Two Acidic, Anticoagulant PLA2 Isoenzymes Purified from the Venom of Monocled Cobra Naja kaouthia Exhibit Different Potency to Inhibit Thrombin and Factor Xa via Phospholipids Independent, Non-Enzymatic Mechanism

Background The monocled cobra (Naja kaouthia) is responsible for snakebite fatality in Indian subcontinent and in south-western China. Phospholipase A2 (PLA2; EC 3.1.1.4) is one of the toxic components of snake venom. The present study explores the mechanism and rationale(s) for the differences in anticoagulant potency of two acidic PLA2 isoenzymes, Nk-PLA2α (13463.91 Da) and Nk-PLA2β (13282.38 Da) purified from the venom of N. kaouthia. Principal Findings By LC-MS/MS analysis, these PLA2s showed highest similarity (23.5% sequence coverage) with PLA2 III isolated from monocled cobra venom. The catalytic activity of Nk-PLA2β exceeds that of Nk-PLA2α. Heparin differentially regulated the catalytic and anticoagulant activities of these Nk-PLA2 isoenzymes. The anticoagulant potency of Nk-PLA2α was comparable to commercial anticoagulants warfarin, and heparin/antithrombin-III albeit Nk-PLA2β demonstrated highest anticoagulant activity. The anticoagulant action of these PLA2s was partially contributed by a small but specific hydrolysis of plasma phospholipids. The strong anticoagulant effect of Nk-PLA2α and Nk-PLA2β was achieved via preferential, non-enzymatic inhibition of FXa (Ki = 43 nM) and thrombin (Ki = 8.3 nM), respectively. Kinetics study suggests that the Nk-PLA2 isoenzymes inhibit their “pharmacological target(s)” by uncompetitive mechanism without the requirement of phospholipids/Ca2+. The anticoagulant potency of Nk-PLA2β which is higher than that of Nk-PLA2α is corroborated by its superior catalytic activity, its higher capacity for binding to phosphatidylcholine, and its greater strength of thrombin inhibition. These PLA2 isoenzymes thus have evolved to affect haemostasis by different mechanisms. The Nk-PLA2β partially inhibited the thrombin-induced aggregation of mammalian platelets suggesting its therapeutic application in the prevention of unwanted clot formation. Conclusion/Significance In order to develop peptide-based superior anticoagulant therapeutics, future application of Nk-PLA2α and Nk-PLA2β for the treatment and/or prevention of cardiovascular disorders are proposed.


Introduction
The Indian monocled cobra (Naja kaouthia) is one of the major venomous snakes of the Elapidae family. This species of snake is responsible for major snakebite fatality in India, Bangladesh, Nepal, Myanmar, and south-western China [1]. Snakebite remains a neglected tropical disease which requires immediate medical attention. The venom of N. kaouthia is enriched with various enzymes and non-enzymatic protein/peptide toxins [1]. Among the snake venom enzymes, PLA 2 s, because they have a crucial role in inducing various pharmacological effects on snakebite victims, are an interesting group of proteins [2,3]. Snake venom PLA 2 enzymes exhibit a wide variety of pharmacological effects such as neurotoxicity, cardiotoxicity, cytotoxicity, membrane damaging activity, myotoxicity, necrotic, anticoagulant, hemolytic, haemorrhage and edema induction by various mechanisms of action [2][3][4][5][6][7][8][9][10][11]. The structure-function relationship and the mechanism of action of this group of small proteins are still subtle, complex and thus impose intriguing challenges to the toxinologists [2,12].
The systemic neuromuscular blockade and other clinical signs of N. kaouthia envenomation have been described. However, the mechanism(s) of anticoagulant action of snake venom PLA 2 enzymes still remains incomprehensible and a detailed molecular mechanism of anticoagulant action of PLA 2 enzymes from N. kaouthia venom has never been explored. Our previous studies have shown the biochemical and pharmacological characterization as well as membrane damaging activities of major PLA 2 isoenzymes purified from the venom of N. kaouthia of different geographical origins [5][6][7]. These Nk-PLA 2 s were found to prolong the Ca-clotting time of citrated platelet poor plasma [7]; however, their mechanism of anticoagulant action was not explored. Snake venom PLA 2 enzymes may prolong the blood coagulation in victims by different ways; either by enzymatic hydrolysis of plasma phospholipids which is required to initiate the coagulation process and/or binding with essential blood coagulation factors through a non-enzymatic mechanism [2][3][8][9]. Studies on the mechanism of such anticoagulants will advance our understanding of 'vulnerable' sites in the coagulation cascade [12]. This study may further help us to design novel strategies towards developing anticoagulant therapeutics for treating cardiovascular disorders (CVDs) and also new functional diagnostic test kits in the field of haemostasis [2,10,13].
In the present study, we report the purification and the mechanism of anticoagulant action of two major PLA 2 enzymes purified from the venom of N. kaouthia of eastern India origin.

Purification of two PLA 2 isoenzymes
All the fractionation steps were carried out at 4uC. Twenty five mg (dried weight) of N. kaouthia crude venom, dissolved in 0.5 ml of 20 mM Tris-HCl, pH 7.4, was injected to a FPLC-HiPrep CM FF16/10 column (20 ml) pre-equilibrated with 20 mM Tris-HCl, pH 7.4 (buffer A). The column was coupled to a Fast Protein Liquid Chromatography (FPLC) system (Akta Purifier 10, Wipro GE Healthcare). After washing the column with 30 ml (1.5 CV) of equilibration buffer, the bound proteins were eluted from 0 to 50% linear gradient of buffer B (20 mM Tris-HCl, pH 7.4 containing 1M NaCl) for 100 min at a flow rate of 1.0 ml/min. One ml fraction was collected in each tube. The elution of protein was monitored at 280 nm and the peaks were then screened for protein content [14], PLA 2 (see below), and plasma clotting activities (see below).
The cation-exchange unbound fractions showing maximum anticoagulant as well as PLA 2 activities were pooled, desalted, lyophilized, dissolved in 0.5 ml of 20 mM Tris-HCl, pH 7.4 and was then applied to an anion exchange FPLC Hiprep DEAE FF16/10 column (20 ml) pre-equilibrated with the above buffer. After washing the column with 30 ml of equilibration buffer (1.5 CV), bound proteins were eluted from 0 to 50% linear gradient of buffer B (20 mM Tris-HCl, pH 7.4 containing 1M NaCl) for 100 min at a flow rate of 1.0 ml/min. One ml fraction was collected in each tube. The elution of protein was monitored at 280 nm, and the peaks were then screened for protein content [14], PLA 2 , and plasma clotting activities. The purity and molecular mass of 1.0 mg protein of pooled fractions showing high PLA 2 and anticoagulant activities were determined by MALDI-TOF-MS (4800 Plus MDS SCIEX, Applied Biosystems) as described previously [3,15].

Protein identification by LC-MS/MS
Twenty five microgram of the lyophilized sample after reduction and alkylation was subjected to in-solution digestion with proteomics grade trypsin (Promega) for overnight at 37uC [15]. The digested peptides were reconstituted in 15 ml of the 0.1% (v/v) formic acid and were subjected to RP-HPLC-MS/MS analysis with collision induced dissociation as the fragmentation method. The data so generated was searched for the identity of tryptic peptides using MASCOT 2.4 search engine on a Proteome discoverer 1.4 platform. The data was searched against Uniprot Swiss-Prot database (non redundant database with reviewed proteins) and Uniprot TrEMBL (database with unreviewed proteins) database against the taxid snake from NCBI. Minimum of two high confident peptides was used as a prerequisite to identify the protein(s) of interest.

PLA 2 activity assay
The PLA 2 activity was assayed as described previously [5,8] by using egg-yolk phospholipids as substrate. One unit of PLA 2 activity has been defined as the amount of protein which produces a decrease in 0.01 absorbance after 10 min at 740 nm. The substrate specificity of purified PLA 2 was determined by using different commercially available phospholipids (1 mM) viz. phosphatidylcholine (PC), phosphatidylserine (PS) and phosphatidylethanolamine (PE). The enzyme activity was assayed by titrametric method with palmitic acid as a free fatty acid standard [6,8]. The PLA 2 activity is expressed as ng of fatty acid released/min/mg of protein.

Anticoagulant activity assay
Platelet poor plasma (PPP) was used for anticoagulant activity assay [3,8]. One unit of anticoagulant activity has been arbitrarily defined as one second increase in clotting of PLA 2 -treated PPP compared to the clotting time of control PPP [5,8]. To determine the dose-dependent anticoagulant activity different amounts (0.1 to 1.0 mg) of Nk-PLA 2 were incubated with PPP (300 ml) for 3 min at 37uC followed by an addition of 40 ml of 250 mM CaCl 2 in order to initiate the clot formation. As a positive control, the anticoagulant activity of heparin and warfarin was also determined under identical experimental conditions. The effect of preincubation on anticoagulant activity was determined by preincubating 300 ml PPP with 0.5 mg PLA 2 for 5-30 min at 37uC and then the anticoagulant activity was determined as stated above.
To determine the release of PLA 2 -induced free fatty acids from PPP, 0.5 mg PLA 2 (dissolved in 20 mM Tris-HCl, pH 7.4) was pre-incubated with PPP (300 ml) for 3-30 min at 37uC, and thus the liberated fatty acids were extracted and micro titrated by using 0.01 N NaOH and phenolphthalein red as indicator [8]. A control (PPP treated with buffer) was also run in parallel.The fatty acids liberated from the PPP (treated with PLA 2 or buffer) were extracted and quantified as described by Doley et al. [6].

Thrombin and Factor Xa inhibition assay
To observe the effect of purified PLA 2 isoenzymes on amidolytic activity of thrombin or FXa, purified PLA 2 (1.0 mg in 20 mM Tris-HCl, pH 7.4) was pre-incubated with thrombin (0.03 NIH U/ml) or FXa (0.15 mM) for 30 min at 37uC and then by assaying the activity of these coagulation factors against their chromogenic substrates (0.2 mM) T1637 and F3301, respectively [3,8]. The release of p-nitroaniline (pNA) was monitored for 15 min at intervals of 1 min at 405 nm in a plate reader (Multiskan TM GO, Thermo Scientific, USA). For every experiment, a control was run in parallel. The activity of thrombin or FXa towards their substrates was considered 100% activity and other values were compared with this.
The dose-dependent (0-0.2 mg) and time-dependent (5-30 min) antithrombin activity of purified PLA 2 against fibrinogen (7.5 mM), which is the physiological substrate for thrombin, was also determined as described previously [3]. In another set of experiments, the FXa inhibition assay was performed by preincubating 20 nM FXa with 1.0 mg of PLA 2 for 60 min at 37uC and then the reaction was initiated by adding prothrombin (1.4 mM) to this mixture. The prothrombin activation was determined by formation of thrombin using its chromogenic substrate T1637 [8]. Thrombin formation was also determined by 15% SDS-PAGE analysis of prothrombin activation products after 1 h incubation of prothrombin (10.0 mg) with 20 nM FXa (pretreated with PLA 2 or buffer) at 37uC, pH 7.4. The formation of thrombin was confirmed by the peptide mass fingerprinting analysis of ,36 kDa protein. For each experiment, a control was also run in parallel where buffer instead of PLA 2 was added.

Determination of inhibitory constant (K i ) for the inhibition of FXa and thrombin
For determining the inhibitory constant (Ki value) of Nk-PLA 2 a on amidolytic activity of FXa, a fixed concentration of FXa (20 nM) was pre-incubated with two different concentrations of Nk-PLA 2 a (50.0 nM and 100.0 nM) in a 96-well microplate at 37uC for 60 min. A control was run in parallel where FXa was incubated with buffer (20 mM Tris-HCl, pH 7.4) under identical experimental conditions. Then graded concentrations (0.1-0.8 mM) of chromogenic substrate for factor Xa (F3301) were added and the final volume of reaction was adjusted to 100 ml with 20 mM Tris-HCl, pH 7.4. After incubation for 10 min, the release of pNA was determined at 405 nM. For kinetic analysis, the reaction rate (V) was plotted against substrate concentrations (S) at each inhibitor concentration and the data was fitted to a hyperbolic Michaelis-Menten model using GraphPad Prism 5.0 software. The inhibitory constant (Ki) was determined by using the uncompetitive model (shown below) for enzyme inhibition by using the same software.
VApp~V max KmApp~K m In the above equations (1) and (2), I indicates the inhibitor concentration; VApp and Vmax represent maximum velocity in the presence and absence of the inhibitor, respectively. The KmApp and Km in equation 2 denote the Michaelis constant in the presence and absence of inhibitor, respectively.
The Ki value for inhibition of amidolytic activity of thrombin by Nk-PLA 2 b was determined by pre-incubating a fixed concentration of thrombin (0.03 NIH U/ml) with two different concentrations of Nk-PLA 2 b (50.0 nM and 100.0 nM) at 37uC, pH 7.4 for 30 min. In control, thrombin was incubated with buffer (20 mM Tris-HCl, pH 7.4) instead of Nk-PLA 2 b under identical experimental conditions. The reaction was initiated by adding graded concentrations (0.1-0.8 mM) of chromogenic substrate for thrombin (T1637) in a final reaction volume of 100 ml adjusted with 20 mM Tris-HCl, pH 7.4. After incubation for 10 min at 37uC, the release of pNA was determined at 405 nM in a microplate reader. The kinetics analysis was done by using the GraphPad Prism 5.0 software and the Ki value was determined as stated above. The experiments were repeated three times to assure the reproducibility.

Spectrofluorometric assay of interaction of PLA 2 with thrombin/FXa and PC
To study the protein-protein interaction, thrombin/FXa was pre-incubated with NK-PLA 2 (1:10 molar ratio) for 20 min at room temperature and the fluorescence spectra were acquired as described previously [3,8,10]. As a control, the fluorescence spectra of individual proteins were also determined. For determination of interaction of Nk-PLA 2 with PC in presence of 0.5 mM EDTA (to prevent the PC hydrolysis), our previously described procedure was followed [3]. The above experiments were repeated three times to assure the reproducibility.

Platelet aggregation inhibition assay
For preparation of platelet rich plasma, citrated blood (1:9, v/v) was centrifuged at 2006 g for 20 min. The in vitro platelet aggregation inhibition property was investigated by pre-incubating different concentrations (0.5 to 2.0 mg) of purified PLA 2 with thrombin (1 NIH U/ml) for 30 min at 37uC followed by induction of platelet aggregation. A control reaction was also set up where instead of purified Nk-PLA 2 , thrombin (1 NIH U/ml) was incubated with 16 PBS (20 mM K-phosphate, 150 mM NaCl, pH 7.4) at 37uC. The platelet aggregation was monitored in a Chrono-log dual channel aggregometer for 10 min. The experiment was repeated three times to assure the reproducibility.

Effect of histidine modifier, heparin and commercial antivenom on enzymatic and anticoagulant activities
The PLA 2 inhibitor (histidine modifier) p-bromophenacyl bromide (5 mM), heparin (1 IU) or graded amounts of monovalent antivenom (against N. kaouthia venom) were incubated with a fixed concentration of PLA 2 (1.0 mg/ml) at 37uC for 30 min. The catalytic, anticoagulant, and thrombin/FXa inhibitory activity of PLA 2 was then assayed as stated above. The activity of PLA 2 in presence of its inhibitor/antivenom was thus compared with control (without inhibitor/antivenom).

Statistical analysis
Results were presented as mean 6 standard deviation (S.D.) of three experiments. Statistical analysis of the data was performed using a Student's t test in Sigma Plot 11 for Windows (version 7.0). Values of p#0.05 were considered significant.

Results
Isolation and purification of two PLA 2 isoenzymes from N. kaouthia venom Fractionation of crude N. kaouthia venom through Hiprep CM FF 16/10 column separated the proteins into 6 peaks, Nk(H)CEXP1 to Nk(H)CEXP6 (Fig. 1a).The peak 1 [Nk(H)CEXP1] eluted with the equilibration buffer showed a significant PLA 2 as well as anticoagulant activity. By re-fractionation of this peak through Hiprep DEAE FF16/10 column, the proteins were separated into 8 fractions (Fig. 1b). The peak Nk(H)AEXP4 and Nk(H)AEXP5 demonstrated significant PLA 2 as well as anticoagulant activity. The proteins of these peaks were found to be homogenous through MALDI-TOF-MS analysis with molecular mass of 13463.91 Da and 13282.38 Da, respectively (Figs. 1c-d). A summary of purification of these purified anticoagulant PLA 2 s, named Nk-PLA 2 a and Nk-PLA 2 b is shown in Table 1.

LC-MS/MS identification of proteins
By LC-MS/MS analysis, both the PLA 2 s under study showed highest similarity (score 463) with PLA 2 (EC 3.1.1.4) III isolated from monocled cobra venom (accession no. 2144440) with 23.5% sequence coverage. Furthermore, the PLA 2 s under study also demonstrated similarity (18-15% sequence coverage) with PLA 2 s isolated from Naja naja (accession no. 395192), anticoagulant PLA 2 from N. n. sagittifera (accession no. 66361231), and an acidic PLA 2 natratoxin from N. atra (accession no. 209573225) venom samples. Besides, both these PLA 2 s were found to possess identical sites for trypsin cleavage and therefore, they demonstrated similar trypic fragmentation pattern. The NCBI BLASTP search of one of the tryptic peptides GGSGTPVDDLDR (Mr. 1188.5 Da) showed putative conserved domains of PLA 2 -like superfamily. Another tryptic peptide sequence NMIQCTVPSR (Mr. 1148.5 Da) demonstrated 100% sequence similarity with an acidic PLA 2 enzymes natratoxin (accession no. A4S04) isolated from the Chinese cobra N. atra venom. Furthermore, another tryptic peptide sequence viz. NLYQFK (Mr. 812.4 Da) was found to contain the N-terminal sequence of both these PLA 2 isoezymes purified from N. kaouthia venom.

Substrate specificity, anticoagulant activity and inhibition of platelet aggregation
Both NK-PLA 2 s showed preferential hydrolysis of PC over PS or PE (data not shown), which is similar to the phospholipid hydrolysis pattern of PLA 2 s, which we previously isolated from the venom of N. kaouthia [5][6][7]. The specific activity of Nk-PLA 2 a and Nk-PLA 2 b towards PC was determined as 2.3610 5 and 4.5610 5 units/mg, respectively. Both enzymes displayed optimum activity at the pH range of 8.0-9.0 and at 37-45uC (data not shown). Like other snake venom PLA 2 enzymes [6][7][8], heating Nk-PLA 2 a or Nk-PLA 2 b at 75uC for 30 min did not result in any significant decrease in catalytic activity of heated enzyme as compared to control (unheated) enzyme (data not shown).
A comparison of anticoagulant activity of the purified PLA 2 s with commercial anticoagulants heparin and warfarin is shown in Fig. 2a. Both these PLA 2 s increased the Ca-clotting time of PPP in a dose-dependent manner, however to a significantly different extent (Fig. 2a). The anticoagulant activity of Nk-PLA 2 b was found to be significantly higher (p,0.05) compared to Nk-PLA 2 a, heparin or warfarin under identical experimental conditions. It is worthy of mention that heparin shows anticoagulant effect by binding with antithrombin-III (AT-III) present in PPP, which in turn inhibits thrombin and activated factor X [16]. Pre-incubation of Nk-PLA 2 a (0.2 mg) and Nk-PLA 2 b (0.2 mg) at a ratio of 1:1 (w:w) for 20 min at room temperature prior to the addition of PPP did not allow the plasma to clot within 10 min. Conversely, the anticoagulant activity of individual Nk-PLA 2 a or Nk-PLA 2 b under identical experimental conditions was found to be much lower (Fig. 2a).
Increasing of the pre-incubation of PPP with Nk-PLA 2 a or Nk-PLA 2 b for 0-5 min before addition of CaCl 2 enhanced the anticoagulant potency of both the enzymes (Fig. 2b). However, increasing of the pre-incubation of Nk-PLA 2 a or Nk-PLA 2 b with PPP from 5 to 30 min did not reveal any significant difference (p. 0.05) in the release of fatty acids from plasma phospholipids (Fig. 2c). This result indicates that only a minor plasma phospholipids hydrolysis done by Nk-PLA 2 isoenzymes is sufficient to exert their anticoagulant effect [17]. The Nk-PLA 2 a did not show inhibition of thrombin-induced platelet aggregation at a dose of 2.0 mg (data not shown). Conversely, Nk-PLA 2 b demonstrated dose-dependent partial inhibition of thrombin-induced platelet aggregation and saturation was observed at 2.0 mg PLA 2 (Fig. 3).

Inhibition of thrombin and FXa
None of the Nk-PLA 2 s showed amidolytic activity against the chromogenic substrate for FXa or thrombin. The Nk-PLA 2 a at a dose of 1.0 mg completely inhibited the amidolytic activity of FXa (0.15 mM); conversely, the Nk-PLA 2 b under identical experimen-  tal conditions displayed partial inhibition to FXa (Fig. 4a). The Michaelis-Menten plot showed (Fig. 4b) that Nk-PLA 2 a decreased in a dose-dependent manner, the Km as well as the Vmax values of factor Xa towards its chromogenic substrate ( Table 2). The turnover number (Kcat value) was also found to decrease with an increase in inhibitor (Nk-PLA 2 a) concentration ( Table 2). The Ki value for the inhibition of amidolytic activity of factor Xa by Nk-PLA 2 a was calculated at 42.1 nM. Furthermore, Nk-PLA 2 a was found to display a greater strength than Nk-PLA 2 b in inhibiting the prothrombin activation property of FXa (Fig. 5a). The SDS-PAGE analysis (under reduced conditions) of prothrombin degradation products by FXa in presence or absence of Nk-PLA 2 a or Nk-PLA 2 b also supports the above observation (Fig. 5b). The Nk-PLA 2 isoenzymes differentially inhibited the amidolytic activity (Fig. 6a) and fibrinogen clotting activity of thrombin (Fig. 6b). The Michaelis-Menten plot for the inhibition of amidolytic activity of thrombin at two different concentrations of Nk-PLA 2 b is shown in Fig. 6c. The Nk-PLA 2 b dose-dependently decreased the Km, Vmax, and Kcat values of thrombin towards its chromogenic substrate ( Table 2). The K i value for thrombin inhibition displayed by Nk-PLA 2 b was determined at 9.4 nM. It was observed that under identical experimental conditions, the inhibition of fibrinogen clotting activity of thrombin done by Nk-PLA 2 b was more pronounced (p,0.01) compared to the same activity displayed by Nk-PLA 2 a (Fig. 6b). Increasing of the time (5-30 min) during which Nk-PLA 2 b was pre-incubated with thrombin (prior to addition of fibrinogen) linearly decreased the fibrinogen clotting activity (or increased the fibrinogen clotting time) of thrombin (Fig. 6d).

Spectrofluorometric analysis of binding of PLA 2 s with thrombin, FXa or PC
The interaction of FXa with Nk-PLA 2 a (at 1:10 ratio) resulted in an increase in the emission of fluorescence signal (measured using excitation at 280 nm) of Nk-PLA 2 a-FXa complex (Fig. 7a). However, the interaction of Nk-PLA 2 b with FXa decreased the fluorescence signal of Nk-PLA 2 b-FXa complex compared to the fluorescence signal produced by Nk-PLA 2 b (Fig. 7a). Moreover, the fluorescence signal at ,345 nm produced by Nk-PLA 2 b was found to be higher than that of Nk-PLA 2 a suggesting presence of more tryptophan residue in the former PLA 2 enzyme. The interaction of Nk-PLA 2 a or Nk-PLA 2 b with thrombin showed that the intrinsic fluorescence of Nk-PLA 2 b was significantly increased in the presence of thrombin (Fig. 7b); conversely, interaction of Nk-PLA 2 a with thrombin influenced the fluorescence signal only to a marginal extent (Fig. 7b).
The effect of interaction between Nk-PLA 2 a/Nk-PLA 2 b and PC is shown in Fig. 7c. The interaction between Nk-PLA 2 a and PC in the presence of EDTA (to prevent the PC hydrolysis) resulted in a small decrease in the emission intensity with respect to the emission strength of PC alone. However, interaction effect of Nk-PLA 2 b with PC was found to be more pronounced because a large decrease in emission intensity was observed (Fig. 7c).

Effect of histidine modifier (PLA 2 inhibitor) and antivenom on catalytic and anticoagulant activity
Like other snake venom PLA 2 enzymes, the catalytic as well as anticoagulant activity of Nk-PLA 2 a and Nk-PLA 2 b was drastically and simultaneously reduced by pBPB, a modifier of histidine ( Table 3). The monovalent antivenom was found to inhibit the catalytic and anticoagulant activity of Nk-PLA 2 a and Nk-PLA 2 b  to a significantly (p,0.01) different extent. However, at 1:10 antigen (PLA 2 ): antivenom (protein:protein) ratio the catalytic activity of both the enzymes were inhibited to 62 to 65% of their original activity whereas their anticoagulant activity was not affected (Table 3). At an antigen (PLA 2 ): antivenom ratio of 1:200, the monovalent antivenom also failed to neutralize the thrombin inhibitory activity of these PLA 2 enzymes (Table 3). Interestingly, heparin showed dissimilar effect in neutralizing the catalytic property of Nk-PLA 2 a and Nk-PLA 2 b isoenzymes; nevertheless, FXa or thrombin inhibitory properties of these PLA 2 isoenzymes in presence of heparin remained unaffected (Table 3).

Discussion
The LC-MS/MS analysis of both these PLA 2 isoenzymes has suggested their identity with PLA 2 enzymes isolated from cobra venom. The molecular mass of Nk-PLA 2 a (13,463.91 Da) and Nk-PLA 2 b (13,282.38 Da) is very close to the molecular mass of previously reported PLA 2 isoenzymes NK-PLA 2 -I (13.6 KDa) and Nk-PLA 2 -A (13,619.36 Da), and Nk-PLA 2 -II (13,346.19 Da) and Nk-PLA 2 -B (13,303.05 Da), isolated from venom samples of N. kaouthia of different geographical locations [5][6][7]. Moreover, the active sites of Nk-PLA 2 a and Nk-PLA 2 b, like classical PLA 2 enzymes from snake venom, also contain histidine which is considered as the most crucial residue for phospholipids hydrolysis [3,[5][6][7][8]10,18]. These combined data suggest that Nk-PLA 2 a and Nk-PLA 2 b are PLA 2 isoenzymes purified from N. kaouthia venom of eastern India origin.
The PLA 2 enzymes on the basis of their potencies to prolong the re-calcification time of PPP have been classified into weak or strong anticoagulant enzymes [2,19]. The anticoagulant strength of Nk-PLA 2 a or Nk-PLA 2 b categorized them to strong anticoagulant PLA 2 enzymes [2][3]8,10]. The significantly higher anticoagulant activity of Nk-PLA 2 b than that of Nk-PLA 2 a suggests that these PLA 2 isoenzymes have evolved to exert pharmacological effects by different mechanisms in snakebite victims. Recently Saul et al [20] have also reported the presence of two PLA 2 isoforms in venom of Vipera ammodytes ammodytes, which differ in sequence by only two amino acid residues (Phe 124 is replaced by Ile 124 and Lys 128 is replaced by Glu 128 ). These natural PLA 2 isoforms exhibit significant differences in toxicity and anticoagulant potencies [20]. Comparison of primary structures of both the Nk-PLA 2 s, which is our next goal of study, may shed light on their structural motifs correlated to the differences in their biological activity.
Snake venom PLA 2 s enzymes most likely prolong the blood coagulation through hydrolysis of pro-coagulant phospholipids of plasma rendering them unavailable for initiating the coagulation process [2][3]8,10]. The role of catalytic activity in anticoagulant Table 2. Kinetics of inhibition of FXa (0.15 mM) and thrombin (0.03 NIH U/ml) by Nk-PLA 2 a and Nk-PLA 2 b at 37uC, pH 7.4, respectively.

Parameters
Factor Xa inhibition Thrombin inhibition  mechanism is evident from the fact that Nk-PLA 2 b compared to Nk-PLA 2 a demonstrating superior PC binding as well as PC hydrolysis property also shows higher anticoagulant potency [2][3]8]. Furthermore, inhibition study with heparin and pBPB also supports the role of catalytic activity in anticoagulant property of NK-PLA 2 isoenzymes. It has been mentioned that very low but specific plasma phospholipids hydrolysis is the characteristics feature of strongly anticoagulant PLA 2 s whereas non specific, nonanticoagulant PLA 2 enzymes hydrolyze the plasma phospholipids indiscriminately and therefore, they fail to retard the blood coagulation [2][3]10,17]. Furthermore, these results are in accordance with our previous observations showing strong anticoagulant PLA 2 enzymes affect blood coagulation by mechanism that is independent of phospholipids hydrolysis [3,8]. Differential inhibition of catalytic activity and anticoagulant property (pharmacological activity) of Nk-PLA 2 a and Nk-PLA 2 b by monovalent antivenom, pBPB (histidine modifier, PLA 2 inhibitor) and heparin suggests that there are separate or perhaps overlapping sites in these PLA 2 molecules that contribute to these activities [2][3]21]. Studies have indicated that heparin acts as a competitive inhibitor for human secretory class II PLA 2 and closely related snake venom PLA 2 enzymes [22]. The non-enzymatic mechanism of anticoagulant action of snake venom PLA 2 has been executed by competing with blood clotting factors such as thrombin, FXa, Va, or prothrombinase complex in the lipid surface [2][3]8,[11][12]20]. The present study strongly indicates that Nk-PLA 2 a or Nk-PLA 2 b prolongs the blood coagulation by targeting different coagulation factors. The anticoagulant mechanism of Nk-PLA 2 b is partially dependent on FXa inhibition suggesting FXa is the major pharmacological target for Nk-PLA 2 a to disrupt the coagulation cascade through a non-catalytic, phospholipids independent mechanism. This significant difference in potency of FXa inhibition by two PLA 2 isoenzymes purified from the same snake venom is in accordance with the report of Saul et al. [20]. It has been shown that a point mutation of Lys 128 to Glu 128 perturbs the interaction of PLA 2 isoform with FXa results in a 10-fold decrease in the affinity of mutated enzyme for FXa [20]. Therefore, it may be anticipated that small surface residue changes in snake venom isoenzymes can give rise to very significant changes in their biological activity [15,20,23]. The kinetics of inhibition suggests that the Nk-PLA 2 a binds to the enzyme (FXa)-substrate complex (uncompetitive mode of inhibition) rather than competing with the substrate for binding at the active site of the enzyme. In contrast, strongly anticoagulant PLA 2 enzyme CM-IV purified from N. nigricollis venom is a non-competitive inhibitor of FXa [24]. Furthermore, inhibition of FXa by Nk-PLA 2 a is more pronounced compared to CM-IV (Ki = 250 nM) [24]. Moreover, the mechanism of FXa inhibition by Nk-PLA 2 a or Nk-PLA 2 b and human group II secretory PLA 2 (hsPLA 2 ) seems to be different because hsPLA 2 does not inhibit the amidolytic activity of FXa [25].
The Nk-PLA 2 b exerts potent anticoagulant activity via targeting thrombin without the requirement of any cofactor. Till date, only two snake venom PLA 2 enzymes, one purified from venom of N. haje [11] and the other from the venom of Daboia russelii russelii [3], have been demonstrated to prolong PPP clotting via thrombin inhibition. The Ki value indicates that Nk-PLA 2 b is a more potent inhibitor of thrombin compared to the previously reported thrombin-inhibitor PLA 2 s isolated from snake venoms [3,11]. Furthermore, Nk-PLA 2 b inhibited thrombin by uncompetitive mechanism whereas TI-Nh, a PLA 2 purified from N. haje venom is a mixed-type inhibitor of thrombin [11]. Therefore, the mechanism and strength of thrombin inhibition by PLA 2 enzymes purified from venoms of N. haje and N. kaouthia are different. The spectrofluorometric data also vouch the interaction of Nk-PLA 2 a and Nk-PLA 2 b with FXa and thrombin [3,[8][9]. However,  the significantly different strength of binding of these PLA 2 isoenzymes to thrombin or FXa reinforces the fact that these PLA 2 isoenzymes have different pharmacological targets in order to exert their anticoagulant action [2]. Moreover, like RVAPLA 2 isolated from D. r. russelii venom [3], the binding of Nk-PLA 2 b to thrombin is not a very rapid event. In a sharp contrast, a Kunitztype protease inhibitor isolated from D. r. russelii venom binds to thrombin quickly [26]. The final stage of the blood coagulation cascade is catalyzed by thrombin, a 36 kDa trypsin-like serine protease, which is synthesized in the liver as its precursor prothrombin. Inactive prothrombin is proteolytically cleaved by the prothrombinase complex involving the activated factor X (FXa)/Va to form thrombin which is responsible for physiologic homeostasis and pathologic thrombosis [27]. Thrombin and FXa are, therefore, important pharmaceutical targets for the treatment and prevention of thrombotic disorder, a leading cause of cardiovascular death throughout the world [28][29]. Furthermore, platelet aggregation inhibitor aspirin prevents intravascular clot formation and it has therefore been used as prophylaxis to prevent myocardial infarction and stroke for more than 100 years now. However, several limitations of aspirin have been reported and the best dose of aspirin for the prevention of cardiovascular disease has not been well established [30]. The Nk-PLA 2 b by virtue of its thrombin inhibitory property partially prevented the thrombinmediated aggregation of human platelets which may have therapeutic implications for averting unwanted clot formation.
Despite their widespread use, the anticoagulant drugs heparin and warfarin are associated with significant complications [28][29].
This underscores the search for new, safe and potent anticoagulant drugs that do not depend on vitamin K antagonism [29]. Notably, the cardiovascular therapeutic area has been undergoing unprecedented changes with the clinical introduction of new drugs targeting thrombin, factor Xa and inhibition of platelet aggregation. The direct thrombin inhibitors compared to indirect thrombin inhibitor such as heparin may offer a more specific anticoagulant effect [26]. Furthermore, our previous studies had vouched the non-toxic nature of N. kaouthia PLA 2 isoenzymes [5,7]. Taken together, it may be anticipated that future therapeutic application of Nk-PLA 2 a and Nk-PLA 2 b as combined platelet aggregation inhibitors and anticoagulant agents for the treatment and/or prevention of cardiovascular disorder is highly promising [13,28].