The prothrombin-activating principle from echis carinatus venom. I. Preparation and biochemical properties

Beyond the 'big four': Venom profiling of the medically important yet neglected Indian snakes reveals disturbing antivenom deficiencies

Background

Snakebite in India causes the highest annual rates of death (46,000) and disability (140,000) than any other country. Antivenom is the mainstay treatment of snakebite, whose manufacturing protocols, in essence, have remained unchanged for over a century. In India, a polyvalent antivenom is produced for the treatment of envenomations from the so called ‘big four’ snakes: the spectacled cobra (Naja naja), common krait (Bungarus caeruleus), Russell’s viper (Daboia russelii), and saw-scaled viper (Echis carinatus). In addition to the ‘big four’, India is abode to many other species of venomous snakes that have the potential to inflict severe clinical or, even, lethal envenomations in their human bite victims. Unfortunately, specific antivenoms are not produced against these species and, instead, the ‘big four’ antivenom is routinely used for the treatment.

Methods

We characterized the venom compositions, biochemical and pharmacological activities and toxicity profiles (mouse model) of the major neglected yet medically important Indian snakes (E. c. sochureki, B. sindanus, B. fasciatus, and two populations of N. kaouthia) and their closest ‘big four’ congeners. By performing WHO recommended in vitro and in vivo preclinical assays, we evaluated the efficiencies of the commercially marketed Indian antivenoms in recognizing venoms and neutralizing envenomations by these neglected species.

Findings

As a consequence of dissimilar ecologies and diet, the medically important snakes investigated exhibited dramatic inter- and intraspecific differences in their venom profiles. Currently marketed antivenoms were found to exhibit poor dose efficacy and venom recognition potential against the ‘neglected many’. Premium Serums antivenom failed to neutralise bites from many of the neglected species and one of the ‘big four’ snakes (North Indian population of B. caeruleus).

Conclusions

This study unravels disturbing deficiencies in dose efficacy and neutralisation capabilities of the currently marketed Indian antivenoms, and emphasises the pressing need to develop region-specific snakebite therapy for the ‘neglected many’.

Snakebite is a ‘neglected tropical disease’ that majorly affects the rural populations in developing countries. India bears the brunt of snakebites with over 46,000 deaths and 140,000 disabilities, annually. A significant number of these bites are attributed to the widely distributed ‘big four’ snakes, namely spectacled cobra (Naja naja), common krait (Bungarus caeruleus), Russell’s viper (Daboia russelii), and saw-scaled viper (Echis carinatus). The commercial antivenoms marketed in India are only manufactured against these four species, while neglecting many other medically relevant snakes with restricted geographic distribution. Snakebite pathology is dependent on the venom composition of the population/species, which can, in turn, vary intra- and inter-specifically. Though this variation severely limits the cross-population/species antivenom efficacy, envenomations by the neglected snakes in India are treated with the ‘big four’ antivenom. Therefore, to unravel the underlying venom variability, we investigated venom proteomic, biochemical/pharmacological and toxicity profiles of the major neglected Indian snakes and their ‘big four’ relatives. To assess the effectiveness of the ‘big four’ antivenom in treating bites from these neglected snakes, we performed preclinical experiments, which revealed alarming inadequacies of the commercial antivenoms. Our findings accentuate the compelling necessity for the innovation of highly efficacious next-generation snakebite therapy in India.

Neutralisation of Local Haemorrhage Induced by the Saw-Scaled Viper Echis carinatus sochureki Venom Using Ethanolic Extract of Hibiscus aethiopicus L

The objective of the study is to investigate the anti-snake venom activities of a local plant, Hibiscus aethiopicus L. The H. aethiopicus was dried and extracted with ethanol. Different assays were performed according to standard techniques, to evaluate the plant's acute toxicity and its antivenom activities. The results of evaluating the systemic acute toxicity of the H. aethiopicus extract using “oral and intra-peritoneal” route were normal even at the highest dose (24 g/kg) tested. All guinea pigs (n = 3) when treated with venoms E. c. sochureki (75 μg) alone induced acute skin haemorrhage. In contrast, all guinea pigs (n = 18) treated with both venom and the plant extract at a concentration between 500 and 1000 mg/kg showed no signs of haemorrhage. Moreover, all guinea pigs (n = 18) treated with venom and the plant extract below 400 mg/kg showed acute skin haemorrhage. All guinea pigs treated with venom E. c. sochureki (75 μg) alone induced acute skin haemorrhage after both 24 and 32 hours. In contrast, all guinea pigs treated with both venom and the plant extract (administered independently) at concentrations between 500 and 1000 mg/kg showed no signs of haemorrhage after 32 hours. However, after 24 hours all tested guinea pigs showed less inhibition (<60%) compared to that obtained after 32 hours. The outcome of this study reflects that the extract of H. aethiopicus plant may contain an endogenous inhibitor of venom induced local haemorrhage.

Battery of tests for profiling abnormalities of vitamin K-dependent coagulation factors in drug-toxicity studies in rats

A battery of simple tests for profiling abnormalities of vitamin K-dependent coagulation factors encountered in drug-toxicity studies was verified in rats treated with warfarin (3 and 10 mg/kg, p.o). The thrombotest, or hepaplastin-test, is useful as a follow-up test after routine screening tests for coagulation abnormalities based on PT and APTT, to rule out other coagulation-factor abnormalities. Measurement of coagulation factor activities (factors II, VII, IX and X) using factor-deficient human plasmas provides direct evidence of decreased activities of vitamin K-dependent factors. Furthermore, Echis carinatus venom coagulation time, together with factor II activity, allows us to confirm the generation of PIVKA-II.

Snake venom hemorrhagins

Viperine and crotaline snake venoms contain one or more hemorrhagic principles called hemorrhagins. These are zinc-containing metalloproteases characterized by the presence of a protease domain, with additional domains in some of them. They act essentially by degrading the component proteins of basement membrane underlying capillary endothelial cells. The toxins also act on these cells causing lysis or drifting apart, resulting in hemorrhage per rhexis or per diapedesis. Some of these toxins have been found to exert additional effects such as fibrinogenolysis and platelet aggregation that facilitate hemorrhage. The structural and functional features of this class of toxins have been discussed in this review in an attempt to get a better understanding of their toxicity. This can be of immense therapeutic value in the management of snake venom poisoning, as hemorrhagins are among the major lethal factors in snake venom.

Purification and characterization of multisquamase, the prothrombin activator present in Echis multisquamatus venom

The venom of Echis multisquamatus (Central Asian sand viper) contains a single prothrombin activator, designated multisquamase, which is structurally and functionally different from ecarin, the prothrombin activator from the venom of Echis carinatus (saw-scaled viper). Multisquamase is comprised of a 58000 Mr and a 23000 Mr subunit that consists of two disulfide-linked chains of 12000 Mr and 10000 Mr, respectively. In contrast to ecarin, which activates prothrombin and prethrombin 1 at comparable rates, and whose activity is hardly affected by Ca2+ or by changes in ionic strength, multisquamase hardly activates prethrombin 1; prothrombin activation requires Ca2+ and is strongly inhibited at high ionic strength. The most favourable kinetic parameters are observed at 1 mM Ca2+ and at low ionic strength (Km=0.085 microM and kcat=0.68 s(-1) at I approximately 0.04). An increase in ionic strength considerably reduces the rate of prothrombin activation, due to an increase of the Km (Km=0.8 microM and kcat=1.03 s(-1) at I approximately 0.2). Studies in plasmas from patients on oral anticoagulant therapy show that E. Multisquamatus venom only activates carboxylated prothrombin, whereas E. carinatus activates both prothrombin and descarboxyprothrombin. Thus, multisquamase-dependent prothrombin activation appears to require post-translational modification of the gla-domain. This venom prothrombin activator may, therefore, become a useful tool to quantitate prothrombin and descarboxyprothrombin in cases where vitamin K-dependent carboxylation of prothrombin is impaired.

Characterization of snake venom components acting on blood coagulation and platelet function

Snake venoms can affect blood coagulation and platelet function in various ways. The physicochemical properties and the mechanisms of actions of the snake venom components affecting blood coagulation and platelet function are discussed.

A novel one-step purification of human alpha-thrombin after direct activation of crude prothrombin enriched from plasma

Crude prothrombin enriched from human plasma was directly activated to generate alpha-thrombin without prior exhaustive purification of the proenzyme using a combination of several different types of chromatographic techniques, as in all previously described methodologies. Activated thrombin was separated from other components in a single step by taking advantage of its highly specific affinity to heparin immobilized on a matrix support of Sepharose CL-6B. On the basis of the data presented herein, we have demonstrated the ease with which at least 25 mg of a highly purified enzyme (greater than 97% homogeneous by laser densitometry) can be obtained per litre of plasma. Our product exhibits a specific activity of at least 4000 National Institutes of Health units/mg and is stable after being freeze-dried for the purpose of long-term storage.

Procoagulant activities in venoms from central Asian snakes

The venoms from central Asian snakes (Echis carinatus, Echis multisquamatus, Vipera ursini, Vipera lebetina, Agkistrodon halys halys and Naja naja oxiana) contain several enzymes with amidolytic- and procoagulant activity. We have characterized the activities and the mol. wts of the venom enzymes that are able to convert a number of commercially available chromogenic substrates for activated coagulation factors. The chromogenic substrate cleavage patterns obtained for the crude venoms may be helpful tools in the further identification of venom fractions and venom enzymes with procoagulant activity. The crude venoms were also tested for their ability to clot fibrinogen, to lyse fibrin polymers and to activate the coagulation factors prothrombin, factor X and factor V. The products of venom-catalyzed coagulation factor activation were structurally characterized by SDS gel electrophoresis and were compared with activated coagulation factors that are generated under physiological conditions.

Effects of snake venoms on hemostasis

Proteins found in venoms, especially of the Viperidae snake family, exert, often with a narrow specificity, activating, inactivating, or other converting effects on different components of the hemostatic and fibrinolytic systems, respectively. Some purified snake venom proteins have become valuable tools in basic research and in diagnostic procedures in hemostaseology. "Procoagulant" as well as "anticoagulant" venom components have been identified in in vitro test systems. "Procoagulant" snake venom components may cause in vivo, upon massive application as in the case of snake-bite of small prey animals, intravascular coagulation leading to circulatory arrest and rapid death. Smaller doses of procoagulant venom components applied to large organisms as in the case of snake-bite accidents in humans, may cause a consumption coagulopathy with localized or generalized bleeding. Highly purified, specific fibrinogen coagulant venom proteinases are used in human medicine to produce therapeutic defibrinogenation. These practically nontoxic venom enzymes may act synergistically with other components aggravating their toxic effects.

Effects of venom proteases on peptide chromogenic substrates and bovine prothrombin

Eighteen proteases were isolated from six hemorrhagic venoms of snakes belonging to the families of Crotalidae and Viperidae. According to their actions, they are classified as thrombin-like enzymes, alpha-fibrinogenases, beta-fibrinogenases, Factor X activator, prothrombin activator, hemorrhagins and esterases. Thrombin-like enzymes, beta-fibrinogenases, hemorrhagins and esterase hydrolyzed Phe-Pip-Arg-pNA (S-2238, substrate for thrombin) more strongly than CBZ-Ile-Glu-Gly-Arg-pNA (S-2222, substrate for Factor Xa), CBZ-Phe-Val-Arg-pNA (B-7632) or CBZ-Pro-Phe-Arg-pNA (B-2133). Thrombin-like enzymes, beta-fibrinogenase and esterase hydrolyzed tosyl-L-arginine methyl ester and benzoyl-L-arginine ethyl ester. S-2238 is the most susceptible chromogenic substrate for most venom proteases. Thrombin-like enzymes degraded prothrombin molecule progressively down to prethrombin 2 while alpha- and beta-fibrinogenases degraded it only to prethrombin 1. Factor X activator of Vipera russelli venom and esterase of T. mucrosquamatus venom did not have any effect on prothrombin. Thus, the effects of venom proteases on prothrombin are not parallel to their amidolytic or esterolytic effects.

Purification and properties of a proteinase from Vipera lebetina (snake) venom

A proteinase from the venom of Vipera lebetina was purified by chromatography on Sephadex G-100 and CM-cellulose. The purified proteinase was homogeneous on SDS-polyacrylamide gel electrophoresis and consisted of a single chain with molecular weight of 37,000 +/- 1500. The isoelectric point of the proteinase was over 10. The enzyme was active on casein but not on esters and amides of arginine. It split the oxidized insulin B-chain at the peptide bonds of Tyr16-Leu17, Phe24-Phe25 and Phe25-Tyr26, and glucagon at the bonds Tyr10-Ser11, Leu14-Asp15 and Leu26-Met27. The enzyme was inhibited by DFP and PMSF, and partially by soybean trypsin inhibitor, but not with EDTA.

The action of Echis carinatus venom on the blood coagulation system. Demonstration of an activator of factor X

It is shown that Echis carinatus venom activates both coagulation factor II and coagulation factor X. The activation of both proenzymes by the venom is Ca++-dependent; phospholipids are not necessary. The activation of factor II by the venom is a slow process and, in the absence of factor X, the clotting activity towards fibrinogen is generated only very slowly. Because Echis carinatus venom clots plasma readily, we postulate that under conditions where the prothrombinase complex can be formed from the factor X activated by the venom it is this complex, rather than the venom itself, that is responsible for the major part of the thrombin formation.

The action of prothrombin activated by Ecarin on fibrinogen

Highly purified human prothrombin was activated by Ecarin, a prothrombin activating principle of Echis carinatus venom and the generated thrombin-like activity was investigated. Kinetics of the release of fibrino-peptides A and B (FPA, FPB) from human fibrinogen was estimated using radioimmunoassay technique. The direct proteolytic action of Ecarin on fibrinogen was studied by means of polyacrylamide gel electrophoresis and by radioimmunoassay for fragment B beta 15-42 of fibrinogen. In a system containing Ecarin, fibrinogen and prothrombin both fibrinopeptides were cleaved off at a rate that was essentially similar to that observed with thrombin, the cleavage off of FPB being in the initial stage always slower than the release of FPA. At physiological concentration of prothrombin and at low concentration of Ecarin all FPB were liberated while less than 1% of B beta 15-42 immunoreactivity was released. It was possible to accelerate the release of FPB when prothrombin was activated by Ecarin in the presence of Fibrin I gel (formed from fibrinogen by Batroxobin which liberates FPA only) instead of fibrinogen. Fibrin I thus constituted a more favourable substrate than fibrinogen with regard to release of FPB. The results indicate that the coagulant activity generated from prothrombin by Ecarin is thrombin-like. No intermediate product capable of marked cleavage off of FPA only was detected.

Isolation of the prothrombin-converting enzyme from fibrinogenolytic enzymes of Echis carinatus venom by chromatographic and electrophoretic methods

Chromatography of crude Echis carinatus venom revealed four enzymes with fibrinogenolytic activity and activity to chromogenic substrates, specific for proteases of the coagulation and fibrinogenolytic systems. By employing three-step chromatography on DEAE-Sephacel and Sephacryl S-200, this venom afforded the procoagulation enzyme Ecarin in good recovery (73%) and purification (53-fold). This highly purified preparation has proved to be a single-chain glycoprotein, occurring in two isomers differing in electrical charge and having a low fibrinogenolytic activity.

Post-translational carboxylation of preprothrombin

In summary, in this review on the function of vitamin K in post-translational modification of precursor proteins by carboxylation of certain glutamyl residues, I have tried to cover in particular the recent work on the reaction, the enzymes involved and the mechanisms being considered. In doing this I have also considered vitamin K, its discovery, its functional form and the possible relation of its metabolism to the carboxylation reaction. Equally the various vitamin K-dependent gla-containing proteins currently known have been described. The carboxylation of synthetic small molecule exogenous substrates and the synthesis and metabolism of the products of carboxylation are of great help in studying the reaction. Structural specificity of vitamin K analogs in vivo and in vitro has been compared and the use of various antagonists in vivo and in vitro considered in attempts to gain an understanding of the overall reaction. The reactions subsequent to carboxylation, e.g., the activation of prothrombin to thrombin via serine proteases and the related activation of the other vitamin K-dependent proteins have not been considered in this review. The review has not covered prothrombin or other vitamin K-dependent protein isolation, nor the determination of these proteins. As the vitamin K-dependent protein carboxylation story has developed over the past six years, a number of reviews have been written which help in keeping up with the various aspects of the field as it has expanded. These reviews refer to many of the papers I have had to eliminate due to space limitations. They are referenced as 469-489. The review is in no sense comprehensive and many papers have been missed or only mentioned. I have tried to concentrate on the more recent work and, thus, much of the very fine work of the 1940's on vitamin K chemistry is hardly mentioned. Some redundancy has been built into the organization of the review so that a reader can obtain a reasonable view of any one section without having to search the whole review for all possible relevant information on any particular part of the field.

Factor II antigen in liver disease and warfarin-induced vitamin K deficiency: correlation with coagulant activity using Echis venom

Four coagulation factors (II, VII, IX, X) are synthesized in the liver as precursor forms. Vitamin K is necessary for their conversion to functional factors. Factor II precursor is antigenically similar to factor II and can be activated to form thrombin by Echis carinatus venom. Forty-one patients with liver disease and 30 patients with vitamin K deficiency (Warfarin treated), were compared with 51 normal patients. Rats with CCl4-induced hepatic necrosis and animals given Warfarin were also studied. The following measurements were made: standard prothrombin time; Echis clotting time; factor II coagulant assay with thromboplastin; factor II assay with Echis venom; and factor II antigen (human) by electroimmunoassay. In animals and humans with liver disease, factor II was reduced, as measured by all techniques. With vitamin K deficiency functional factor II was reduced, but factor II antigen and Echis factor II activity were normal. THe data suggest that the prothrombin time and Echis coagulation methods can be used to differentiate whether the coagulopathy in liver disease is due to reduced precursor levels of factor II, vitamin K deficiency (ie, impaired formation of functional from precursor factor II), or both.

Purification and characterization of the anticoagulant principle of Trimeresurus mucrosquamatus venom

By means of CM-Sephadex column chromatography, Trimeresurus mucrosquamatus venom was separated into 20 fractions. Fraction XX had the marked anticoagulant action. This fraction was refractionated three times on Sephadex G-75, and a single peak was obtained. The patterns of microzone and disc electrophoresis also showed a single band. A single, symmetrical boundary with a value of 1.61 S was obtained by ultracentrifugation. It was a single peptide chain with a molecular weight of 11 700. The isoelectric point was higher than pH 10. The anticoagulant principle possessed phospholipase A activity and was calcium ion dependent. It did not possess proteolytic, tosyl-L-arginine methyl ester esterase, phosphodiesterase and alkaline phosphomonoesterase activities of the crude venom. The phospholipase A activity was heat-labile at pH 7.4, but was heat-stable at pH 5.6. The anticoagulant activity was more resistant to heat treatment as compared with phospholipase A activity. The anitoagulant action of the purified principle was competitively inhibited by platelet phospholid, tissue thromboplastin and cephalin, and was neutralized by antiserum. The anticoagulant principle inhibited platelet aggregation induced by ADP. It did not destroy fibrinogen, Factor X, prothrombin and thrombin; nor did it induce fibrinolysis nor interfere with the interaction between thrombin and fibrinogen. It is concluded that the anticoagulant action of this phospholipase A was due to the inhibition of the activations of Factors X and II through the inactivation of the procoagulant activity of phospholipids mediated partly by phospholipid-binding activity of this venom enzyme and partly by its enzymatic hydrolysis of phospholipids.

Disseminated intravascular coagulation caused by the carpet viper (Echis carinatus): trial of heparin

Heparin has been advocated for the treatment of poisoning by Echis carinatus, a snake whose venom causes disseminated intravascular coagulation. Fourteen patients with proven E. carinatus bite who had incoagulable blood were treated with specific Echis antivenom. Seven of them were also given low-dose heparin, initially 50 units/kg body weight by i.v. injection, followed by 10 units/kg/h by i.v. infusion for 22 h. Response to treatment was assessed clinically and by repeated tests of blood coagulation. All patients showed a rapid return to normal blood coagulability after treatment and the heparinized group were not significantly different in any respect from the group given antivenom alone. Heparin did not reduce the local effects of envenoming. There appears to be no place for heparin in the treatment of E. carinatus poisoning provided that potent antivenom is available. The in vivo results were supported by in vitro studies in which it was found that Echis-induced thrombin was less sensitive to the inhibitory effect of heparin than physiological thrombin.

Viper venoms and coumarin-induced prothrombin. A comparison of several one-stage methods employing three different venoms as thromboplastins

Three venoms obtained from three vipers, namely Echis carinatus, Notechis scutatus scutatus and Oxyuranus scutellatus, have been used as thromboplastin in a one-stage assay of coumarin-induced prothrombin. Regardless of the venom used, prothrombin resulted to be low in coumarin-treated patients. The mean values obtained were 27.2, 33.6, and 24.2%, respectively. These values were comparable to those obtained by means of the classical one-stage method (24.8%). A good correlation was observed among the different methods. However, the levels observed using the Notechis scutatus scutatus venom method were slightly higher as compared to those obtained by means of the other viper venoms and by means of the classical one-stage method. The three viper venoms used seem unable to activate coumarin-induced prothrombin. The levels obtained were in fact, in each instance, definitely lower than those observed immunologically. Methods which employ these viper venoms may be used in the evaluation of prothrombin in coumarin-treated patients.

The effect of several viper venoms on prothrombin Padua

The effect of four viper venoms (Oxyuranus scutellatus, Notechis scutatus scutatus, Echis carinatus, Naja nigricollis) on prothrombin Padua has been studied. Using Oxyuranus scutellatus venom and Notechis scutatus scutatus venom, prothrombin activity resulted to be moderately decreased similarly to what observed with other one-stage and two-stage methods. On the contrary, using Echis carinatus venom a normal level was obtained. No clotting was observed using the Naja nigricollis venom, regardless of the concentration used. The normal level of factor II obtained with Echis carinatus venom as compared with the low levels obtained with the other venoms, suggests that it acts on a different site of the prothrombin molecule.

Purification and properties of the anticoagulant principle of Trimeresurus gramineus venom

By means of DEAE-Sephadex A-50 column chromatography, Trimeresurus gramineus venom was separated into 12 fractions. Fraction 8 had marked anticoagulant action in the tests of whole blood clotting time, calcium clotting time and plasma prothrombin time. Fraction 8 was rechromatographed on Sephadex G-100, then on DEAE-Sephadex A-50 again, and finally on Sephadex G-100, and a single peak was obtained. The patterns of microzone and disc electrophoresis also showed a single band. A single symmetrical boundary with 1.70 Svedberg units was obtained by ultracentrifugation. The estimated molecular weight was 19 500. The isoelectric point was pH 4.5. Chemical analysis showed that the anticoagulant principle was a glycoprotein and that it was thermolabile. The anticoagulant activity of this purified principle was 3.5 times higher than that of the crude venom. Fraction 5 potentiated its anticoagulant activity to 10 times higher than that of the crude venom. This principle did not possess caseinolytic, tosyl-L-arginine methyl ester esterase, phospholipase A, phosphodiesterase, alkaline phosphomonoesterase, fibrinolytic, hemorrhagic or local irritating activities. The purified anticoagulant principle did not destroy fibrinogen, induce fibrinolysis, inactivate thrombin nor interfere with the interaction between thrombin and fibrinogen. However, a marked inhibition of prothrombin activation was caused by the anticoagulant principle. The inhibition of prothrombin activation was not due to the destruction of prothrombin or its activation factors, but due to an interference in the interaction between prothrombin and its activation factors because of the reversible binding of these factors with the anticoagulant principle of the venom.

Changes in phylloquinone epoxidase activity related to prothrombin synthesis and microsomal clotting activity in the rat

The oxidation of phylloquinone to the 2,3-epoxide (by phylloquinone epoxidase) was studied in liver from control and warfarin-resistant rats. The reaction requires microsomal fraction, soluble protein, a heat-stable soluble factor and O(2). It is not inhibited by CO or CN(-). Epoxidase activity was stimulated if plasma prothrombin was lowered either by anticoagulants or the absence of vitamin K. The activity of the enzyme rapidly returned to normal values after the administration of vitamin K to hypoprothrombinaemic rats. These differences in the activity of the enzyme occur in the microsomal fraction and not the cytosol. A thrombin-generating polypeptide that accumulates in microsomal fraction of hypothrombinaemic rats correlated directly with epoxidase activity. These data support the view that enzymic interconversion of phylloquinone and its 2,3-epoxide participates in the biological activity of vitamin K.

Mechanism of action of vitamin K: demonstration of a liver precursor of prothrombin

Extracts of sonicated liver microsomes that are prepared from rats deficient in vitamin K or from rats given vitamin K antagonists contain a factor that liberates a thrombin-like activity when it is incubated with venom from Echis carinatus. The amount of this factor is low in control rats and in hypoprothrombinemic rats given vitamin K 1 hour before they were killed. These data indicate that this factor is a protein precursor of prothrombin, which is synthesized in the liver.