mutation of Asp-189, Glu-217, Asp-222, and Tyr-225. Asp-189 fixes the orientation of one of the four water molecules ligating Na ϩ and provides an important link between the Na ϩ site and the P1 residue of substrate (10). Glu-217 makes polar contacts with Lys-224 and Thr-172, which helps to stabilize the intervening 220-loop in the Na ϩ site. The ion pair between Arg-187 and Asp-222 latches the 186-loop onto the 220-loop to stabilize the Na ϩ site and the pore of entry of the cation to its binding site (11). Tyr-225 plays a crucial role in determining the Na ϩ -dependent allosteric nature of serine proteases (4) by allowing the correct orienta-tion of the backbone oxygen of residue 224 (12), which contributes directly to the coordination of Na ϩ . The side chain of Tyr-225 also secures the integrity of the water channel embedding the primary specificity pocket (12), and the backbone around Tyr-225 is oriented like the selectivity filter of the KcsA K ϩ channel (3,13). Of these four residues, Glu-217 and Tyr-225 are conserved in thrombin from all species sequenced to date, from hagfish to human (14). Asp-189 is a Ser in the sturgeon, and the D189S mutant of human thrombin has impaired Na ϩ binding and substrate recognition (10). Asp-222 is the least conserved residue among the four. It is Ser in the sturgeon, Lys in the mouse, and Asn in the rat (14). The substitution in the mouse is particularly interesting, as it involves a charge reversal in the 220-loop that has the potential to destabilize the Na ϩ binding environment.
Does the presence of Lys-222 in murine thrombin preclude Na ϩ binding? If so, what compensates for the potential loss of catalytic activity? The D221A/D222K mutant of human thrombin is devoid of Na ϩ activation (15) due to complete disruption of the Na ϩ binding site (16). However, the mutant has functional properties intermediate between those of the Na ϩ -free and Na ϩ -bound forms of the wild type (15), which suggests that murine thrombin can retain significant catalytic activity even in the absence of Na ϩ activation. On the other hand, Lys-222 is present in factor Xa, which is a Na ϩ -activated enzyme (17). Hence, the alternative possibility exists that the D222K substitution in murine thrombin is inconsequential on Na ϩ binding. To address these questions directly, we expressed and purified to homogeneity murine thrombin. The results presented here fill an important gap in our understanding of an enzyme that has been the subject of numerous transgenic studies in recent years (18 -21) but for which no biochemical characterization has been reported.

MATERIALS AND METHODS
Construction of the Murine Pre-thrombin 1 Vector-The HPC4-pNUT expression vector containing the murine pre-thrombin 1 gene was constructed using the ribocloning technique (22). The murine prethrombin 1 gene was amplified by PCR from the PCDNA3 vector (kindly provided by Dr. Jay Degen) using the following primers: 5Ј-TC-TGGAGGTTCCAAGGACAATCTGTCACCTCC-3Ј and 5Ј-CTAT-CCAAATTGATCAATGACTTTCTG-3Ј. The gene was prepared for insertion immediately following the HPC4 epitope in the HPC4-pNUT vector using two riboprimers and two "bandaid" primers. Klen-Taq long and accurate DNA polymerase (22) was used at pH 7.9 for the PCR. The HPC4-pNUT target was prepared using two riboprimers that amplified the vector with an insertion gap for the murine pre-thrombin 1 gene. Following PCR, DpnI was added to eliminate parent DNA. The amplified products were precipitated with PEG 2 and washed with 70% ethanol. After resuspension in DNA buffer (10 mM Tris, 10 mM NaCl, * This work was supported in part by National Institutes of Health Research Grants HL49413, HL58141, and HL73813. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed. Tel.: 314-362-4185; Fax: 314-747-5354; E-mail: enrico@wustl.edu. 0.1 mM EDTA, pH 7.9), RNase A was added to cut the plasmid/gene at the ribobases resulting in sticky ends on the target plasmid and the amplified murine pre-thrombin 1 gene. RNase A was digested by addition of proteinase K. The prepared HPC4-pNUT target and the prepared murine pre-thrombin 1 gene insert were mixed at a 1:1 molar ratio in DNA buffer with 0.5 M NaCl and annealed by heating at 75°C for 4 min followed by slow cooling to 25°C over 60 min. The product was then transformed directly into XL1-Blue cells (Stratagene) and plated on low salt LB media containing 100 g/ml ampicillin for selection. Plasmid DNA was isolated and purified from single colonies using Maxipreps from Qiagen. The K222D and W215A/E217A mutants were constructed with the QuikChange system (Stratagene) using the HPC-4-pNUT murine pre-thrombin 1 plasmid as a template. Transformation and plasmid purification were done as described for the wild type.
Expression and Activation of Murine Thrombin-Expression of wildtype and mutant K222D and W215A/E217A murine thrombins was carried out in a baby hamster kidney cell system as described for human thrombin (12). The expressed murine thrombin was concentrated using a Quixstand tangential flow system and purified on an HPC4 antibody column before activation. Activation was carried out overnight using the snake venom enzyme ecarin (American Diagnostica). Activated thrombin was purified to homogeneity by FPLC using Resource Q and S columns with a linear gradient from 0.04 to 1 M choline chloride (ChCl), 5 mM MES, pH 6.0, at room temperature. The active site concentration was determined by titration with hirudin.
Functional Assays-Wild-type trypsin and tissue-type plasminogen activator were expressed, purified, and tested for activity as described (23,24). Chymotrypsin was from Sigma. Activated protein C, factor Xa, factor XIa, and plasmin were from Enzyme Research. The substrates H-D-Phe-Pro-Phe-p-nitroanilide (FPF) and H-D-Phe-Pro-Arg-p-nitroanilide (FPR) were from MidWest Biotech, and Spectrozyme-TH was from American Diagnostica. Measurements of the Michaelis-Menten parameters k cat and s ϭ k cat /K m were carried out from direct integration of progress curves of substrate hydrolysis taking into account product inhibition (25). The dependence of these parameters on [Na ϩ ] was fit according to the equations (26), where x ϭ [Na ϩ ], s 0 and s 1 are the values of s for x ϭ 0 and x ϭ ∞, k cat,0 and k cat,1 are the values of k cat for x ϭ 0 and x ϭ ∞, and K a and K a Ј are the equilibrium association constants for Na ϩ binding to the free enzyme and the enzyme-substrate complex. Experimental conditions were 50 mM Tris, 0.1% PEG 8000, pH 8.0, at 25°C. The ionic strength was kept constant at 200 mM by replacing NaCl with ChCl. Under these conditions, murine thrombin was subject to very significant product inhibition in the presence of FPR, which prevented accurate measurements of K m . The values of s ϭ k cat /K m could not be estimated with confidence, because even substrate concentrations as low as 1 M were apparently saturating. In the process of screening alternative substrates that would retain a Na ϩ effect in the human enzyme for comparison with murine thrombin, we selected Spectrozyme-TH for our titration studies. We also found that thrombin would cleave FPF with values of s comparable with those of chymotrypsin, in agreement with studies on ester substrates published more than 40 years ago (27)(28)(29).
Fibrinogen and protein C (murine and human; from Hematologic Technologies) were used for the characterization of murine thrombin under experimental conditions of 5 mM Tris, 0.1% PEG, 145 mM NaCl, pH 7.4, at 37°C. The assays were performed as described for the human enzyme-substrate interactions (30 -32). The interaction of murine thrombin with murine PAR1 and PAR4 was studied as for the analogous human enzyme-substrate interactions (33,34) from the kinetics of cleavage of soluble fragments corresponding to the extracellular portion of the murine receptors. These fragments were synthesized by solid phase chromatography, purified to homogeneity by high pressure liquid chromatography, and tested for purity by mass spectrometry. Their sequences were (2 ϭ site of cleavage) 32 ERTDATVNPR2SFF-LRNPSENTFELVPLGDEE 63 for PAR1 and 51 KSSDKPNPR2GY-PGKFCANDSDTLELPASSQA 81 for PAR4. Fluorescence measurements of Na ϩ binding were carried out as reported (11), using a FluoroMax-3 SPEX spectrophotometer, under experimental conditions of 5 mM Tris, 0.1% PEG 8000, pH 8.0, at 10°C, I ϭ 800 mM. Control experiments of FPF hydrolysis by murine thrombin were carried out under identical solution conditions. Fig. 1 shows the effect of Na ϩ on the hydrolysis of the synthetic substrate Spectrozyme-TH by human thrombin. With this substrate, the value of s ϭ k cat /K m increases 70-fold from 1.

RESULTS
The effect is observed at an ionic strength of 200 mM kept constant with ChCl. The value of k cat for this substrate also changes significantly from 15 s Ϫ1 at [Na ϩ ] ϭ 0 to 91 s Ϫ1 at saturating [Na ϩ ]. The values of K a and K a Ј derived from titration of s and k cat are 33 and 45 M Ϫ1 , respectively. When the same titrations are carried out with murine thrombin, the values of s and k cat are constant over the entire [Na ϩ ] range and fall in between those of the slow ([Na ϩ ] ϭ 0) and fast ([Na ϩ ] ϭ ∞) forms. This shows that murine thrombin is not a Na ϩ -activated enzyme. However, the enzyme retains significant catalytic activity because the value of s is almost 20-fold higher than that of the slow form of human thrombin.
We surmised that the presence of Lys-222 in the Na ϩ binding loop of murine thrombin prevents Na ϩ binding but, at the same time, provides a partial functional mimicry of the Na ϩ effect by stabilizing the enzyme in a conformation similar to that of the fast form of the human enzyme (see below). In the human enzyme, the D222A replacement compromises Na ϩ binding (9). The D221A/D222K double mutant is devoid of Na ϩ activation (15), and its structure shows no evidence of bound Na ϩ (16). However, the mutant has functional properties intermediate to those of the slow and fast forms of the wild type (15). To directly test the role of Lys-222 in murine thrombin, we replaced it with Asp. The K222D mutation restores a great deal of Na ϩ activation, as revealed most eloquently by the k cat (Fig. 1); the value of s increases from 39 M Ϫ1 s Ϫ1 at [Na ϩ ] ϭ 0 to 69 M Ϫ1 s Ϫ1 at saturating [Na ϩ ], and the value of k cat increases from 21 s Ϫ1 at [Na ϩ ] ϭ 0 to 80 s Ϫ1 at saturating [Na ϩ ]. The Na ϩ affinity of this mutant is very close to human thrombin, with K a and K a Ј values of 25 and 100 M Ϫ1 , respectively.
Another intriguing property of murine thrombin is its ability to cleave amide substrates specific for chymotrypsin. Thrombin specificity is considered trypsin-like because of the presence of Asp-189 in the primary specificity pocket and the fact that all known physiologic substrates carry Arg at P1, with the sole exception of Lys in PAR3 (35). However, murine thrombin cleaves FPF with a value of s comparable with that of chymotrypsin and several orders of magnitude higher than that of trypsin (Fig. 2). This property is shared by the human enzyme and stands in contrast to other clotting and fibrinolytic proteases carrying Asp-189 (Fig. 2). As for Spectrozyme-TH carrying Arg at P1, cleavage of FPF shows a significant Na ϩ effect in the human enzyme but not in murine thrombin. Interestingly, murine thrombin cleaves FPF with a value of s similar to that of the fast form of human thrombin, and the Na ϩ effect is restored in the K222D mutant. These changes confirm the importance of the linkage between Na ϩ binding and the S1 site of the enzyme documented by structural (9) and mutagenesis (10) studies. Although FPF lacks a charged P1 moiety, the side chain of Phe at P1 is still capable of sensing the Na ϩ induced changes around the environment of Asp-189 in the primary specificity pocket. The ability of thrombin to cleave chymotrypsin-specific substrates was discovered more than 40 years ago; bovine thrombin was found capable of hydrolyzing ester substrates carrying Phe at P1 with values of s comparable with those of chymotrypsin (27)(28)(29). Surprisingly, this important observation has had little impact on subsequent studies of thrombin specificity toward synthetic amide substrates (36,37), including recent large scale profiling (38,39). The trypsin-like specificity of thrombin has never been put to experimental test by changing the P1 position of amide substrate to residues other than Arg or Lys.
Fluorescence measurements of Na ϩ binding to murine thrombin fail to report any significant change up to 800 mM [Na ϩ ] (Fig. 3). In the case of K222D, the fluorescence change is Ͻ2% relative to the base line, even though the mutant experiences up to 5-fold increases in the kinetic parameters of cleavage of FPF under identical solution conditions. This suggests that binding of Na ϩ to the mutant is not linked to significant changes in the environment of Trp residues, unlike what is observed with the human enzyme (33). Trp-215 may be locked in a rigid conformation in murine thrombin, or concomitant changes at other Trp residues may cancel out the effects seen in the human enzyme.
The ability of murine thrombin to retain high catalytic activity in the absence of Na ϩ activation is demonstrated by cleavage of Spectrozyme-TH (Fig. 1) and especially by the hydrolysis of FPF (Fig. 2). Functional mimicry of the fast form of human thrombin by the murine enzyme is supported by its interaction with physiologic substrates (Table 1). Under physiologic conditions of salt, pH, and temperature, cleavage of fibrinogen occurs with a value of k cat /K m ϭ 27 M Ϫ1 s Ϫ1 , which is higher than the value measured for the human enzyme and very close to the value of 31 M Ϫ1 s Ϫ1 measured for the fast form (30). Cleavage of protein C under saturating concentrations of thrombomodulin occurs with a value of k cat /K m ϭ 0.38 M Ϫ1 s Ϫ1 , which is again higher than the value of 0.22 M Ϫ1 s Ϫ1 measured for the human enzyme (35). Cleavage of PAR1 and PAR4 occur with k cat /K m values of 120 and 8.9 M Ϫ1 s Ϫ1 , which are, respectively, 4-and 26-fold higher than those reported for the human enzyme and exceed those pertaining to the fast form (34). The preferential cleavage of PAR1 relative to PAR4 is almost 100-fold for the human enzyme but is reduced to 13-fold in murine thrombin. Interestingly, the same preference is retained in the cleavage of human PAR1 and PAR4 by murine thrombin, although the rates are significantly slower than those measured for the murine substrates. On the other hand, cleavage of murine PAR1 and PAR4 by human thrombin occurs with rates comparable with those of the human substrates (Table 1).
In principle, the lack of Na ϩ effect on murine thrombin could be due to the lack of Na ϩ binding (K a ϭ 0 in Equation 1) or to the presence of Na ϩ binding not linked to transduction of this event into enhanced catalytic activity (s 0 ϭ s 1 in Equation 1). We favor the first scenario,  because mutation of Asp-222 in thrombin abolishes Na ϩ binding (9) and the crystal structure of the D221A/D222K mutant shows no evidence of bound Na ϩ (16). Factor Xa carries Lys-222-like murine thrombin but is a Na ϩ -activated enzyme (17). This property, however, is likely due to the different composition of the 186-loop. In fact, Ala mutations in this loop compromise or abrogate Na ϩ binding in factor Xa (40), although they have no such effect in thrombin (11).
The activity of murine thrombin toward FPF and physiologic substrates is the same or higher than that of the human enzyme, which suggests that murine thrombin is stabilized in a conformation similar to that of the fast form even though it is devoid of Na ϩ binding. Essentially, murine thrombin has constitutively replaced Na ϩ activation. This reinforces the importance of Na ϩ activation in thrombin and proves that an enzyme devoid of Na ϩ binding and stabilized in the low activity slow form would be incompatible with physiologic function. We propose that the constitutive replacement of Na ϩ activation in murine thrombin has an important evolutionary advantage. Several mutations of human thrombin, e.g. Frankfurt (41), Salakta (42), Greenville (43), and Scranton (44), occur naturally at residues important for Na ϩ recognition and cause bleeding. It is therefore possible that pressure to constitutively replace Na ϩ activation arose in the mouse to counter the effect of such mutations or of more disruptive ones. To test this hypothesis, and to complement ongoing transgenic studies by Dr. Jay Degen, we constructed the murine W215A/E217A mutant. The W215A/E217A mutant of human thrombin is the most striking anticoagulant mutant engineered to date; the mutation compromises selectively fibrinogen and PAR1 cleavage but leaves protein C activation in the presence of thrombomodulin almost intact (45). The anticoagulant effect of the mutation is seen most convincingly in vivo (46) and has been explained structurally in terms of the total collapse of the locale for fibrinogen and PAR1 recognition (47). The murine W215A/E217A mutant has largely (Ͼ1,000-fold) compromised fibrinogen cleavage, but the reduction is almost 20-fold less pronounced than in the human variant ( Table 1). Cleavage of PAR4 is compromised 600-fold. Notably, cleavage of PAR1 is reduced only 60-fold as opposed to Ͼ1,000-fold in the human enzyme. On the other hand, cleavage of protein C is reduced Ͻ7-fold, just as for the human variant. Mutation of Trp-215 and Glu-217 affects the activity of murine thrombin toward fibrinogen, PAR1 and PAR4 because these residues make direct contacts with substrate (31,33,34). However, unlike the human enzyme, murine thrombin is spared an additional ϳ20-fold drop in activity caused by stabilization of the Na ϩfree slow form upon the Ala replacement of Trp-215 and Glu-217 (9). Because of the constitutive Na ϩ activation in murine thrombin, the W215A/E217A mutation is significantly less anticoagulant in the mouse than in humans.

DISCUSSION
Murine thrombin lacks Na ϩ activation, in sharp contrast to the human enzyme. Although the enzyme does not bind Na ϩ , it is locked in a conformation that resembles functionally the fast form of the human enzyme. Molecular mimicry of Na ϩ activation has general relevance to enzymes activated by monovalent cations, and its elucidation can facilitate the rational engineering of more proficient proteases (3). The task is nontrivial, as demonstrated by attempts to mimic Na ϩ or K ϩ activation in thrombin (48), pyruvate kinase (49), and the molecular chaperone Hsc70 (50). In all cases the ability of the enzyme to bind and discriminate among monovalent cations was successfully replaced by a Lys side chain. However, in no case was the replacement associated with high catalytic activity. In the case of murine thrombin, the replacement of Asp-222 with Lys and possibly other changes in the 186-loop lining the Na ϩ pore (Fig. 4) resulted in efficient mimicry of Na ϩ activation and full catalytic activity toward physiologic substrates.
There is an evolutionary advantage to constitutively replacing Na ϩ activation for optimal catalytic activity if pressure exists to mutate residues important for Na ϩ binding. Evidence has been presented that the evolutionary divergence of serine proteases can be explained fully in terms of the C-terminal sequence of the catalytic domain (51). This is  1SFQ (9). The residues within 10 Å of the bound Na ϩ that are different in the murine enzyme are marked. In murine thrombin, residue 222 is Lys, and the corresponding sequence in the 186-loop is VNDTKR. There is enough variation between the two species in this critical region of the enzyme to expect differences in Na ϩ binding and activation.

for the hydrolysis of physiologic substrates by wild-type and mutant thrombins
Experimental conditions: 5 mM Tris, 0.1% PEG, 145 mM NaCl, pH 7.4, at 37°C. Human and murine substrates were used for human and murine thrombin, respectively, unless noted otherwise. The values in parentheses refer to the slow (͓Na ϩ ͔ ϭ 0) and fast (͓Na ϩ ͔ ϭ ∞) forms of human wild type (WT) under the same solution conditions. RAP denotes the relative anticoagulant potency, calculated as the ratio of k cat /K m for protein C activation relative to that of fibrinogen, in units of wild-type values. The W215A/E217A mutant is significantly less anticoagulative in mouse than in human. ND, not detectable, i.e. too low to measure.

Substrate
Human WT W215A/E217A Murine WT W215A/E217A the sequence in which pressure to mutate is the highest and, coincidentally, where all residues important for Na ϩ binding are located. It is therefore possible that murine thrombin endorses molecular mimicry of Na ϩ activation to counter the effect of mutations affecting Na ϩ binding that would compromise the ability to cleave fibrinogen or the PARs. Murine thrombin cleaves fibrinogen and protein C with rate constants comparable with those of the human enzyme-substrate interactions, but important differences have emerged from the results on PAR1 and PAR4. PAR1 is responsible for platelet activation in humans at low thrombin concentrations, and its action is reinforced by PAR4 at higher concentrations of enzyme (19). In the mouse, signaling in platelets is mediated entirely by PAR4, with PAR3 facilitating PAR4 cleavage at low thrombin concentrations (52,53). PAR1 is by far the best physiologic substrate for murine thrombin, its k cat /K m being almost 6-fold higher than that of fibrinogen and 4-fold higher than the analogous interaction in humans. PAR4 has a k cat /K m 13-fold lower than that of PAR1 in the mouse, which contrasts the 100-fold preference of PAR1 versus PAR4 seen in the human enzyme. This is because the thrombin-PAR4 interaction is 26-fold more specific in the mouse than human. These findings are consistent with PAR4 being the major signaling pathway in mouse platelets (19,54). The relatively low k cat /K m for PAR4 cleavage by human thrombin explains the ancillary role of this receptor in human platelet activation that is mostly dependent on PAR1 (19,54). A value of k cat /K m for thrombin-PAR4 interaction in the mouse similar to that seen in humans would be incompatible with the physiologic requirements of platelet activation. That is presumably the reason why the thrombin-PAR4 interaction in the mouse occurs with a k cat /K m value similar to that of the thrombin-PAR1 interaction in humans.
A potentially important property of murine thrombin, also shared by human thrombin, is the ability to cleave chymotrypsin-specific substrates such as FPF. Although thrombin closely resembles trypsin in its three-dimensional architecture, it differs significantly in the accessibility and hydrophobicity of the S1 pocket. Heparin cofactor II carries Leu-444 in the P1 position, which penetrates the S1 pocket of thrombin and effectively inactivates the enzyme, although the L444R substitution produces more potent inhibition (55,56). The S1 pocket of factor Xa can also accept hydrophobic residues (57), consistent with the data reported in Fig. 2. Medicinal chemists have long recognized that the S1 pocket of thrombin is wider than that of trypsin and can accommodate bulky hydrophobic groups (58,59), a feature that has been exploited to achieve selective inhibition of thrombin over trypsin (60,61). Recently, a crystal structure of thrombin complexed with a synthetic inhibitor has documented the mode of interaction of a Tyr side chain within the primary specificity pocket (62). The hydroxyl group of Tyr engages the carboxylate of Asp-189 in favorable polar interactions that are both direct and mediated by a water molecule. Hence, thrombin possesses the necessary structural requisites to effectively cleave substrates carrying Phe or Tyr at P1. The physiologic importance of this dual trypsin-like and chymotrypsin-like specificity of murine and human thrombin is presently unclear and requires further investigation. The preference of thrombin for P1 residues may be influenced by cofactors or by interaction with membranes or extracellular matrix components, which opens up the possibility that the environment in which thrombin acts may change the repertoire of substrates available for cleavage and may include targets that remain to be identified.