This Thesis describes the purification and characterization of several components of the venom of Australian snakes which are active in the coagulation and fibrinolytic pathways. The hypothesis driving these studies has been that venom components or their mimetics may prove useful as therapeutic and diagnostic agents.
Chapter 1 is the General Introduction, commencing with a brief historical aside and including an overview of haemostasis, anticoagulation pathways, phospholipase A₂, the fibrinolytic system, biological significance of snake venoms, snake venom proteinase inhibitors and summary of the scope of the Thesis. Chapter 2 is Materials and Methods, Chapters 3 to 6 are Results Chapters and Chapter 7 outlines conclusions and future prospects.
In Chapter 3, six anticoagulant components have been purified from the venom of the Australian King Brown snake, all of which possess phospholipase A₂ (PLA₂) activity. These have been purified using a combination of heparin Sepharose affinity chromatography, CM-Sepharose ion exchange chromatography and gel filtration on Sephacryl S-100. Six fractions were defined and shown to be homogeneous by SDS-PAGE (with and without β-mercaptoethanol) and anticoagulant activity (measured using prothrombin time (PT) and activated thromboplastin time (aPTT)) but they differed in specific PLA ₂ activity. Two groups were defined (AC I and AC II), based on separation on heparin Sepharose. The AC II group demonstrated the more potent anticoagulant activity and weaker PLA₂ activity.
Inhibition of PLA₂ activity by heparin (1.0 U) led to significant enhancement of anticoagulant activity. These six anticoagulant have not been related as yet to the fifteen PLA2 fractions purified and sequenced by Takasaki et al (1990). The six anticoagulant fractions had varied effects on platelet function. The AC II group showed the most significant effects but all showed significant toxicity when injected into the tail vein of a mouse. The cause of death as ascertained by post mortem examination appeared to be excessive bleeding into the viscera. Lower concentrations of the AC II anticoagulant IV injected into rats resulted in significant prolongation of the PT.
Kinetic studies of AC I and AC II showed no inhibition of thrombin or Factor Xa. These observations led to the hypothesis that these anticoagulants interfere with the formation of prothrombinase and TENase. They may also interfere with the formation of other activation complexes in both the extrinsic and intrinsic blood coagulation pathways.
Chapter 4 deals with the development of a drug candidate with potential to manage blood loss during major surgery or after severe trauma, where haemorrhage is a concern.
The data presented show that two polypeptide components isolated from the venom of the Brown snake, Pesudonaja textilis, have been identified as inhibitors of the serine proteases plasmin and trypsin. These inhibitors were named Textilinin (Txln) 1 and 2. Txln 1 and 2 were shown to be non-toxic (in mice) and was as efficient as aprotinin (Trasylol, Bayer) in controlling blood loss when equivalent concentrations were used. A purification protocol for both Txlns is presented: it includes a combination of gel filtration on Sephacryl S-300, ion exchange on DEAE-Sepharose CL-4B and gel filtration on Sephacryl S-100, with a yield of 1.4 and 2.1% (based on A₂₈₀ units) respectively. SDS-PAGE with and without β-mercaptoethanol, showed a single band of approximate molecular weight 7,000.
N-terminal sequencing showed that Txln 1 and 2 are similar, only differing by 7 amino acids over the 59 residues. The primary sequence data for Txln 1 and 2 show that they contain six cysteines (probably forming three intra-chain disulphide bonds), similar to the structure observed in other (Kunitz-type) serine protease inhibitors such as aprotinin. Txln 1 and 2 showed, respectively, 45 and 43% homology with aprotinin. There was also 58 and 55% homology, respectively, with another closely related naturally occurring plasmin inhibitor that is present in the venom of the Eastern Taipan, Oxyuranus scutellatus. The Ki values for anti-plasmin activity of Txln 1 and 2 did not differ significantly, being 2nM and 7 nM respectively. Reverse phase chromatography on a Waters RP-8 column showed single peaks for both Txln 1 and 2. Recovery and reconstitution in buffer established that the material from each chromatographic peaks retained anti-plasmin activity. The full amino acid sequences of Txln 1 and 2 have been determined, both being 59-amino acid polypeptides with calculated molecular mass of 6681.4 and an observed molecular mass of 6682.4 Da. Ion spray mass spectroscopy analysis indicated a molecular mass of 6686 Da for Txln 1 and 6692 Da for Txln 2. Comparison of the amino acid sequences of Txln 1 and 2 with the reported sequences of aprotinin-like antiplasmin proteins purified from venoms of Haemachatus hemachatus and Vipera russellii, showed significant homology.
Comparison of inhibitor constants indicated that aprotinin bound very tightly (Ki of about 10¯¹¹ to plasmin, while Txln 1 and 2, both bound somewhat less tightly (Ki of about 10 ¯⁹). Both Txln 1 and 2 reduced blood loss by about 60% in a murine tail vein-bleeding model. It is proposed that the kinetic profiles for Txln 1 and 2 interacting with plasmin, together with their inability to inhibit kallikrein, should allow the arrest of haemorrhage without the threat of thrombosis and vein-graft occlusion.
The cloning and expression of Txln is also presented in this Chapter. Recombinant Txln refolded easily into an active protein, which inhibited plasmin and effectively reduced loss in the mouse tail-vein bleeding model. This genetically engineered venom gland protein is potentially a new therapeutic agent. This is particularly timely addition in view of consideration of the current world-wide anxiety about possible viral or prion contamination of both autologous blood and aprotinin (TrasylolR, a bovine lung extract) used in these types of surgery.
In Chapter 5, a study is presented aiming to relate the high prevalence of fatal bites by Brown snakes (Genus: Pseudonaja) to snake venom yields and prothrombin activator activity. The amount of venom obtained from repeated milkings of 66 snakes (over 15 months), representing four Pseudonaja species, was measured. The LD₅₀ for P. inframacula venom has been determined for the first time. SDS-PAGE and Western blotting studies have shown that Brown snake antivenom (BS-AV) recognizes the prothrombin activator (PA) and also other venom components, including the neurotoxins. Neutralisation of the prothrombin activator of the Brown snake (P. textilis) (Pt-PA) by BS-AV was found to be time dependent, even with excess antivenom added.
Adult rats administered 8 microgram quantities of Pt-PA (IV) died within minutes with acute disseminated intravascular coagulation (DIC). Rats could be made resistant to Pt-PA either by preheparinization or by inducing tolerance to increasing quantities of Pt-PA. This tolerance regime causes exhaustion of the haemostatic system by prior exposure to small quantities of activator. These responses are characteristic of procoagulants in general. From earlier work, there is no evidence that Pt-PA has any other intrinsic toxicity apart from being a procoagulant (see candidate's publications). The experimental findings recorded here implicate two major factors which probably influence envenomation outcomes: (1) the dose of venom obtained by milking was found to be greater in some specimens than expected from earlier work and (2) a more significant role than previously acknowledged seems likely for the prothrombin activator component. If these considerations are valid, it seems that either the current commercial BS-AV should be packaged in larger volumes, or that increased dosage should be advised when it is used in certain geographical localities. Another significant option is to spike the antivenom with specific antibodies or other specific antidotes or compounds (such as low molecular weight Factor Xa-like inhibitors), which inhibit the prothrombin activator activity with greater affinity and shorter interaction time.
In Chapter 6, studies focussed on investigations of five recent fatal snakebites by Brown snakes (four) and Taipan (one). All cases showed very similar disturbed coagulation and fibrinolytic systems, particularly the presence of disseminated intravascular coagulation and severe pathological secondary fibrinolysis. Brown snake venom could not be detected in the plasma of one patient bitten five times by a Brown snake, using Western blotting studies and visualising with an enhancement protocol able to detect low nanogram amounts. However, immunohistochemical analysis of tissues from the same patient showed the presence of venom bound to various tissue components.
These studies have highlighted the inadequacies of CSL Brown snake and Taipan antivenom, since all five patients had received copious amounts of specific antivenom during intensive care. Nonetheless all five died.
Another common feature of all five cases was a protracted renal failure, even subsequent to the correction of coagulation and fibrinolytic disturbances, suggesting that significant renal damage had occurred. This non-resolving renal failure is a real concern for most physicians who have had to manage envenomation patients.
Finally, the studies presented here highlight the enormous potential of snake venoms as source of (or blueprint for) novel agents to manage abnormalities in haemostasis as well as other diseases.