Alpaca Immunoglobulins

Home , Monkey bite, Papuan black snake, Scorpion sting

OCTOBER 2012

Alpaca Immunoglobulins

by Andrew Padula

Oct 2012

RIRDC Publication No 14-065 RIRDC Project No PRJ-007770

ISBN 978-1-74254-681-0 ISSN 1440-6845

Alpaca immunoglobulins Publication no. 14-065 Project No. PRJ-007770

The information contained in this publication is intended for general use to assist public knowledge and discussion and to help improve the development of sustainable regions. You must not rely on any information contained in this publication without taking specialist advice relevant to your particular circumstances.

While reasonable care has been taken in preparing this publication to ensure that information is true and correct, the Commonwealth of Australia gives no assurance as to the accuracy of any information in this publication.

The Commonwealth of Australia, the Rural Industries Research and Development Corporation (RIRDC), the authors or contributors expressly disclaim, to the maximum extent permitted by law, all responsibility and liability to any person, arising directly or indirectly from any act or omission, or for any consequences of any such act or omission, made in reliance on the contents of this publication, whether or not caused by any negligence on the part of the Commonwealth of Australia, RIRDC, the authors or contributors.

The Commonwealth of Australia does not necessarily endorse the views in this publication.

This publication is copyright. Apart from any use as permitted under the Copyright Act 1968, all other rights are reserved. However, wide dissemination is encouraged. Requests and inquiries concerning reproduction and rights should be addressed to the RIRDC Publications Manager on phone 02 6271 4165.

Researcher Contact Details

Name: Andrew Padula Address: 26 Howitt Ave, Bairnsdale, Victoria, 3875

Email: [email protected]

In submitting this report, the researcher has agreed to RIRDC publishing this material in its edited form.

RIRDC Contact Details

Rural Industries Research and Development Corporation Level 2, 15 National Circuit BARTON ACT 2600

PO Box 4776 KINGSTON ACT 2604

Phone: 02 6271 4100 Fax: 02 6271 4199 Email: [email protected]. Web: http://www.rirdc.gov.au

Foreword

This research explores an alternative use for alpaca, other than for fibre production. Fibre production has been the traditional saleable product from alpaca, along with trade in live animals and, to a lesser extent, alpaca meat. Blood contains molecules called immunoglobulins, which have a number of medical uses, including antivenom production. Australia has over 130,000 alpaca which represent a vast, untapped resource for using the animals to produce unique therapeutic immunoglobulins. This research has examined the potential of the alpaca immunoglobulins for a specific application: snake antivenom production.

There are two main benefactors of this research. Firstly, the Australian alpaca industry stands to benefit from having an alternative use for the alpaca, as a producer of high value niche market products. Secondly, the benefits are to the global community in terms of safer and more effective therapeutic immunoglobulin products.

Alpaca belong to the camelid family. Camels have recently been shown to produce a unique form of immunoglobulins in their blood. These immunoglobulins have been shown in the camel to be more heat stable, less allergenic than horse or sheep immunoglobulins, and better at inhibiting some enzymes. These properties, especially the lower allergenicity, identify camelid immunoglobulins as potentially having significant advantages over horse or sheep immunoglobulins.

This project was undertaken as a proof of concept study. No similar work has been conducted anywhere in the world. Much of what was undertaken was developed on basic science principles.

The key findings of this research are that alpaca can be used successfully to produce therapeutic immunoglobulins. This project explored one specific type of immunoglobulin product; antibodies produced against snake venoms, for antivenom production. In this project, we have demonstrated that alpaca can be immunised, without harm to the animal, against highly toxic and low molecular weight compounds. We have shown that alpaca produce strong immune responses to low doses of a range of snake venoms. We have developed the technology to monitor these specific immune responses. We have developed a system for harvesting of the alpaca blood and separating the serum. We have successfully adapted a process used in horses to concentrate these immunoglobulins during processing. We have obtained the necessary animal ethics approvals to test the protective nature of these immunoglobulins in mice that are challenged with doses of venom. Finally, we have shown that these immunoglobulins are truly protective against the lethal effects of snake venoms.

This study has proved the concept that alpaca can make potent therapeutic antibodies against snake venoms. However, further work is required before this could be commercialised.

This report is an addition to RIRDC’s diverse range of over 2000 research publications and it forms part of our Rare Natural Animal Fibres R&D program, which aims to conduct RD&E for new and developing animal industries that contribute to the profitability, sustainability and productivity of regional Australia.

Most of RIRDC’s publications are available for viewing, free downloading or purchasing online at www.rirdc.gov.au. Purchases can also be made by phoning 1300 634 313.

Craig Burns Managing Director Rural Industries Research and Development Corporation

iii

About the author

Andrew Padula is a veterinarian with a PhD in veterinary physiology from the University of Melbourne. He owns, and runs, a veterinary practice in Bairnsdale, Victoria. Acknowledgments

This project would not have been possible without the enthusiasm and financial support of the Australian Alpaca Association (AAA). The AAA loaned various equipment for use in the project. In particular, Fiona and Ian Vanderbeek, of Alcazar Suri Stud, have been outstanding in helping to make this project successful, and in supplying alpaca. Thanks also to Jenny McDavitt, Merungle Alpaca Stud, Paul and Fran Haslin of Elysion Alpaca, Paul Cramley and Linda Davies of Pacofino, and Ros and Michael Davis of Elimbari Alpaca for supplying animals for use at various stages in the project. .

Technical aspects of the project were supported by Frank Madaras and Peter Mirtschin of Venom Science Pty Ltd, Tanunda, South Australia. The uniqueness of the alpaca necessitated starting from scratch with some aspects of this work, and Frank in particular helped with numerous issues.

Thank you Dr Tim Kuchel of the Institute of Veterinary and Medical Sciences, Adelaide, South Australia, for advice on various technical aspects of antibody production. Thank you to Prof. Jose Maria Gutierrez, University of Costa Rica, Instituto Clodomiro Picado, Costa Rica for enthusiastic technical advice. Thanks to Nathan Dunstan of Venom Supplies Pty Ltd, South Australia for supply of the venoms used in the project. The venoms are the essential component required to make antibodies, and without the help of those who place themselves at risk with venomous snakes, this work could not happen.

Thank you to Dr Ken Winkel, Director of the Australian Venom Research Unit, University of Melbourne. Ken has been part of many fruitful and informed discussions on the topic of snake venoms and treatments. We are fortunate to have such expertise in Australia.

Thank you to Max Campbell and Ian Temby, of the Department of Primary Industries Wildlife and Small Institutions Animal Ethics Committee, for working with me the get the necessary approvals to undertake this work. I am most appreciative of the ‘working together’ approach with the delicate animal ethics issues associated with snake venoms and their application in animals. Thank you to Catheryn Obrien, veterinarian at Walter Eliza Hall Institute, for generous donation of technical knowledge and experience in working with mice. Thank you to the Animal Welfare Science Centre, University of Melbourne, for assistance with obtaining animal ethics approvals.

Thank you to staff at Bairnsdale Animal Hospital: Kylee, Lisa, Debbie and Stephanie, for putting your time into this project. In particular, I would like to thank Allan Quirke, aka “the human clamp”, for the hours of helping with the field work with the alpaca. His abilities in the field and many useful tips are most appreciated.

Thank you to Julie Bird of RIRDC for commissioning this project.

Abbreviations

µg: Microgram

µL: Microlitre

µm: Micrometre

AAA: Australian Alpaca Association

CSL: Commonwealth Serum Laboratory

ELISA: Enzyme Linked Immunosorbent Assay g: Gram h: hour

HCIgG: Heavy chain only immunoglobulin

IgA: Immunoglobulin A

IgG: Immunoglobulin G

Kg: Kilogram

M: Molar mg/mL: milligram per millilitre mg: Milligram mL: millilitre

MW: molecular weight

NaCl: Sodium Chloride nm: Nanometre

PCV: packed cell volume

PBS: Phosphate Buffered Saline

PLA2: Phospholipase A2

U/mL: Units per millilitre

U: Units

°C: degrees Celsius

Contents

Foreword ...... iii

About the Author...... iv

Acknowledgments...... iv

Abbreviations ...... v

Tables ...... ix

Figures ...... x

Executive Summary ...... xiii

Introduction ...... 1

Aims of this Project ...... 1 The Alpaca and Other Camelids ...... 1 An Alternative Use for Alpaca in Australia ...... 2 Unique Aspects of Camelid Immunoglobulins ...... 3 Discovery of the structure of heavy-chain only camelid antibodies ...... 3 VHH ...... 4 Rationale for the Development of Camelid Immunoglobulin Products ...... 6 Previous Work on Antivenom Production in Camelids ...... 7 Merits of Alpaca for Antiserum Production ...... 9 Polyclonal Antibody Products ...... 10 Snake Antivenoms for Human Use in Australia ...... 13 Snake Antivenoms for Veterinary Use in Australia ...... 15 Medically Important Venomous Australian Snakes ...... 17 Tiger Snakes (Notechis sp.) ...... 19 Taipans (Oxyruanus sp.) ...... 22 Brown Snakes (Pseudonaja sp.) ...... 24 Black Snakes (Pseudechis sp.) ...... 27 Death Adders (Acanthophis sp.) ...... 30 Sea Snakes ...... 31 Actions of Australian Snake Venom Toxins ...... 32 Neurotoxins ...... 32 Myotoxins ...... 32 Coagulalopathy ...... 32 Nephrotoxicity ...... 33 Cardiotoxicity ...... 33 Local Effects ...... 33

Overview of the Antivenom Production Process ...... 33 Hyperimmunisation of Animals ...... 33 Monitoring the Immune Response ...... 33 Harvesting of Serum or Plasma ...... 34 Processing and Concentration of Antibodies ...... 34 Quality Control ...... 35 World Antivenom Situation ...... 35

Objectives ...... 36 Specific Objectives as Defined in the RIRDC Research Agreement ...... 37

Methodology ...... 38

Animal Ethics Approval ...... 38 Alpaca Sourcing and General Management ...... 38 Handling of Alpaca ...... 39 Health Monitoring Post-Venom Immunisation...... 40 Snake Venom Preparation for Immunisation ...... 40 Treatment Groups ...... 40 Immunisation Program ...... 41 ELISA for Venom Antibody Measurement ...... 41 Packed Cell Volume ...... 42 Venom Toxicity Determination ...... 42 Neutralisation of Lethality in Mice ...... 43 Serum Protein Electrophoresis ...... 43 Processing of Serum ...... 43 Centrifugation and Filtration ...... 43 Caprylic Acid Fractionation ...... 45 Ammonium Sulphate Fractionation ...... 46 Study Trip to Instituto Clodomiro Picado, San Jose, Costa Rica...... 47

Results ...... 48 Summary of Alpaca Immunoglobulin Project Key Achievements ...... 48 Effects on Alpaca Health ...... 49 Serum Biochemical Changes ...... 49 Serum Muscle Enzymes ...... 49 Serum Inflammation Markers...... 49 PCV Response to Whole Blood Collection ...... 49 Antibody Responses to Venom - ELISA Results ...... 57 Serum Processing ...... 65 Ammonium Sulphate Precipitation ...... 68 Sodium Sulphate Precipitation ...... 70 Heat Treatment of Serum ...... 70 Median Lethal Dose Determination of Venoms ...... 71

vii

Tiger Snake ...... 71 Papuan Taipan ...... 72 Neutralisation of Lethality and Potency Determination ...... 72

Discussion of Results ...... 74

Implications...... 80

Recommendations ...... 82

Phase 2 Projects ...... 83 Improving the Efficacy and Safety of Immunisation and Bleeding ...... 83 Optimising Final Product Quality, Clarity and Stability ...... 83 Infectious Disease Risk Assessment and Biosecurity Plan ...... 83 Regulatory Compliance Issues Management ...... 83 Patent Landscape Search ...... 83 Development of Commercial and International Linkages...... 83 Documentation of Camelid Immunoglobulin Properties in Alpaca ...... 84

References ...... 85

viii

Tables

Table 1. Prices offered to growers for alpaca fleece by Alpaca Ultimate (www.alpacaultimate.com.au) in 2012...... 2 Table 2. Prices offered to alpaca fibre growers by House of Alpaca (www.thehouseofalpaca.com.au) in 2010...... 3 Table 3. Percentage of conventional and heavy chain IgG typically found in camelids [data from ((Hamers-Casterman et al. 1993, Maass et al. 2007)] ...... 4 Table 4. Therapeutic immunoglobulin products currently available worldwide for treating various envenomation’s (antivenoms)...... 11 Table 5. Currently registered therapeutic immunoglobulin products for treating various infectious diseases and bacterial toxins...... 12 Table 6. Currently registered immunoglobulin products for treating non-infectious human disease conditions...... 13 Table 7. Emerging applications for immunoglobulin and serum products at various stages of registration...... 13 Table 8. Product range of CSL land snake antivenoms...... 14 Table 9. Commonwealth Serum Laboratories, Australia, antivenom product range. One Unit is capable of neutralising 0.01mg of venom. CSL supplies snake antivenoms with a variable fill volume but each vial is filled with a minimum number of Neutralising Units per vial. This can also be expressed as potency in mg of venom neutralised per mL of antivenom which is more widely used internationally...... 15 Table 11. Toxicity of various Australian snake venoms in 18-21 gram laboratory mice listed in decreasing order of toxicity (Broad et al. 1979). The authors used over 5,000 mice to establish this dataset. (Note the shaded entries, these were used in the alpaca project.) ...... 18 Table 12. Summary of the major toxins in Tiger Snake Venom (Notechis scutatus)...... 21 Table 13. Major toxins isolated from the venom of the Coastal Taipan...... 23 Table 14. Major toxins isolated from the Brown Snake (Pseudonaja sp.) ...... 26 Table 15. Summary of major toxins from the King Brown Snake a member of the black snake family...... 29 Table 16. Summary of major toxins from the Common Death Adder...... 31 Table 17. Treatment groups used for alpaca immunisation...... 41 Table 18. Mean packed cell volume (PCV) of alpaca prior to starting venom immunisation and at 21 days after collection of 800mL of whole blood...... 50 Table 19. Final results of Median Lethal Dose (LD50) and Media Effective Dose (ED50) for the experimental alpaca polyvalent antivenom compared to commercial monovalent antivenom products from the Commonwealth Serum Laboratory...... 73

Figures

Figure 1. Schematic representation of conventional antibody and naturally occurring heavy- chain antibody in Camelidae. The folded domains of the polypeptide chains are shown by cylinders. ∼S∼ denotes the inter-chain disulfide bonds. (Nguyen et al. 2001)...... 5 Figure 2. Schematic diagram of the different subclasses of camelid immunoglobulin molecules showing IgG1, IgG2 and IgG3. Note that short ‘hinge’ arrangement for IgG3...... 6 Figure 3. Photograph of CSL snake antivenom product range. Included at far right is a vial of CSL Sea Snake Antivenom...... 15 Table 10. Snake antivenom products with veterinary registration in Australia (2012)...... 16 Figure 4. Veterinary antivenoms available within Australia. AVSL (Lismore, NSW) Brown Snake Avtivenom, Summerland Serums (Lismore, NSW) Tiger Multi-Brown Snake Antivenom and Pfizer (Sydney, Australia) Tiger-Brown Combined Antivenom...... 16 Figure 5. Photograph of Australian Tiger Snake and map showing its distribution within Australia...... 20 Figure 6. Photograph of the Australian Coastal Taipan (Oxyuranus scutellatus) and its distribution in Australia...... 22 Figure 7. Photograph of the Eastern Brown Snake (P. textilis) and the distribution of the various Brown Snakes (Pseudonaja species) within Australia...... 25 Figure 8. King Brown Snake, a member of the black snake family (Pseudechis australis) ...... 28 Figure 9. Distribution of the Red Bellied Black Snake and King Brown Snake within Australia...... 28 Figure 10. Photograph of Common Death Adder (Acanthophis antarcticus) and map showing its distribution within Australia...... 30 Figure 11. Alpaca research herd...... 39 Figure 12. Manual restraint of alpaca...... 40 Figure 13. The author and assistant Allan Quirke harvesting whole blood from an alpaca under field conditions...... 42 Figure 14. Freshly collected bag of alpaca blood ready for clotting and serum separation...... 44 Figure 15. Alpaca serum after centrifugation and prior to filtration (LEFT) and immediately after filtration through 0.22um depth filter (RIGHT). The serum on the left still contains a small percentage of red blood cells which are removed by the filter...... 45 Figure 16. Precipitation of immunoglobulin proteins using 50% ammonium sulphate...... 47 Figure 17. Effect of Tiger Snake Venom on serum the muscle enzyme Creatinine Kinase (CK) at time 0 and 24 hrs after immunisation for the first 3 months. Results presented as mean for the Tiger Snake venom treatment group...... 50 Figure 18. Serum CK levels in alpaca receiving the polyvalent venom mixture...... 51 Figure 19. Serum CK levels in alpaca receiving Brown Snake venom only...... 51 Figure 20. Serum CK levels in alpaca receiving both Tiger Snake and Brown Snake venom...... 52 Figure 21. Serum CK levels in alpaca receiving Papuan Taipan venom...... 52 Figure 22. Serum AST levels in alpaca receiving Tiger Snake venom...... 53 Figure 23. Serum AST levels in alpaca receiving polyvalent venom...... 53

Figure 24. Serum AST levels in alpaca receiving Papuan Taipan venom...... 54 Figure 25. Serum AST levels in alpaca receiving Brown Snake venom...... 54 Figure 26. Serum AST levels in alpaca receiving Tiger Snake and Brown Snake venom...... 55 Figure 27. Serum fibrinogen levels in alpaca at Time 0 and Time 24 hrs after the first immunisation...... 55 Figure 28. Total white blood cell counts in alpaca at Time 0 and Time 24 hours after the first venom immunisation...... 56 Figure 29. Antibody responses of alpaca to immunisation with Brown Snake venom (P. textilis) in the polyvalent treatment group...... 57 Figure 30. Antibody responses to King Brown (Black) snake (Pseudechis australis) venom in the polyvalent treatment group...... 58 Figure 31. Antibody responses of alpaca to immunisation with Papuan Taipan venom (O. scutellatus canni) in the polyvalent treatment group...... 58 Figure 32. Antibody responses of alpaca to immunisation with Common Death Adder venom (Acanthophis antarcticus) in the polyvalent treatment group...... 59 Figure 33. Antibody responses of alpaca to immunisation with Tiger Snake venom (Notechis scutatus) in the polyvalent treatment group...... 59 Figure 34. Antibody responses of alpaca to immunisation with Papuan Taipan venom (O. scutellatus canni) in the monovalent Taipan treatment group...... 60 Figure 35. Antibody responses of alpaca to immunisation with Brown Snake venom (P. textilis) in the monovalent treatment group...... 60 Figure 36. Antibody responses of alpaca to immunisation with Tiger Snake venom (Notechis scutatus) in the monovalent treatment group...... 61 Figure 37. Antibody responses of alpaca to immunisation with Brown Snake venom (P. textilis) in the bivalent Tiger/Brown treatment group...... 61 Figure 38. Antibody responses of alpaca to immunisation with Tiger Snake venom (N. scutatus) in the bivalent Tiger/Brown treatment group...... 62 Figure 39. Antibodies to Tiger Snake Venom in animals receiving the polyvalent venom mixture. Data shown out to 300 days...... 63 Figure 40. Average antibody levels to Tiger Snake Venom in animals receiving the polyvalent venom mixture. Data shown out to 300 days...... 64 Figure 41. Appearance of alpaca serum after processing with caprylic acid with serum diluted in a 0.1M sodium acetate buffer. Note the distinct separation of the non-IgG pellet and clear supernatant containing the IgG...... 65 Figure 42. Undiluted alpaca serum mixed with caprylic acid at 3% (left) and 6% (right) caprylic acid concentration. Note the incomplete separation of the IgG and non-IgG proteins compared to the diluted serum...... 66 Figure 43. Serum protein electrophoresis pattern of alpaca serum prior to processing serum (scanned by Frank Madaras, Venom Science Pty Ltd)...... 67 Figure 44. Serum protein electrophoresis pattern of caprylic acid fractionated alpaca serum (scanned by Frank Madaras, Venom Science Pty Ltd). Note the reduction in the albumin peak...... 67 Figure 45. Further purification of the caprylic acid precipitated serum with Protein G. Note the single large immunoglobulin peak on the serum protein electrophoresis gel scan (scanned by Frank Madaras, Venom Science Pty Ltd)...... 68

Figure 46. Serum protein electrophoresis density scans of Polyvalent Alpaca Antivenom. Serum was heat treated and processed using ammonium sulphate. Note the large reduction in albumin content of the processed product...... 69 Figure 47. Effect of heat treatment of raw serum to 56°C prior to processing. Vial on the left has been heat treated while that on the right received no heat treatment. Note the ‘murky’ appearance to the vial on the right which has higher turbidity...... 70 Figure 48. Heat treatment of raw serum. Bottle on left has been heated to 56°C for 60 minutes whilst bottle on right has not been heated. Note the colour change...... 71 Figure 49. Current map produced by the National Arbovirus Monitoring Program managed by Animal Health Australia. Note the FREE zone located in the southern half of Australia. This would be the preferred location for any serum producing herds...... 81

xii

Executive summary

What the report is about

This report documents the findings on the use of alpaca for producing a therapeutic immunoglobulin product (snake antivenom), for treating snake bite envenomation.

Who is the report targeted at?

This report is targeted at those interested in alternative uses to fibre production for alpaca in Australia. The report has been prepared to be read by someone with a basic understanding of science.

Where are the relevant industries located in Australia?

Globally, there are 3.5 million alpaca. Within Australia, there are approximately 2,500 stud farms with over 100,000 registered stud alpaca. This is the largest registered herd of alpaca worldwide.

Alpaca studs are located throughout New South Wales (NSW), Victoria (VIC), South Australia (SA), Western Australia (WA), Queensland (QLD) and Tasmania (TAS), but not the Northern Territory due to the dry, arid climate and wet summers. The majority of registered breeders are located in NSW and VIC.

NSW is divided into five alpaca breeding regions; central coast and Hunter region, western NSW, the Blue Mountains, southern NSW, and Sydney. The southern region of NSW produces some of the nation’s finest fibre, and contains the state’s largest number of registered breeders, at approximately 150. The central coast and Hunter region is also a thriving area for alpaca breeding, with over 90 breeders, the majority of which have and average herd size of 25 to 40 alpaca. However, there are several studs with in excess of 80 alpaca. Although Sydney is the birthplace of the alpaca industry, limited land prevents large herd sizes and consequently, there are 40 breeders each with small herds. The Blue Mountains and western NSW regions have 68 and 46 registered breeders respectively, the majority of which have relatively small herds.

Victoria contains almost 25% of all registered alpaca studs in Australia. It is divided into three alpaca breeding regions; central VIC, eastern VIC and western VIC. Central VIC has over 300 registered alpaca studs, making up more than 10% of all Australian breeders. The eastern region is geographically diverse, comprising many Australian terrains, fromrich farming country to vineyards. It has over 300 registered breeders, some which are hobby farmers with just five or six animals, and others are established breeders with over 300 animals. Western VIC produces some of the finest alpaca fleece in Australia. It has just 90 breeders, but their animals represent just over 5% of the national herd.

SA has a broad range of soil types and climates. The most suitable region for alpaca breeding is the southern half of the state, where the climate is cooler than the desert region. There are over 100 registered breeders in SA, and the numbers continue to grow. Whilst some of these breeders have small herds of 25 animals, there are a large number of breeders with large herds. One of the largest single herds of alpaca is found in SA.

In QLD, alpaca breeding is concentrated in the southern region where the climate is cooler and less tropical. There are just 60 registered breeders in QLD, the majority of which have small herds, although some have in excess of 80 alpaca. The number of registered breeders is similar in TAS, reaching just 50. In WA, alpaca breeding is concentrated in the coastal plains north and south of Perth. There are no breeders in the northern part of the state, due to the hotter, drier climate. There are over 200 registered breeders in WA, though most are small hobby farms with small herd sizes.

xiii

The benefactors of this research will be alpaca producers, biopharmeutical industry investors, and the Australian alpaca industry as a whole. Alpaca growers may have a new market for the sale of live animals that fit the required specification for serum production. Typically, this would be a young mature, large framed, low disease risk male castrate or female alpaca. These alpaca would be located in herds set up for serum production, that have applied appropriate biosecurity measures and have the required handling facilities.

A possible payment system could be implemented by a pharmaeceutical manufacturer, whereby serum could be purchased on a contracted, per litre volume basis, in much the same way milk is purchased from dairy farmers by the various milk processors. If the alpaca serum industry expanded to similar numbers to the sheep serum industry in Australia, over 10,000 animals could be required for a large scale commercial operation, and also require regular herd replacements. This creates an opportunity for growers to supply animals, or formulate their own serum producing herd.

Investors in the global biopharmaceutical industry will have a new production system for making low disease risk camelid-type antibodies from alpaca serum. Australia’s low disease risk status for the spongiform brain diseases means it is an ideal location for these herds. The global therapeutic antibody industry is also rapidly expanding, and is in search of new products for both human and animal therapies. The Australian alpaca industry as a whole stands to benefit from greater recognition of the unique properties of these animals, and their potentially important contribution to global medicine.

Background

Australia has a large population of alpaca. The traditional product from this animal is its fleece, which is processed to produce high quality fibre. In recent years, a market has existed for alpaca as flock herd protector animals, inserted into sheep and goat flocks. A market exists for meat products from alpaca, but is yet to realise any significant commercial potential in Australia. Alpaca belong to the South American Camelid family along with llama, vicuna and guanaco. Other members of the camelid family include the Bactrian camel (two humped) and dromedary camel (single humped).

All members of the camelid family have been shown to produce unique immunoglobulin molecules, which are different from conventional species such as sheep and horses. Immunglobulins are produced by the body in response to foreign proteins such as virus, bacteria and toxins. Immunoglobulins are a key part of the mammalian immune system, and can bind to various substances, which are then presented to other parts of the immune system for clearance. Camelids have been shown to produce an immunoglobulin molecule, in varying proportions between the different species, which is smaller and devoid of the ‘light-chain’. This unique class of immunoglobulin has been shown to be more heat stable, more active in neutralising certain toxic enzymes, and less likely to cause allergic reactions in humans. These properties are intriguing to explore as they may pave the way for safer and more effective therapeutic antibody products.

Immunoglobulin products have been used in human and veterinary medicine for over 100 years. There are many well developed applications. Prior to the development of antibiotics (i.e. penicillin), the only treatments available for severe infections were horse immunoglobulin products. These proved very effective for infections, and were widely adopted. The development of antibiotics in the 1930s and 1940s provided newer, non-animal derived treatments, and the animal immunoglobulins were largely superseded. However, there has been renewed interest in using immunoglobulins for the treatment of multi-drug resistant bacterial infections. Other uses for immunoglobulins include for tetanus prophylaxis, and rabies post-exposure prophylaxis. Perhaps the most well-known uses are in the antivenom for various snakes, spiders, scorpions, marine creatures, and other toxic and venomous creatures.

Camelid immunoglobulins have been shown to result in a smaller immune response when examined under various testing models. Immediate type allergic responses may occur following administration

xiv

of equine proteins. This can range from mild skin itchiness, to fatal anaphylaxis. A second type of immune response called Serum Sickness can occur 7-14 days post-administration of foreign protein. This is due to the development of antibodies in the recipient against the foreign protein. Whilst not life threating generally, it is very unpleasant, and results in significant morbidity. Camelid immunoglobulins have the potential to reduce these untoward side effects.

The heat stability of camelid immunoglobulins is another property that offers some advantages over conventional immunoglobulins. Heating of camelid immunoglobulins to 70-80 degrees Celsius has been shown to have far less detrimental effect than it would on equine or ovine products. This may offer hope of developing products that do not require refrigeration, such as in tropical regions of the world. It also may lead to a thermal pasteurisation step being included in the processing. Sterilisation of the final product by heat is a very economical means of inactivating potential pathogens and contaminants.

The World Health Organisation currently recognises a crisis in the poor availability of antivenoms, in parts of the world where snake envenomation is a serious public health issue. Typically, this is in developing nations. Major geographical areas where availability is limited include sub-Saharan Africa, south east Asia and India.

Aims/objectives

This project was undertaken as a proof-of-concept study, to demonstrate that alpaca could be used to make potent therapeutic antibodies. The specific therapeutic application chosen was snake antivenom. Antivenom must demonstrate effective anti-lethality properties. This was considered to provide perhaps the strongest form of evidence that the alpaca antibodies are effective in neutralising lethal toxins. To test this hypothesis, a number of other technologies and methods had to be developed. These included a method of safely immunising the alpaca against these highly toxic substances. The health and welfare of the animals had to be monitored. The antibody responses of each animal had to be measured. A method for large volume blood collection, serum separation and filtration had to be tested. An ELISA was developed to measure individual antibody responses over time. Appropriate licenses were obtained to perform the necessary mouse testing of the final antivenom. Physiochemical tests were also performed. A method was developed for processing the raw alpaca serum into a concentrated immunoglobulin product. All of these objectives were successfully met.

Methods used

This was a world first study in alpaca immunoglobulins, and many methods had to be developed. A group of 20 alpaca, located in Bairnsdale, Victoria, were randomly allocated into five groups of four animals each. Each group received a different type of venom. One group received Tiger Snake venom, one group received Brown Snake venom, one group received Taipan venom, and one group received both Brown and Tiger venom. Finally, another group received five types of venom including Tiger, Brown, Mulga, Death Adder and Taipan. Animals were immunised regularly using an oil based adjuvant. Blood samples were collected regularly to monitor animal health and immune responses. Serum was harvested into specialised blood collection bags. The serum was separated and processed using different methods. A final, stable antivenom product was developed and formulated. This polyvalent product was tested in mice. Mice received a multi-lethal dose of venom and varying amounts of antivenom. A measurement of the potency was made by direct comparison to results obtained in mice at the same time, using a commercial antivenom product made by CSL.

Results/key findings

There are a number of key findings. Alpaca responded well to venom immunisation, with minimal effects on their general health and welfare. The ELISA results showed that every animal responded to venom immunisation with high levels of specific antibodies by three months. Antibody levels gradually increased as more immunisations were given. Bulk blood harvesting was performed

multiple times, without any health effects on the animals. Serum was separated and processed using three different methods. The caprylic acid method was found to be unsuccessful on raw serum, and would only fractionate correctly using very diluted serum. This was unexpected, as the method works well in other species. An alternative method of concentrating the immunoglobulin fraction using ammonium sulphate was found to be successful. Various physiochemical tests on the final polyvalent alpaca antivenom product showed it contained 96% immunoglobulins, and 4% serum albumin. This product was used for mouse testing. Nine mouse challenge tests were performed, to determine venom lethality and antivenom potency. Mice were challenged with five lethal doses of venom mixed with antivenom at different dilutions. The results of the mice challenge study were used to determine the final potency of the product. A reference antivenom was used in each bioassay for comparison. The final results of testing revealed equivalent potency to CSL Brown Snake antivenom, but a lower potency for Tiger Snake and Taipan venom. The antivenom product was stable when stored at 4 degrees Celsius, and no precipitate formed in the vial.

Implications for relevant stakeholders

The key implication from this work is that alpaca can be used successfully to make potent, and physio-chemically stable, immunoglobulin products. Further work is required to improve the potency of the antivenom products against venoms containing a high proportion of low molecular weight toxins, such as Tiger Snake and Taipan.

The implications for the alpaca industry are that these animals are potentially commercially viable for serum production on a large scale. More work is required to upscale the production system to a commercial size. Due to disease risk concerns, it is likely that only alpaca herds located outside of the designated Arbovirus zone in Australia would be able to participate. However, the bulk of the alpaca industry is located in the free region.

Although this project has successfully produced antivenom that can effectively neutralise venom, significantly more work is required before such a product could be commercialised. Snake antivenom are perhaps the most difficult of the therapeutic antibody products to develop, because of the detailed testing and high level immune response that is required from the animal. Nevertheless, alpaca can make antivenom, and this paves the way for exploring other therapeutic antibody products that do not require as rigorous immunisation and testing.

Recommendations

A number of projects are required for Phase 2 to fill the knowledge gaps, and bring alpaca derived immunoglobulins closer to commercial reality. Perhaps the most important of these new projects is to understand the patent landscape relating to camelid antibodies. If there is not freedom to operate, then commercialisation would be difficult. Other key pieces of work required are a comprehensive disease risk assessment, and the development of risk based biosecurity management protocols. Disease transmission risks must be identified and managed, for commercialisation and marketplace confidence to occur in alpaca derived serum products. In respect of the physiochemical properties of alpaca immunoglobulins, there are almost no studies specifically relating to alpaca, despite the fact that alpaca belong to the camelid family. Work must be done to show that alpaca immungolobulins do indeed possess the unique properties described for camel immunoglobulins, such as lower allergenic potential, and increased thermal stability. Once these properties are better defined for alpaca, this sets out the advantages of the alpaca serum as a production system for immunoglobulins, whatever the therapeutic application. There are many applications for alpaca immunoglobulins other than snake antivenom. It is necessary to start exploring these, to better understand the market for such products. To bring a product to market requires meeting all the necessary regulatory compliance issues. These need to be understood, and strategies developed to align with these.

Alpaca serum also appears to require different processing conditions than what is typically used for horse immunoglobulins. The laboratory processing method developed in this project resulted in a

xvi

satisfactorily concentrated immunoglobulin product. However, more work needs to be done to translate this bench-top method into a commercial scale system. This could be done on a contract basis with a suitable bioprocessing partner company that has the necessary regulatory and technical know-how.

xvii

Introduction

Mankind’s primordial fear of snakes, and the dire consequences of snakebite, have driven research to develop ever more effective treatments. Snake antivenoms were first developed in the 1890s. These early products were no more than crude serum obtained from horses. The horses were progressively injected with ever increasing doses of venom. Although effective, large volumes of this antiserum were required and some patients developed immediate and life threatening allergic reactions to the foreign horse proteins. Some patients subsequently developed an unpleasant delayed illness called ‘Serum Sickness’, 10-14 days after receiving the antivenom.

Over time, antivenoms have been made more potent through better immunisation regimes and immunoglobulin concentration methods, thus requiring lower therapeutic volumes, and consequently less foreign protein is administered. Chemical treatments to the raw serum were developed to concentrate the specific molecules (immunoglobulins), and reduce the likelihood of immediate and delayed type allergic reactions.

The horse has been the dominant animal used throughout the history of antivenoms. Horses are ideal in some ways because large amounts of serum can be harvested, they can be trained to tolerate blood collection and injections, and they respond well to a range of venom types. This project has explored an alternative animal – the alpaca – for snake antivenom production. This work is a world first, as never before has an alpaca been used for such a purpose.

Aims of this project

This project was designed as a ‘proof of concept’ study, to determine if alpaca can be successfully used for production of therapeutic antibodies. The specific product to be evaluated was snake antivenoms. Snake antivenoms are well defined immunoglobulin products, with lifesaving effects. However, the bigger picture is an exploration of polyclonal antibody production.

Polyclonal antibodies are molecules that are produced against a specific target, be that a toxin or other substance. Currently, horses are the species that has been used for the majority of products. Other species such as sheep and goats have been used, but to a lesser extent. There has been no work done on alpaca.

On one level, this project is an exploration of antivenom production. However at another level, it is a first look at the issues associated with more widespread use of the alpaca for other immunoglobulin products.

This project has broken new ground in the field of antibody production. Never before has the alpaca been used for snake antivenom production. Consequently, there have been a lot of new techniques and methods to develop for this species.

The alpaca and other camelids

The alpaca is a member of the camelid family. The camelid family is divided into New World Camelids which includes Alpaca, Llama, Vicuňa and Guanaco. The Old World Camelids include the Bactrian camel (double hump) and Dromedary camel (single hump), and the limited population of Mongolian camels. The Vicuna is the smallest of the New World Camelids, whilst the Llama is the largest (Fowler 2011).

The alpaca can be divided into two distinct breeds. The Suri has fibre which lacks a crimp, and hangs in ringlets. The Huacaya has fleece with fibres with a light crimp, and the fleece grows at 90 degrees to the body.

The alpaca originates in the high Andean plains in Peru. The alpaca has evolved in a mountainous environment, with a typical elevation of 4,000-4,900m.

An alternative use for alpaca in Australia

The key driver for the alpaca immunoglobulin project was to explore an alternative commercial use for alpaca in Australia. Within Australia, there are approximately 2,500 stud alpaca farms, with over 100,000 registered stud alpaca. Australia has the largest registered herd of alpaca worldwide. However, the average stud herd size is only 40 animals. Thus, the alpaca industry is very diverse, and unit holdings are generally small relative to other production animal industries.

Many mature industries have originated from this model. For example, the dairy industries in Australia, Europe and the UK, were originally based on many small herds. As competition increased, and margins diminished, this family farm business model has largely given way to bigger herds and fewer farmers. For many industries, this is an inevitable pathway to ensure commercial survival.

The primary products of the alpaca are its fleece and, to a lesser extent, its meat. In Australia, the market for alpaca meat products is only in its infancy. The other market is in elite genetic material: high value animals that are offered for sale. This elite genetic market is probably not sustainable as the industry develops. The sale of animals as herd protector animals is another market for live animals.

Fibre from alpaca can be of very high quality. Very low micron (measure of fibre diameter) alpaca fleece can be sold for relatively high prices. Prices paid vary. Fleeces with fibre diameters under 16 micron attract the highest prices. Fibre diameter below 19 micron is highly desirable. Fleece weights vary between 1kg and 5kg, with average around 2kg. Prices paid for fleeces depend on the fibre diameter (see prices in Tables 1 & 2 below from two alpaca wool buyers).

Table 1. Prices offered to growers for alpaca fleece by Alpaca Ultimate (www.alpacaultimate.com.au) in 2012.

Fibre Diameter in Micron Price per kg Under 20 $14.00 20 – 21.99 $14.00 22 – 23.99 $10.00 24 – 25.99 $8.00 26 – 28 $6.00

Table 2. Prices offered to alpaca fibre growers by House of Alpaca (www.thehouseofalpaca.com.au) in 2010.

Fibre Diameter in Micron Price per kg 16-18 $44.00 18.1-20 $36.00 20.1-22 $21.60 22.1-24 $15.60 24.1-26 $12.00 26.1-28 $9.00 28+ $3.00

Worldwide, there is a global biopharmaceutical industry already producing various types of therapeutic antibodies. These antibodies are recovered from the serum of animals that have been ‘hyperimmunised’ to produce higher levels than normal of specific immunoglobulins. These hyperimmune serum industries primarily use horses for antibody production. Sheep are used to a lesser extent. Because alpaca belong to the camelid family, it is likely they possess some or all of the unique properties discovered with camel antibodies.

Unique aspects of camelid immunoglobulins1

Discovery of the structure of heavy-chain only camelid antibodies

In 1989, a project student working at Vrije Universiteit Brussel in Belgium for a team led by Raymond Hamers, was researching how camels fight off parasites. In one of the tests, it was noticed that the camel seemed to have a class of antibodies that were smaller than the conventional Immunoglobulin G (IgG). By 1993, the preliminary investigations were completed and the results published: camelids possess a unique class of IgG lacking light chains (Hamers-Casterman et al. 1993).

Conventional antibodies are produced by B cells, with each cell producing antibodies specific for one epitope. Individual specificity is due to variations in the residues that form the loops within both the heavy and light regions towards the tip of each arm of the antibody, shown as VL or VH in Figure 1. These regions, known as complementarity-determining regions (CDRs), are responsible for forming the unique shape that will bind a particular toxin or other molecule type. However, since the CDRs from both the VL and VH form the antigen binding site, then both heavy and light chains together are involved in binding, rather than individually.

From this, it would be expected that, since the unique camelid antibodies known as heavy-chain only IgG (HCIgG) lack the light chain, they would have a poorer ability to bind. This is not the case: it has been shown that HCIgGs can bind with high specificity and affinity to a wide range of antigens, from the small (haptens and peptides) to the large (viruses and proteins) [for references see (Cook 2010)].

1 Content within this section has been adapted from the thesis and published work of Dr Darren Cook, University of Liverpool, UK.

Despite containing only three of the six CDRs from conventional antibody antigen binding sites, HCIgG have comparable affinities. HCIgG have evolved to overcome this lack of three CDRs by elongating the remaining CDR loops. CDR1 and 2 are longer and more flexible in HCIgG than conventional IgG, where their canonical structure limits the number of possible conformations (De Genst et al. 2006). CDR3 is also extended, providing a greater area for antigen binding, and allowing for the recognition of a greater array of antigens.

Conventional antibodies have an antigen binding surface that is either flat or concave, requiring flat or concave antigens. The CDR3 loop of the HCIgGs can protrude past the surface, enabling binding with clefts and cavities which are inaccessible to conventional antibodies. Such clefts and cavities are of biological significance, often creating the site of interactions such as those between enzymes and substrates, or between ligands and receptors. It has been shown that the CDR3 loop can be shaped similarly enough to mimic the substrate, and enable the VHH to become a potent competitive inhibitor of the relevant enzyme (Lauwereys and Ghahroudi 1998).

The protruding binding site of the HCIgG also may allow recognition of epitopes (antibody binding sites) that are densely packed, and that conventional IgG would be unable to access. Both qualities are of interest when hoping to neutralise the effects of the many proteins and enzymes found in snake venom. HCIgG has successfully been isolated from camels, llamas, and most recently alpaca (Maass et al. 2007). The ratio of conventional IgG to heavy chain IgG varies between the three species (Table 3), with camels containing the greatest proportion of HCIgG.

Table 3. Percentage of conventional and heavy chain IgG typically found in camelids [data from ((Hamers-Casterman et al. 1993, Maass et al. 2007)]

IgG Camel Llama Alpaca Conventional ~25% 55-75% ~50% Heavy Chain Only ~75% 25-45% ~50%

Why these animals have evolved this unique class of antibodies remains a mystery, and whilst immune responses generate antigen specific antibodies of both classes, it is likely that the HCIgG is able to recognise different epitopes to the conventional IgG. Indeed, studies show that the two classes of antibody can respond with different affinites to the same protein. It is therefore likely that the HCIgG broadens the ability of camelids to recognise a greater range of epitopes, and inhibit a greater number of enzymes (Lauwereys and Ghahroudi 1998).

VHH

Most of the research surrounding camelid antibodies has not focused on the whole intact molecule of HCIgG but instead on the binding region referred to as VHH (Figure 1). This is due to the fact that VHH have many unique properties, which may make them ideal for use in a wide range of applications.

VHH are the smallest known intact antigen binding unit. At 15 kDa, they are smaller than Fab fragments (50 KDa), and ten times smaller than conventional IgG. This small size may allow for greater penetration into tissues and organs. Since VHH have evolved not to require attachment to light chains, they are naturally more stable, soluble and much less likely to form aggregates than the VH domains of conventional antibodies or single chain variable fragments (ScFv). VH from conventional IgG uses hydrophobic areas to interact with the CH1 and VL domains. When the VH is separated from these groups, the hydrophobic areas cause the fragment to become sticky, making VH much less effective (Revets et al., 2005).

This structural stability of VHH also extends to extremes of temperature and pH. VHH clones retained binding affinity and specificity during, and after, heating to 90°C (van der Linden et al. 1999) . Similarly, a VHH clone has been developed to treat E. coli infection in pigs via oral application, and has been shown to survive the harsh effects of low pH and proteolytic enzymes (Harmsen et al., 2006).

The simple, sturdy structure of VHH combined with the ease of production of recombinant forms has opened up the potential for improvements in many fields. These include cancer screening (Cortez- Retamozo et al., 2002), cancer treatment (Cortez-Retamozo et at., 2004), affinity chromatography (Kloosteret at., 2007) and as an antiviral (Pant et at., 2006).

It is hoped that some of these properties may also aid the treatment of snake envenoming

IgG1 IgG2 IgG3

Figure 2. Schematic diagram of the different subclasses of camelid immunoglobulin molecules showing IgG1, IgG2 and IgG3. Note that short ‘hinge’ arrangement for IgG3.

Rationale for the Development of Camelid Immunoglobulin Products

Much of the research into camelid HCIgG has focused on the properties of the smaller recombinant VHH. However, there may be properties of the intact HCIgG (derived from the animal) that are favourable to developers of antivenom against snake envenoming and other anti-toxin activities.

Lower immunogenicity

Camel derived IgG has been shown to be less immunogenic when injected into other mammalian species. Camel IgG (containing both conventional and HCIgG) was compared with IgG prepared from horses and sheep in a number of experiments, to determine its immunogenicity (Herrera et al. 2005). To test the likelihood of causing early adverse reactions (EAR) upon the administration of either one of these antivenoms, each species' IgG was incubated with human plasma in order to determine the level of complement activation (thought to play a key role in EAR). Both the horse and the sheep IgG caused complement activation, whilst the camel IgG did not. Another mechanism proposed to initiate EAR is the sensitisation of humans to various animals, through close contact, which allows the person to develop antibodies against the animal. When human IgG was added to ELISA plates containing the antivenoms, horse and sheep IgG again showed a much greater response than camel IgG.

In order to examine late onset reactions known as serum sickness, researchers injected the three antivenoms into mice, and measured the production of mouse IgG that reacted with the foreign IgG. The results showed that camel IgG generated a significantly smaller response than both horse and sheep IgG. These results show that camel IgG has the potential to reduce the incidence of both early and late adverse reactions in patients administered antivenom.

Heat stability

Camel IgG has been shown to be able to withstand higher temperatures for longer times than conventional mammalian IgG. The smaller, recombinantly created VHH is able to withstand even more heat. The heat stability of recombinant camelid VHH is very impressive (Goldman et al. 2008, Cook 2010), retaining the ability to bind even when heated to 90°C (van der Linden et al. 1999).

A heat-stable antivenom would confer a number of advantages. In the PhD work of Darren Cook at the University of Liverpool, a study was completed to explore the heat stability of different classes of camel IgG (Cook 2010). Cook separated camel IgG subclasses against the African snake Echis ocellatus, and heat treated these to either 60 °C for ten hours or 80°C for five minutes. The immune

reactivity of these heated samples was then compared to that of non-heat similar samples. Cook found that the heavy chain only subclasses of IgG2 and IgG3 were markedly less affected by the ten hour heat treatment at 60°C than the conventional IgG subclass (Cook 2010). A similar result was found for the 80°C treatment, with the IgG3 subclass more heat resistant than either IgG1 or IgG2. However, Cook reported that the effort required in separating the IgG subclasses using currently available technology would likely be too slow and costly to justify in a commercial setting.

Safety by heat inactivation of infectious agents

The thermostable properties of camelid IgG2 and IgG3 may prove useful for using heat based methods to reduce the risk of infectious disease spread. There is a theoretical risk of zoonotic transfer of diseases from the immunised animal to the patient. Whilst many viruses are destroyed by caprylic acid precipitation of serum during production (Burnouf et al. 2004), the ability to heat treat HCIgG (coupled with low pH) would greatly improve the safety of the antivenom. Such a step may also reduce the need for further, more complex, purifications, and may help to reduce the price of treatment.

Distribution and shelf life

Certain clones of camel VHH have displayed 100% binding affinity after incubation for one week at 37°C [see (Cook et al. 2010) for references]. It is worthwhile investigating if such a property exists in the native HCIgG, as this could lead to the possibility of an antivenom (or immunoglobulin products) that requires neither refrigeration or lyophilisation (freeze drying). A room-temperature stable antivenom would allow greater distribution to clinics with no refrigeration, making the medicine more available to the rural areas where snakebite is more common. This could reduce the travel time for patients (which can sometimes be days) and improve their prognosis.

It has already been shown that immunisation of camels with a small, weakly antigenic but potent scorpion neurotoxin, enabled the isolation of immunoglobulins that offered good neutralising abilities (Meddeb-Mouelhi et al. 2003).

It has already been demonstrated that camels raise an immune response against E. ocellatus (West African CarpetViper) venom, and that IgG extracted from this offers a protective effect in animal models (Harrison et al. 2006, Harrison and Wernery 2007). Furthermore, following the good response to a poor immunogen in the scorpion study, it is of interest as to whether a camel could mount a better response to N. nigricollis (Black-Necked Spitting Cobra) venom, which has historically produced poor antibody titres in horses (Petras et al. 2011).

Enzyme inhibition

A hypothesis has been put forward that camelid IgGs may be more effective than conventional IgGs for inhibiting enzymes. In particular, the smaller VHH fragments have this property. No evidence for this was found in llama derived IgG (Ferrari et al. 2007). Immunoglobulin obtained from the dromedary camel has been shown to be highly effective in neutralising the enzymatic properties of the enzymes carbonic anhydrase, and pancreatic amylase (Lauwereys and Ghahroudi 1998).

Previous work on antivenom production in camelids

Following the discovery in 1993 of the unique aspects of camel immunoglobulins (Hamers-Casterman et al. 1993), interest has been focussed on how these properties can be harnessed for various therapeutic antibody preparations, including antivenom.

In 2003, a paper was published describing the use of camels for producing scorpion antivenom in Tunisia (Meddeb-Mouelhi et al. 2003). This study was the first to demonstrate the potential of camel

antibodies to bind toxins from a venomous creature. This study demonstrated, for the first time, that camels could indeed mount an immune response to the lethal toxins of the scorpion Androctonus australis. The study demonstrated that a substantial proportion of polyclonal heavy chain antibodies bind to the low MW scorpion toxins. They also showed that these antibodies were capable of preventing death in mice exposed to scorpion venom (Meddeb-Mouelhi et al. 2003).

Following the Tunisian work with scorpion venoms in camels, Dr Robert Harrison, from the University of Liverpool in the UK, immunised camels against snake venom. Harrison used the venom from the African Viper (Echis ocellatus) to immunise camels and llama (Harrison et al. 2006). The study showed that camels (n=2) and llamas (n=2) responded well to immunisation with E. ocellatus venom. The study also demonstrated that these antibodies were effective in neutralising venom induced haemorrhages in mice. More interestingly, they showed that there appeared to be a non- immunoglobulin component in camel serum, that was able to neutralise this venom-induced pathology (Harrison et al. 2006).

Harrison saw the potential of the camelid IgG molecule, and considered it worth pursuing further for antivenom production (Harrison and Wernery 2007). He focussed on camels because they produced over 50% of the camelid type IgG molecule. Building on earlier work, Harrison began immunising camels with venom from three types of African snakes. These were the Puff Adder (Bitis arietans), Desert Cobra (Naja nigricollis) and the Saw-Scaled viper (Echis ocellatus). These snakes were considered the most medically important in Africa. Harrison enlisted a PhD student, Darren Cook, to investigate the properties of the antivenom that was produced. Cook’s thesis was titled “Potential of camelid antibodies to improve the treatment of snake envenoming” (Cook 2010). Around this time, there was also considerable interest in producing recombinant VHH fragments in yeast for various purposes. Much of the work on VHH was patented by the Belgium based biotech company Ablynx.

Researchers in Bolivia also separately investigated the potential of llama antibodies for making antivenom (Fernandez et al. 2010). The large numbers of llama in South America were obviously appealing. The snake venom used was from Bothrops mattogrossensiss. The results were very good. The llama antibodies proved capable of neutralising the Bothrops venom at four times lower protein concentration than the equivalent antivenom created in a donkey.

In the first paper published by Darren Cook from his PhD work, he described the serological responses of the camels that were immunised against combinations of the three African venoms (Cook et al. 2010). The immunisation program started off using an adjuvant (GERBA) that appeared to result in too rapid release of venom into the camel’s circulation, and signs of toxicity occurred in the animals. This adjuvant was later changed to Freund’s Incomplete Adjuvant for all subsequent injections (Cook et al. 2010). Camels were immunised for 60 weeks and received a total of 13 injections, of various doses of venom. Seven of the eight camels immunised were considered to have developed satisfactory immune responses to the venom.

In a second paper published by Cook on the camel antivenom work, he described the capacity of the camel antiserum to neutralise venom induced pathology in mice (Cook et al. 2010). Neutralisation of the pathological effects of venoms from E. ocellatus, B. arietans and N.nigricollis by IgG from the venom-immunised camels, or commercial antivenom, was compared using assays of venom lethality, haemorrhage and coagulopathy. The E. ocellatus venom challenge results of the E. ocellatus monospecific camel IgG antivenom were broadly equivalent to comparable commercial ovine (EchiTAbG™, Micro-Pharm Ltd, Wales) and equine (SAIMR Echis, South African Vaccine Producer, South Africa) antivenoms. Although the equine antivenom required half the amount of IgG, the B. arietans monospecific camel IgG neutralised the lethal effects of B. arietans venom at one fourth the concentration of the SAIMR polyspecific antivenom (a monospecific B. arietans antivenom is not available). The N. nigricollis camel IgG antivenom was ineffective (at the maximum permitted dose, 100mL) against the lethal effects of N. nigricollis venom (Cook et al. 2010).

Cook’s results of testing the various antivenoms against the venoms showed encouraging results, however it was disappointing that the camel serum was not effective against N. nigricollis. The venom of this snake is very difficult to make antivenom against, presumably due to the low MW toxins. Antibodies against neurotoxins from the Cobra (Naja sp) snake family appear to be particularly difficult to make.

In the third paper published by Cook, he described the separation of different subclasses of camel IgG, and the ability of each subclass to neutralise venom (Cook et al. 2010). The paper also described the effect of heat treatment on the different classes of IgG. The findings were that all subclasses of IgG contributed to the neutralisation of venom induced pathology. The study also showed that heating to 60°C for ten hours, or heating to 80°C for five minutes, had either no effect or minimal effect on camel antiserum when assessed in an ELISA. In contrast to the camel serum, sheep serum heated to 80°C rapidly deformed and turned into a gel like substance (Cook et al. 2010). Cook also described the successful use of caprylic acid for fractionating serum, but details were limited, withno information on the filtering process.

Cook has not published any further papers on the use of camel serum for antivenom production. There have been no further publications by any author on the use of camelids for antivenom production.

Merits of alpaca for antiserum production

Alpaca have great potential for use as ‘workhorses’ in the antiserum production industry. Firstly, the animal itself has advantages in terms of its physical resilience and simple management. Secondly, alpaca immunoglobulin molecules are unique, offering inherent properties of thermal stability and molecular structure.

The traditional animal for large volume antibody production has been the horse. However, many animal species have been used to make antivenoms (Russell 1988). Of nearly 200 different antivenoms produced worldwide, only six are not produced in horses (Landon and Smith 2003). Other species used have included sheep, goats, donkeys, pigeons, rabbits, horses, cattle, chickens, guinea pigs, llamas, camels, cats and dogs. Each species has advantages and disadvantages. The following is a discussion of the merits of alpaca.

Alpaca are hardy animals, having evolved at 17,000 feet elevation on the Peruvian Altiplano, and so are somewhat adapted to harsh environmental conditions. In comparison to sheep, alpaca are very resistant to fly strike. Fly strike is a major problem for sheep, and requires preventative measures such as chemicals and mulesing. Alpaca do not require mulesing or crutching. Alpaca do not suffer with the same intestinal parasitism problems that afflict sheep. Resistance to anthelmintic drugs has not been described for alpaca. Alpaca are also resistant to prion diseases which typically affect cattle (Bovine Spongiform Encephalopathy) and sheep (Scrapie). There have been no documented cases of prion disease in alpaca. This makes them very low risk for spreading the spongiform diseases. There are few zoonotic diseases that occur in alpaca. This makes alpaca serum very safe to use in humans and other animals.

Alpaca have a relatively long lifespan. Typical lifespans of 15 to 20 years are reported (Fowler 2011). Sheep are reported to only have a useful lifespan for antibody production of five to eight years (Landon and Smith 2003).

At current prices alpaca are a relatively low cost animal, with male castrates at $200-$400 each. Due to their low cost, poor responding animals could be readily replaced.

A mature alpaca weighs around 75 to 90kg and can readily donate 15-20% of circulating blood volume without any health problems. Alpaca blood contains around 65% plasma and 35% cells. In contrast, horses contain closer to 60% plasma and 40% cells. This means it is possible to recover

more plasma per litre of blood. The lower red blood cell count is probably an evolutionary adaptation to high altitude; red cells are more efficient at carrying oxygen. In diseased states, where the red blood cell counts have fallen to less than 10%, alpaca often appear remarkably healthy (Foster et al. 2009). The shape of the alpaca red cell is such that its surface area to volume ratio is greater than that of most mammals. This improves exchange rate of oxygen across the cell membrane (Foster et al. 2009).

Alpaca can be handled with minimal facilities. Existing sheep and cattle yards can be readily modified to suit alpaca. Alpaca are gentle animals, and require only minimal physical restraint. Alpaca are also easy on fences.

Alpaca contain the ‘camelid’ type immunoglobulin molecule that is devoid of light chains. This has been shown to account for 40% of serum immunoglobulins (Maass et al. 2007).

Procedures such as immunisation and blood collection can be performed by a team of two or three people. Our experiences have shown that 20 animals can be multi-site immunised, and blood sampled, in around 45 minutes. Large volume blood collection (800mL bags) can be accomplished at a rate of around 8-12 alpaca per hour using a team of three people. Under ideal conditions, 60-80 alpaca could have serum harvested per day, using a team of three, with basic facilities.

Polyclonal antibody products

Therapeutic antibody products have been used in human medicine for over 100 years. The ‘father’ of the antiserum industry was the German man, Ernest Behring. In 1890, Behring first published work on the use of antiserum created in horses for treatment of tetanus and diphtheria (Gronski et al. 1991). Human patients with diphtheria were first treated in 1893, with sheep serum that Behring had manufactured. By 1994, over 20,000 doses had been despatched from his production facility.

The first snake antivenoms used in humans were described by the Frenchman Albert Calmette in the 1890s (Hawgood 1999). Calmette used horses for the production of antiserum against the Indian Cobra. Motivated by the number of soldiers dying from Cobra bites, he documented the use of his anti-serum extensively in the military.

These early antibody products were not without problems. A disease syndrome called ‘serum sickness’ often followed about ten days after administration of the horse serum. This was due to the development of antibodies by the patient against the horse serum. Some considered this sickness worse than the original problem that was being treated.

Methods to concentrate the serum and remove the unwanted proteins were developed in the 1890s and early 20th century (Homer 1916). These more refined immunoglobulin products reduced the incidence of serum sickness, allowed the industry to develop, and expanded confidence in the this therapeutic approach. When antibiotics were developed during the 1940s, the use of serum therapy for infectious diseases declined. However, many products still remained.

A wide range of immunoglobulin products has since been developed (see Tables 4-7 below).

Table 4. Therapeutic immunoglobulin products currently available worldwide for treating various envenomation’s (antivenoms).

PRODUCT TARGET

Snake Antivenom AUSTRALIA & PNG: Brown Snake, Tiger Snake, Taipan, Black Snake, Death Adder, Polyvalent ASIA: Cobra(s), Kraits, Russels Viper, Other Vipers.

AFRICA: Cobras, Vipers, Boomslang, Mamba EUROPE: European Vipers USA: Rattlesnake(s) (Crotalus), Coral Snakes CENTRAL/SOUTH AMERICA: Crotalus sp, Bothrops, Coral Snakes, Bushmaster (Lachecis) JAPAN: Mamushi sp. INDONESIA: Multiple species of Cobra, Vipers CHINA: Cobras, Kraits, various other species MIDDLE EAST: Cobras, Vipers

INDIA/SRI LANKA: Cobras, Vipers, Kraits

Spider Antivenom Red Back Spider (Australia) Sydney Funnel Web Spider (Latroectus hasseltii; Australia) Brazilian Wandering Spider (Phoneutria sp, Sth. America) Brown Recluse Spider (Loxosceles laeta; USA)

Black Widow Spider (Latrodectes sp.; USA)

Scorpion Antivenom Various species of scorpions in Mexico, Brazil, Egypt, India, Tunisia, Iran, South Africa.

Tick Antivenom Tick paralysis (Ixodes holocyclus; Australia)

Marine Antivenom Box Jellyfish (Class Cubozoa; Chironex sp; Australia, Indo-Pacific.)

Stonefish (Synanceia sp; Australia, Indo-Pacific Regions)

Irukandji jellyfish sting (no antivenom as yet but needed).

Table 5. Currently registered therapeutic immunoglobulin products for treating various infectious diseases and bacterial toxins.

PRODUCT TARGET Tetanus antitoxin Post-exposure prophylaxis. Antibodies raised against Clostridium tetani toxin (150kDa). Commonly used product for deep tissue penetrations that are at risk of developing tetanus.

Botulism antitoxin Post-exposure prophylaxis. Seven different types of toxins produced by Clostridum botulinium. Toxins named A, B, C, D, E F, and G. Trivalent (A,B,E) Botulinum Antitoxin is derived from equine sources utilizing whole antibodies. The second antitoxin is Heptavalent (A,B,C,D,E,F,G) Botulinum Antitoxin, which is derived from equine IgG antibodies. This is available from the United States Army. On June 1, 2006 the United States Department of Health and Human Services awarded a $363 million contract with Cangene Corporation for 200,000 doses of Heptavalent Botulinum Antitoxin over five years for delivery into the Strategic National Stockpile beginning in 2007. Botulism defined as a Category A biodefense threat.

Rabies antiserum Post-exposure prophylaxis. Given to patients who have been bitten by dog or other animal species that may carry rabies. Reduces risk of infection developing. Worldwide requirement estimated by WHO at 9 million vials per year. Currently undersupply.

Diptheria antitoxin Upper respiratory tract illness caused by Corynebacterium diphtheria. Bacteria produces a toxin (60kDa). Diphtheria is a serious disease, with fatality rates between 5% and 10%. In children under five years, and adults over 40 years, the fatality rate may be as much as 20%. Antitoxin made in horses against the toxin.

Antibiotic resistant New research field to find solutions to the problem of multi-drug bacterial infections resistant antibiotics. Antibodies promising as an alternative to antibiotics.

Human infectious diseases Cytomegalovirus, Hepatitis-B, Rhesus factor, Respiratory Syncitial Virus, Staph auerus and others.

Table 6. Currently registered immunoglobulin products for treating non-infectious human disease conditions.

PRODUCT TARGET Cancer therapies Various products at different stages of entering marketplace. Aim to control specific types of cancers by targeting with antibodies.

Digoxin antitoxin Medication used to treat heart failure. Accidental overdose can be (DigiFab™) fatal. Antibodies used to bind up excess digitalis. Currently derived from sheep. Registered in Canada, USA, UK and Switzerland.

Table 7. Emerging applications for immunoglobulin and serum products at various stages of registration.

PRODUCT TARGET Bioterrorism & Antibodies useful against a wide range of infectious diseases that could be used for bioterrorism. biodefense Additions to the U.S. Strategic National Stockpile.

Smallpox (‘Vaccinia’) antibodies produced by Cangene Corporation, USA. Also sold to other countries as part of biodefense programs.

*Blood clotting factors *Not directly an immunoglobulin product but large demand in human medicine.

Factor VII, VIII, IX.

Snake antivenoms for human use in Australia

Historically in Australia, the Commonwealth Serum Laboratories (CSL) has been the only supplier of antivenoms registered for human use. The first CSL antivenom developed was for Tiger Snake, and was released to the market in the early 1930s (Anon 1957). CSL has continued to be the only supplier of snake antivenoms that are registered for use in humans. CSL uses horses to produce antivenoms. CSL produces a range of Monovalent snake antivenoms, and a five-snake species Polyvalent antivenom. CSL has a long and rich history in the development of antivenoms. There exists a supply agreement between CSL and the Australian Government for CSL to supply these essential medications free of charge to the health care system.

CSL antivenoms are chemically treated to reduce the allergic response in humans. This process involves enzymatic digestion of the whole IgG molecule to remove the Fc portion, resulting in an antivenom that contains only the F(ab)2 portion (Cameron-Smith et al. 2000). However, this process reduces the potency of the antivenom (Madaras et al. 2005), compared to whole IgG antivenom.

Potency is also a function of the amount of protein per mL of antivenom. The lower the amount of protein, the less likely an allergic reaction is. A whole IgG Taipan antivenom produced in Costa Rica was shown to be as potent as CSL Taipan antivenom, but contained only one third the amount of protein (Vargas et al. 2011). This further emphasises the increased neutralising capacity of whole IgG antivenoms. The Costa Rican antivenom is currently undergoing a human clinical trial in Papua New Guinea.

CSL perform their potency testing in guinea pigs (Grasset 1957). Like many rodents, the guinea pig is very sensitive to the neurotoxins contained in venoms. The use of death or survival following a constant dose of venom, and varying dilutions of antivenoms, is problematic for determining potency. Snake venoms are a complex mix of toxic proteins, with varying chemical properties. The bioassay, whether guinea gigs or mice, probably only gives a reliable measure of the antivenom’s ability to neutralise neurotoxins. Other venom components that affect blood clotting are not necessarily assessed using this bioassay. More specific tests of clotting function show that up to 20 times more antivenom is required to inhibit clotting abnormalities than to prevent death. In a study in anaesthetised dogs injected with Brown Snake Venom and CSL Brown Snake Antivenom, a dose of 25 times the recommended antivenom dose was required to prevent clotting disturbances (Tibballs and Sutherland 1991). A similarly markedly increased dose of CSL Brown Snake Antivenom was required to prevent clotting disturbances in citrated human plasma (Madaras et al. 2005).

Monovalent CSL antivenoms have also been shown to have polyvalent activity to varying degrees (O'Leary and Isbister 2009). Using an ELISA against specific venoms, Isbister showed that all CSL’s range of monovalent antivenoms have activity similar to the CSL polyvalent product (O'Leary and Isbister 2009). This is likely to reflect the fact that CSL has a limited pool of horses, and that they are recycled through different immunisation programs. CSL does not make a label claim that their monovalent antivenoms have polyvalent activity. This is an interesting issue, as if all monovalents are truly polyvalent this could lead to significant problems. For example, if a patient is bitten by a Death Adder but the treating hospital only has CSL Monovalent Tiger and Brown Snake Antivenoms, these would be withheld and either the patient transported to another hospital or complex and expensive antivenom transport arrangements made. However, existing stock the CSL Brown or Tiger antivenoms would likely neutralise Death Adder with similar potency to CSL Monovalent Death Adder.

Table 8. Product range of CSL land snake antivenoms.

Product Snake Species used for Immunising

CSL Tiger Snake Antivenom Notechis scutatus

CSL Brown Snake Antivenom Pseudonaja textilis CSL Taipan Antivenom Oxyuranus scutellatus

CSL Black Snake Antivenom Pseudechis australis

CSL Death Adder Antivenom Acanthophis antarcticus

CSL Polyvalent Antivenom All of the above.

Figure 3. Photograph of CSL snake antivenom product range. Included at far right is a vial of CSL Sea Snake Antivenom.

Table 9. Commonwealth Serum Laboratories, Australia, antivenom product range. One Unit is capable of neutralising 0.01mg of venom. CSL supplies snake antivenoms with a variable fill volume but each vial is filled with a minimum number of Neutralising Units per vial. This can also be expressed as potency in mg of venom neutralised per mL of antivenom which is more widely used internationally.

Snake Vial Volume (mL) Units per Vial Potency (mg/mL) Death Adder 25-26 6,000 2.3 – 2.4 Tiger Snake 9-12 3,000 2.5 – 3.3 Brown Snake 4.5-9 1,000 1.1 – 2.2 Black Snake 30-50 18,000 3.6 – 6.0 Taipan 43-50 12,000 2.4 – 2.8

Snake antivenoms for veterinary use in Australia

Snake bite in animals occurs far more commonly than in humans. In a survey of veterinary practices across Australia, it was estimated that over 6,000 cases of snakebite occur annually in animals (Mirtschin et al. 1998). In comparison, the number of human snakebite cases in Australia is significantly less, at 1,000-2,000 cases (Sutherland 1979, White 1998). The number of human fatalities from snakebite in Australia averages two deaths each year (White 1998).

The snake species responsible for most animal deaths in Australia is the Tiger Snake and Brown Snake (Mirtschin et al. 1998). The Brown Snake is now responsible for the majority of human envenomations in Australia. In published surveys, cats and dogs are the most frequently reported bite victims. Black Snakes accounted for a large proportion of bites in animals, but fewer deaths (Best 1998, Mirtschin et al. 1998).

Veterinarians in Australia may purchase and use the CSL range of human antivenoms. However, recent price increases have made these very expensive medications. For example, a single vial of CSL Polyvalent antivenom can be purchased by veterinarians at a wholesale price of over $2,000 per vial (excluding GST). This product would be the ideal product for all veterinarians to use, as it does not require identification of the snake species, but is a large volume to administer. However, in most cases in southern Australia veterinarians would use a bivalent Tiger and Brown Snake Antivenom because these are the main snake species found.

Table 10. Snake antivenom products with veterinary registration in Australia (2012).

Product Snake Species Product Presentation Preservative Pfizer Tiger-Brown Combined Tiger Snake 20mL glass vial Phenol Antivenom Brown Snake Protein concentration of 2.2mg/mL [equine protein] 17%.

AVSL Multi-Brown Snake Pseudonaja species of 10mL glass vial Phenol Antivenom snakes 5mL glass vial 2.2mg/mL [ovine protein]

Summerland Serums Tiger Tiger Snake 20mL glass vial Phenol Multi-Brown Snake Antivenom Pseudonaja species of 2.2mg/mL [equine protein] snakes

Summerland Serums Multi- Pseudonaja species of 10mL glass vial Phenol Brown Snake Antivenom snakes 2.2mg/mL [equine protein]

Figure 4. Veterinary antivenoms available within Australia. AVSL (Lismore, NSW) Brown Snake Avtivenom, Summerland Serums (Lismore, NSW) Tiger Multi-Brown Snake Antivenom and Pfizer (Sydney, Australia) Tiger-Brown Combined Antivenom.

Medically important venomous Australian snakes

Australian snakes are amongst the most venomous in the world. The toxicity of snake venoms are usually compared based on their Median Lethal Dose in mice. The dose of venom that kills 50% of mice is called the Median Lethal Dose (LD50). A lower LD50, means more toxic venom. The LD50 dose is usually expressed as mg of venom per kg of animal. A typical LD50 test involves injecting six groups of six mice, with graded doses of venom. The mice are usually observed for 24 or 48 hrs, and total number of deaths is recorded. The proportion of mice that die per dose level is analysed and presented as an average dose, plus a 95% confidence interval for the measurement. The confidence interval depends greatly upon the numbers of mice used per venom dose; the more mice, the narrower the confidence interval.

Table 11 contains LD50 data generated by CSL in mice for a range of Australian snakes.

The following section contains a discussion of the different medically important snakes in Australia. Much of this section has been adapted from review papers by Prof. Julian White, Women’s and Children’s Hospital, North Adelaide, Australia (White 1991, White 1998).

Table 11. Toxicity of various Australian snake venoms in 18-21 gram laboratory mice listed in decreasing order of toxicity (Broad et al. 1979). The authors used over 5,000 mice to establish this dataset (note the shaded entries; these were used in the alpaca project).

LD50 mg/kg (95% confidence limits) Snake (common name in parenthesis) 0.1 % bovine serum Saline albumin in saline Parademansia microlepidotus 0.025 (0.020-0.029) 0.010 (0.007-0.014) (Small-scaled snake) Pseudonaja textilis 0.053 (0.041-0.065) 0.041 (0.033-0.051) (Brown snake) Oxyuranus scutellatus 0.099 (0-0814.123) 0.064 (0-052-0-078) (Taipan) Notechis scutatus 0.118 (0.095-0.146) 0.118 (0.088-0-157) (Tiger snake) Notechis ater niger 0.131 (0.107-0.163) 0.099 (0.083-0.120) (Reevesby Island Tiger snake) Enhydrina schistosa 0.164 (0.149-0.185) 0.173 (0.144-0.210) (Beaked sea snake) Notechis ater occidental's 0.194 (0.161-0.234) 0.124 (0.102-0.152) (Western Australian Tiger snake) Notechis ater serventyl 0.338 (0.278-0.414) 0.271 (0.220-0.335) (Chappell Island Tiger snake) Acanthophis antarcticus 0.400 (0-336-0.472) 0.338 (0.2784.413) (Death adder) Pseudonaja nuchalis 0.473 (0.393-0.570) 0.338 (0.271-0-423) (Gwardar) Austrelaps superbus 0.560 (0-448-0.700) 0.500 (0-415-0.605) (Australian Copperhead) Pseudonaja affinis 0.660 (0.550-0.800) 0.560 (0.505-0.620) (Dugite) Pseudechis papuanus 1.09 (0.865-1.35) 1.36 (1.23-1-51) (Papuan black snake) Hoplocephalus stephensii 1.36 (1.12-1.66) 1.44 (1.27-1.63) (Yellow banded snake) Tropidechis carinatus 1.36 (1.19-F56) 1.09 (0.980-1.21) (Rough scaled snake) Pseudechis guttatus 2.13 (1.79-2.51) 1.53 (1.24-1.89) (Blue-Bellied black snake) Pseudechis colletti 2.38 (2.08-2.74) Not done (Collett's snake) Pseudechis australis 2.38 (1.93-2.92) 1.91 (1.57-2.33) (King brown snake) Pseudechis porphyriacus 2.52 (2.09-3.04) 2.53 (2.05-3.14) (Red-Bellied black snake) Cryptophis nigrescens 2.67 (2-40-2.96) Not done (Small-Eyed snake) Demansia olivacea >14.2 Not done (Spotted snake)

Tiger snakes (Notechis sp.)

The tiger snake group contains a diverse collection of species, spread across four genera. Tiger snakes, genus Notechis, generally live in cooler, moister habitats in eastern Australia, including many southern offshore islands, and Tasmania. They are less frequently encountered than previously, but still cause significant numbers of bites. Their fangs are larger and they produce, on average, more venom than brown snakes, but the venom is slightly less toxic. They have a lower dry bite rate and prior to antivenom development, had a 45% fatality rate for bites, compared to 8% for brown snakes.

Bites cause local pain and often local erythema, mild swelling, and bruising. General systemic effects are similar to those for brown snakes, but specific effects include flaccid paralysis. This is common, often severe (caused by predominantly presynaptic acting neurotoxins), and so not reversible with antivenom. Defibrination coagulopathy is seen in almost all cases, with systemic envenoming. Early cardiac collapse is not reported, unlike brown snake bites. However, early collapse (presumed noncardiac) with recovery, sometimes associated with generalized convulsions, does occur. Muscle damage is common and can be severe. Renal failure is uncommon, but does occur occasionally, though not always linked to muscle damage. It is uncommon, but long- term alteration or loss of taste/smell can occur with tiger snake bites.

All cases with significant systemic envenoming require antivenom therapy. In humans, CSL tiger snake antivenom is used with an initial dose of two vials, and possibly more for large specimens found on some Bass Strait islands.

The rough-scaled snake (Tropidechis carinatus) is limited to pockets of distribution along eastern coastal Australia, from north eastern NSW through to north east Queensland. In both clinical and treatment terms, it is essentially identical to the tiger snakes.

Copperheads, genus Austrelaps, are restricted to southeast Australia, Tasmania and Kangaroo Island. Their venom, and particularly their clinical effects, are not well reported. They can cause flaccid paralysis. It is not clear if they are likely to cause coagulopathy or muscle damage. Treatment is similar to that for tiger snakes.

The broad-headed snakes, genus Hoplocephalus, have restricted ranges in portions of eastern coastal NSW and Queensland. Clinically they are similar to brown snakes, causing defibrination coagulopathy, but not paralysis or muscle damage. However, their venom is closer to that of tiger snakes, and treatment is as for tiger snakes.

Tiger Snake venom is highly toxic. LD50 studies in mice reveal a toxicity of 0.118 mg/kg (Broad et al. 1979). There appears to be some variation between animal species in the toxicity with an LD50 described in horses of 0.005 mg/kg (Best 1998), and in monkeys of 0.02 mg/kg (Best and Sutherland 1991). The average venom yield per milking of the Tiger Snake is around 35 mg (Mirtschin et al. 2006).

The most significant components of Tiger Snake venom are the low MW neurotoxic molecule called notexin, and a blood clotting activator (Prothrombin Activator). See table 12. In terms of antivenom production, the low MW toxins are the most difficult to make antibodies against.

Figure 5. Photograph of Australian Tiger Snake and map showing its distribution within Australia.

Table 12. Summary of the major toxins in Tiger Snake Venom (Notechis scutatus).

Snake Species Major Toxins Specific Toxins / MW Actions Tiger Snake Neurotoxins Notexin Pre-synaptic neurotoxin. Single chain PLA2. LD50 in MW =13.5kDa mice of 0.025mg/kg (iv) and 0.006mg/kg. Comprises 6% of crude venom. Similar to Taipoxin from Taipan. (Notechis scutatus) Action is to arrest recycling of vesicles at axolemma, prevents neurotransmitter release. Acts to inhibit release of acetylcholine, resulting in paralysis. Notechis II-5 Pre-synaptic neurotoxin. Seven amino acids different (total 119) from notexin. Also exhibits PLA2 activity and lethality. Scutoxin A and B Toxic peptides of. An isoform of Notexin. Similar MW=13kDa toxicity. Notexin II-5b Non-toxic PLA2 activity. Considerable homology with Notechis II-5 and Notexin. Notechis 11’2 Non-toxic PLA2 activity. Notechis N Similar biological properties to Notexin but differs by one amino acid. Toxins 1 and 2 Post-synaptic neurotoxins. LD50 (iv) in mice of 0.1mg/kg (Toxin 1) and 0.15 mg/kg (Toxin 2). Fast acting toxins, mice dead within 2 hours. MW 6kDa and 7kDa respectively. Effects readily reversed by antivenom (in vitro). Hta-i Toxic proteins. Promote hypotension and MW =18-21kDa haemorrhage in mice. The has 125 amino acids, homology to taipoxin-y. Myotoxins Notexin (see above) Myolytic activity well defined. Causes necrotising myopathy. Potent PLA2 capable of hydrolysing phospholipids in micelles. Myotoxicity is likely to result when treatment with antivenom has been delayed or inadequate. Injection of notexin in monkeys results in CPK elevations within 2 hours. Myoglobinuria and secondary renal damage may occur. Procoagulants Prothrombin activator Depletion of fibrinogen in plasma by activation of MW = 54kDa prothrombin. Prothrombin activator with heavy and light chains. Caused severe depression in blood pressure and cardiac output in anaesthetised dogs. Thrombi formation in the heart. Partial thromboplastin time prolonged in monkeys. Antivenom reversed effects. Hemolysins Less haemolytic activity than Pseudechis family. Red blood cell damage. Hemoglobinuria. Other Hyaluronidase. PLA. PLB. Capillary permeability factor. Lymphotoxic factor. Senory nerve ending toxin. Smooth muscle stimulant. Acetylcholinesterase. Heat-stable coagulant. Peptidase activity. Purine compounds

Taipans (Oxyruanus sp.)

Taipans are among the most feared venomous snakes. They are large, have very potent venom delivered in large amounts through long fangs, have a very low dry bite rate, and a pre-antivenom case fatality rate of >80%.

There are currently at least three recognized species, occupying niches in northern and eastern coastal Australia from NSW to Western Australia (WA) (Oxyuranus scutellatus), in New Guinea (O. s. canni), in central and eastern inland Australia (O. microlepidotus), and in western inland Australia (O. temporalis) (Figure 6).

Taipan bites are associated with variable local effects, from minor to significant local pain and swelling and rapid onset of severe systemic envenoming, with flaccid paralysis, defibrination coagulopathy, and sometimes, secondary renal failure. Myolysis has been reported, but is uncommon to rare.

Virtually all Taipan bites require antivenom therapy, using either CSL Taipan or CSL Polyvalent Snake Antivenom at an initial dose of one or two vials. CSL uses the Australian Coastal Taipan rather than the more toxic Papuan Taipan (Herrera et al. 2012) for its horse immunisation program. Prior to the introduction of CSL’s Taipan Antivenom, almost all bites resulted in a fatal outcome (Morgan 1954).

Taipans are found in Papua New Guinea where envenomation is a common problem. Port Moresby hospital treats dozens of cases each year (Trevett et al. 1995).

The venom of the Taipan is highly toxic to mice, with a published LD50 of 0.064 mg/kg (Broad et al. 1979). The Taipan produces a large quantity of venom, with the average yield being 120 mg (Mirtschin et al. 2006).

Figure 6. Photograph of the Australian Coastal Taipan (Oxyuranus scutellatus) and its distribution in Australia.

Table 13. Major toxins isolated from the venom of the Coastal Taipan.

Snake Species Major Toxins Specific Toxins Actions Coastal Taipan Neurotoxins Taipoxin Potent pre-synaptic neurotoxin. Highly Pre-synaptic neurotoxic. PLA2 activity. Taipoxin is a glycoprotein with three components. LD50 (Oxyuranus MW=45.6kDa. in mice of 0.002mg/kg. Comprises 17% - scutellatus) 20% of venom by weight. The alpha-subunit possesses neurotoxicity and this is very weak Papuan Taipan (500 times weaker than full toxin). Only the alpha and gamma chains provide antibody response. The Beta subunit devoid of (Oxyuranus toxicity can be separated into two subunits of scutellatus canni) around 14.3kDa. The third subunit (lamba) is 26.9kDa devoid of toxic activity but still

has weak PLA2 activity. Causes omega shaped indentations in the axolemma at the endplates of motor nerves. Toxin interferes with recycling of the synaptic vesicles. Similar to notexin. A latency period before toxin is irreversibly bound to nerve ending. α-oxytoxin 6.770kD Postsynaptic neurotoxin. 9% of whole venom. α-scutoxin 6.782kD Postsynaptic neurotoxin. 6% of whole venom. Taicotoxin Complex toxin. Effects on cardiac calcium channels. Composed of three molecular weight entities. Neurotoxic phospholipase (16kDa), serine protease inhibitor (7kDa), a- neurotoxin like peptide (8kDa). Different from taipoxin. Paradoxin (O. PLA2 neurotoxin. Paradoxin - presynaptic elepidotus) neurotoxin, phospholipase A2 based, Pre-synaptic essentially identical to taipoxin. It accounts for 12% of crude venom, is a sialo- glycoprotein with three subunits and has an LD50 of 2 mg/kg (IV mouse). Cannitoxin Isolated from Papuan Taipan. Pre-synaptic β-neurotoxin. [same as taipoxin?] Myotoxins Taipoxin (as above) Also myotoxic to skeletal muscle. Elevated CPK in monkeys. Myoglobinuria possible (not common). O.s.canni does not seem to have the myotoxicity of O.scutellatus Coagulants Purified activator of Causes DIC. Potent converter of PT to 300kDa. thrombin. No anticoagulant activity. Also causes platelets to aggregate. Hemolysins RBC transformation. Hemolysis and anaemia may result.

Brown snakes (Pseudonaja sp.)

Brown snakes, of the genus Pseudonaja, are currently the leading cause of both bites and snakebite fatalities in Australia (White 1998). A variety of species, with variable coloration, are distributed throughout mainland Australia and parts of New Guinea, but not on islands off the coast, including Tasmania. Brown snakes are active diurnal hunters, easily adapting to urban habitats, and are common in many Australian mainland towns and cities. On warm nights, like all other dangerous snakes, they can be active after dusk. They have small fangs, and some deliver only small amounts of venom. The venom is highly potent.

While the majority of bites are dry (approximately 70–80%), when significant venom is delivered, it can rapidly cause severe or lethal envenoming. In particular, early pre-hospital cardiac collapse is associated with brown snake bite, and is frequently fatal. The bite may be painless and go unnoticed, with bite marks hard to detect, but systemic envenoming can develop in minutes, with defibrination coagulopathy the classic feature. Complete defibrination can occur in 15 to 30 minutes post-bite, though can sometimes develop less rapidly. Major intra-organ bleeding is rare unless there is trauma, such as a fall with a blow to the head. As early collapse is common in severe cases, such trauma is always possible, though fortunately rare. Particularly in children, early collapse can be followed by generalized convulsions. Headache, nausea, vomiting, and abdominal pain are all common systemic symptoms. However in adults, there may be absence of symptoms even though there is complete defibrination (White 1998).

In addition to defibrination coagulopathy, there may be renal failure. Flaccid paralysis can occur in humans, but is rare; it is most likely in patients with otherwise severe systemic envenoming and a long delay in initiating antivenom therapy. Myolysis does not occur, though occasionally very mild transient rises in CK are seen. For patients with significant systemic envenoming, notably defibrination-type coagulopathy, antivenom therapy is required. CSL brown snake antivenom is preferred, with an initial dose of two vials now considered adequate based on ongoing clinical research. This is a significant reduction from the previously advocated dose of five or more vials.

The clinical envenomation syndrome differs markedly between humans and animals. In humans, the clotting disturbances are the predominate pathology (Isbister et al. 2007). In animals, clinical signs of neutrotoxicity are the predominate signs (Best 1983, Best and Sutherland 1991). This has been described as the ‘Brown Snake Paradox’ (Barber et al. 2012). Brown Snake venom contains one of the most potent neurotoxins ever discovered, yet this is rarely seen in humans.

The venom of Pseudonaja textilis highly toxic to mice. The LD50 of P. textilis in mice is 0.04 mg/kg, and in horses is 0.02 mg/kg (Kellaway 1931). Fortunately, the average bite from P. textilis contains relatively little venom; the mean dose is 4 mg (Mirtschin et al. 2006).

Figure 7. Photograph of the Eastern Brown Snake (P. textilis) and the distribution of the various Brown Snakes (Pseudonaja species) within Australia.

Table 14. Major toxins isolated from the Brown Snake (Pseudonaja sp.)

Snake Species Major Toxins Specific Toxins Actions Brown Snake Neurotoxins Textilotoxin Pre-synaptic neurotoxin. PLA2 activity. Approx. MW = 74kDa 3% to 5.7% by weight of crude venom. Accounts for 70% of lethality. LD50 in mice (Pseudonaja textilis) Pre-synaptic (iv) 0.001mg/kg. 20% carbohydrate content including sialic acid.

Five subunits. All sub-units shown to contain PLA2 activity. Antitoxin binds strongest to subunit D. Antivenom from Tiger binds to subunit A. Similar homology of subunit A to notexin. Blocks release of acetylcholine. Neuromuscular blockade. No appreciable effect on muscle. Presynaptic effect due to PLA2. Pseudonajatoxin a Postsynaptic neurotoxins. Binds irreversibly to MW=12.28kDa nicotinic acetylcholine receptors. In mice (ip) Post-synaptic LD50 was 0.3mg/kg. 117 amino acids. 7 disulphide bonds.

Pseudonajatoxin b. Postsynaptic neurotoxins. In mice (ip) LD50 MW=7.762kDa was 0.015mg/kg. Blocks acetylcholine Postsynaptic receptors. Binds weakly and reversibly. Procoagulants PT activator enzyme Enzyme accounts for 40% of venom weight. (major toxin) Strongly coagulative activity well known. MW=200kDa Strongest of all Australian elapids with procoagulant effects. Prothrombin activator. Complex of several polypeptide chains. Converts prothrombin to thrombin. Activity enhanced by calcium. Two enzymes have been isolated that do the conversion of prothrombin to thrombin. DIC commonly results and leads to consumption of coagulation factors and spontaneous haemorrhages. Injection into dogs of the prothrombin activator caused severe depression in cardiac output. Thrombocytopaenia and prolongtation of PT and APTT and reduction in fibrinogen. Proposed to cause cardiovascular depression by myocardial dysfunction secondary to DIC. Textarin Serine protease inhibitor. MW=50-53kDa. Textarin accounts for 5% of venom weight. PT converter enzyme (minor toxin) Renal toxicity Venom not thought to be directly toxic to kidney. Probably secondary damage after DIC etc Myotoxins NOT PRESENT

Black snakes (Pseudechis sp.)

The black snake and mulga snake groups, all within the genus Pseudechis, are widely represented throughout mainland Australia and into parts of New Guinea. The black snakes (Red Belly Black = P. porphyriacus and P. guttatus) are restricted to eastern Australia, from NSW through Victoria and South Australia (P. porphyriacus only). These snakes cause local pain, erythema and swelling, and nonspecific systemic symptoms, especially vomiting, abdominal pain, and sometimes diarrhea. These snakes do not cause paralysis or major coagulopathy in humans. They uncommonly cause minor myolysis, and rarely cause major myolysis.

In many cases, antivenom therapy is not required. If it is used, CSL tiger snake antivenom is preferred, except if there is major myolysis, when CSL black snake antivenom should be considered. For either antivenom, the initial dose is one vial.

The mulga snakes (P. australis and P. butleri) and Collett’s snake (P. colletti) cause more severe envenoming, with marked local pain and swelling, and sometimes severe systemic myolysis. Occasionally they also cause an anticoagulant (non-fibrinogen-consumptive) coagulopathy. Secondary renal failure is possible, but rare. Major paralysis is not reported, but minor paralytic effects, specifically ptosis, do occur in a few cases. Any case with significant systemic envenoming requires antivenom therapy, and CSL black snake antivenom is preferred, with an initial dose of one vial. Rarely is more than one dose required.

The Papuan black snake (P. papuanus) from portions of New Guinea is different again in clinical effects, as it can cause coagulopathy, paralysis, and myolysis. The preferred antivenom is CSL black snake antivenom, but in New Guinea this is rarely available, so the equally effective CSL polyvalent snake antivenom is used. The initial dose is one vial.

Pseudechis australis produces a large quantity of venom, with average yields of around 180 mg per milking described (Mirtschin et al. 2006). The LD50 of the venom in mice is described at 2.38 mg/kg (Broad et al. 1979). The Red Bellied Black snake causes severe muscle damage. The LD50 of the red belly black snake has been reported as 2.52 mg/kg (Broad et al. 1979).

Figure 8. King Brown Snake, a member of the black snake family (Pseudechis australis)

Figure 9. Distribution of the Red Bellied Black Snake and King Brown Snake within Australia.

Table 15. Summary of major toxins from the King Brown Snake, a member of the black snake family.

Snake Species Major Toxins Specific Toxins Actions Black (King Myotoxin Mulgatoxin Myopathy. LD50 in mice of 0.2mg/kg Brown/Mulga) Snake (P.australis) (ip). Basic toxin (PLA2). MW=13.484 kDa. Myoglobinuria. CPK elevated. May also produce local tissue swelling. (Pseudechis australis) Neurotoxin LD50 0.076mg/kg in mice (iv). 118 amino acid sequence. Postsynaptic neuromuscular blocking effects. Only moderate neurotoxicity.

Pseudechotoxin Inhibits current through nucleotide gated MW=24kDa. ion channels.

Pseudexins A,B,C Pseudexins 25% of whole venom by (isolated from red belly weight. LD50 in mice (ip) of 0.48mg/kg. black. P.pseudechis) PLA activity.

MW=16.5kDa.

Death adders (Acanthophis sp.)

Death adders, genus Acanthophis, used to be widespread across mainland Australia and throughout New Guinea. However, in southern Australia, they are becoming less common as a cause of bites, except from captive specimens. They had a fearsome reputation, with a low dry bite rate and a pre- antivenom fatality rate of 50%. However, their principal venom components affecting humans are postsynaptic neurotoxins, so the predominant clinical feature of envenoming is flaccid paralysis. The onset of this paralysis is sometimes rapid, but sometimes delayed >24 hours. They do not cause coagulopathy, myolysis, or renal failure.

Antivenom therapy is required in cases developing paralytic features, the preferred choice being CSL death adder antivenom at an initial dose of one vial. Severe paralysis may occasionally require substantially higher doses before reversal. Neostigmine has been used successfully as an adjunct to antivenom to reduce the severity of paralysis in death adder bites. Some death adders also possess presynaptic neurotoxins.

Death Adder venom is highly toxic with an LD50 in mice of 0.4 mg/kg (Broad et al. 1979). Typical venom yields per milking are around 85 mg (Mirtschin et al. 2006).

Figure 10. Photograph of Common Death Adder (Acanthophis antarcticus) and map showing its distribution within Australia.

Table 16. Summary of major toxins from the Common Death Adder.

Snake Species Major Toxins Specific Toxins Actions

Common Death Neurotoxins Acanthophins Post-synaptic neurotoxicity. Probably only Adder post-synaptic. Acanthophin a MW= 7.7kDa. Single chain of 63 AAs. (Acanthophis LD50 in mice (ip) 0.16mg/kg. antarcticus) Acanthophin a c

Acanthotoxin PLA2 activity MW=13kDa. High degree of homology with other elapid PLA2 neurotoxins. Platelet Acanthin I and II Inhibit platelet aggregation. PLA2 activity. inhibitors Incomplete PT activator.

Weak haemolytic activity. Myotoxins Not found.

Coagulant Feebly coagulant. Weak anticoagulant activity in vitro.

Sea snakes

Sea snakes represent a diverse lineage of elapid snakes that have adopted a marine lifestyle, radiating from their presumed evolutionary origins in the Australian region to inhabit tropical waters from the Middle East, across the Indian Ocean, and much of the western Pacific. One pelagic species even extends across the Pacific to the Americas. Bites generally occur only when the sea snake is molested, notably when caught in fishermen’s nets. Changing fishing practices has reduced the number of bites to fishermen in recent years. Many sea snakes have very potent venom, with two distinct clinical effects: flaccid paralysis, postsynaptic, and systemic myolysis, sometimes with secondary renal failure. Most cases of sea snake bite result in either paralysis or myolysis, but rarely both together.

Cases with significant systemic envenoming require antivenom therapy. CSL sea snake antivenom is used at an initial dose of one to three vials, depending on severity of envenoming. Neostigmine can be used as an adjunct to antivenom in cases with paralytic envenoming. In the past it has been recommended that if CSL sea snake antivenom was unavailable, CSL tiger snake or polyvalent snake antivenom could be used. Changes in antivenom manufacturing processes at CSL mean that this substitution is less certain to be effective now.

Sea snakes are of no relevance to the veterinary profession. Sea snake envenomations are primarily an occupational hazard of fishermen in warmer Indo-Pacific waters.

Actions of Australian snake venom toxins

Neurotoxins

Classic discussions of Australian snakebite have emphasized toxins that affect the nervous system (neurotoxicity) as the leading clinical problem and cause of death (White 1998). One hundred years ago this may have been true, but it is no longer valid (Sutherland 1992). Most Australian terrestrial elapid venoms contain a mixture of both types of neurotoxins, though a few species have only postsynaptic neurotoxins, and a few show no neurotoxicity in humans at all, though they may cause paralysis in some mammal species.

The clinical effects of these different types of neurotoxins are essentially the same, but with one important difference from a treatment perspective (White 1998).

Presynaptic neurotoxicity is associated with damage to the terminal axon at the neuromuscular junction, and therefore established paralysis will not be reversed by antivenom.

Postsynaptic neurotoxicity is not associated with damage to either the terminal axon or the muscle endplate, and so is potentially reversible with antivenom therapy, dependent on the reversibility of binding between the toxin and the acetylcholine receptor on the muscle endplate.

The first signs of flaccid paralysis are usually seen >1 hour post-bite, sometimes as late as 24 hours post-bite. In humans, the progression from first signs to full respiratory paralysis usually takes >6 hours, often >12 hours, though can occasionally occur more rapidly. Only some cases with early signs will show progression to full paralysis. First signs are seen in the cranial nerves, starting with eyelid drooping, progressing to fixed forward gaze, fixed dilated pupils, loss of facial expression, reduced tongue extrusion, and then loss of airway protection and onset of drooling. Paralysis may then progress to involve all limbs, with weakness and loss of deep tendon reflexes. Respiration and diaphragmatic function are last to be affected (White 1998).

Myotoxins

Myotoxicity (muscle damaging toxins) are a prominent feature of envenoming by a limited subset of Australian elapid snakes, most notably the tiger snakes and rough-scaled snake (Notechis spp., Tropidechis carinatus) and mulga snake/Collett’s snake subgroup (Pseudechis spp.).

The myotoxins affect skeletal muscle. Clinically this presents as muscle pain, tenderness, pain on contraction against resistance, and myoglobinuria, often visible to the naked eye.

Onset is variable, but always >1 hour postbite, generally >6 hours postbite, and occasionally >24 hours postbite. Two secondary complications can arise: renal failure and elevated serum potassium (hyperkalaemia). Both of these complications, especially the latter, are potentially lethal (hyperkalemia can cause cardiac arrest).

Coagulalopathy

Snake venoms, notably viperid venoms, target diverse points in the complex hemostatic pathways. The endpoint is usually coagulopathy with increased bleeding, though a few species cause thrombosis (White 1998, Paul et al. 2007). Many, but not all, dangerous Australian elapid snakes have toxins targeting points in the hemostasis process. Indeed, it is current wisdom that the resultant coagulopathy is the leading cause of snakebite deaths in Australia (White 1991, White 1998).

Australian venoms affecting coagulation fall into two broad categories: procoagulants and anti- coagulants. Procoagulants are represented as the clotting enzyme (prothrombin) converter, and cause a frequently rapid syndrome and consequent propensity to bleed. This is the most common clinical problem encountered.

Nephrotoxicity

There are no known primary kidney toxins (nephrotoxins) in Australian elapid venoms. However, kidney damage is not rare in major snakebite, particularly with brown snake (Pseudonaja spp.) bites (White 1991, White 1998).

Cardiotoxicity

No primary cardiotoxins have been reported from Australian elapid venoms. However, secondary cardiotoxicity is known (White 1991, White 1998).

Local Effects

Australian elapid snakes do not cause the moderate to severe local tissue injury commonly seen outside Australia with some elapid bites (such as cobras, Naja spp., in Africa and Asia), and many viperid bites (White 1991, White 1998).

Overview of the antivenom production process

The following section provides a brief overview of the antivenom (immunoglobulin) production process in animals.

Hyperimmunisation of animals

To generate specific antibodies in high concentration, the donor animal must be repeatedly injected with venom from the specie(s) of snake that neutralising antibodies are sought. In most cases, venom is mixed with an immune stimulant (adjuvant) to boost the immune response (Standardization 2010), although some antivenom manufacturers continue to use venom in saline or aqueous adjuvant. Precise protocols are often not published and are considered trade secrets.

The interval between injections is variable depending upon the venom and adjuvant used. The time taken to reach a suitable potency is also variable depending upon the adjuvant, frequency of injection and venom type. Monitoring must be undertaken to ensure adequate response is obtained and non- responders removed from production.

Monitoring the immune response

To monitor the effectiveness of the immunisation regime, antibody levels are usually measured using various types of in vitro immunoassays. In most situations, an ELISA (Theakston et al. 1977) will be used to determine response to whole venom or specific toxic fractions of the venom (Pratanaphon et al. 1997). ELISA is preferable because of its high throughput, precision and automation of some steps of the assay. A weakness of the ELISA is that the optical density readings obtained may not necessarily correlate well with venom neutralisation, or only correlate weakly. Other immunological monitoring methods include small scale affinity chromatography (Jones et al. 1999) to measure specific antibody concentration. Neutralisation of toxic components such as PLA2 enzymes has also been used, although assay throughput when assessing individual animals is low (Gutierrez et al. 1988).

In vivo bioassays using rodents provide the Gold Standard measure of anti-lethality monitoring of hyperimmunisation (Standardization 2010). However, rodent assays are expensive and raise significant animal welfare issues (Sells 2003). Some question has also been made of precisely what properties of antivenom a rodent assay actually measures (Kurtovic et al. 2012). In the case of Australian venoms, whose principal toxicity is neurotoxins, the mouse assay provides a reasonable measure.

Harvesting of serum or plasma

Once a suitable level of potency has been achieved, as determined by the monitoring program, blood can then be harvested.

There are a number of different approaches used by the different manufacturers. Whole blood can be collected based on 15% of circulating blood volume. At this level, minimal changes are expected on the animal’s health. Blood can be collected using an anti-coagulant, in which case it is called plasma. In the case of horses, the red blood cells can be separated by gravity and returned to the horse the next day following collection (called manual plasmapharesis). This improves horse health and overall longevity for antivenom production. This is termed plasmapharesis (Ziska 2012). The process can be automated, using a machine that can simultaneously draw blood, separate off the red cells and return these to the animal whilst harvesting the plasma. However, these machines are very expensive (tens of thousands of dollars) and are slow to process large volumes of blood. The animals must also have a specific catheter inserted into the vein to ensure the process works properly.

In animals other than horses, it is generally impractical to perform manual plasmapharesis, and automated machines are too slow and expensive to be practical. The workaround is to only take blood at safe intervals (one to two months minimum), and that only the minimum amount is collected. In smaller species such as sheep and goats, whole blood is often collected into containers without using an anticoagulant. The blood is then allowed to clot. It is then centrifuged to separate into red cells and serum. Serum contains less contaminating clotting cascade proteins, which are consumed during the clotting process. However, working with serum is more difficult and poses a different set of challenges.

Processing and concentration of antibodies

The first step is to separate the red cells from the plasma or serum. Once plasma or serum is obtained, it is filtered to remove any remaining red cells. After filtering, serum can be either stored at 4°C, or frozen at -20°C, until further processing takes place. Storage at -20°C for five years has no detrimental effect on antibody activity.

To reduce the unwanted protein load and decrease the volume required, a concentration step is performed. The concentration step is usually performed because IgG concentrations in plasma and serum are only 10-15 g/L out of a total serum protein concentration of 60-90 g/L depending upon animal species.

Serum contains a large amount of the protein albumin, which is not desirable in the final product due to its likelihood of causing untoward immune responses in the recipient. Concentration of IgG levels in serum can be done by a number of different methods. Two commonly used methods for concentrating IgG in commercial antibody production are: the precipitation out of IgG using high concentration of salts such as ammonium sulphate (Rojas et al. 1994); or, the precipitation of albumin using octanoic acid (Rojas et al. 1994). Both methods result in a degree of contamination of unwanted proteins. The octanoic acid method has advantages in terms of simplicity and higher recovery of IgG activity (Rojas et al. 1994). To achieve a higher degree of purity, chromatographic methods must generally be applied. Perhaps the ultimate method is to perform affinity chromatography, whereby only the specific antibodies of interest are extracted from the serum (Madaras et al. 2005). Affinity

chromatography requires the preparation of columns containing a matrix in which the venom is bound to. Serum is flowed through the column and specific antibodies bind to the venom. The bound antibodies can be eluted from the column by adjusting the pH, and recovered.

The final bulk antivenom product can then be dispensed as a liquid into smaller vials (5-50mL) and stored. A preservative is often added (phenol). The need for a preservative is questionable when the product can be manufactured under completely aspetic conditions. The preservative has been shown to lead to problems with product stability, and increase the likelihood of untoward reactions when administered intravenously (Rojas et al. 1993, Meyer et al. 2007).

Quality control

To ensure that the final product is suitable for neutralising snake venom, it must be tested in a number of ways. Basic physiochemical tests of the product are performed which include protein content, albumin content, pH, sodium chloride content and assessment of other proteins present.

Of key importance is the anti-lethality activity or potency of the antivenom. Potency information is obtained by injecting mice with a standard dose of a multi-lethal venom dose, and incubating with various dilutions of antivenom. Generally six to eight different concentrations of antivenom are used, and typically four to eight mice per dose level are injected. Deaths are recorded at either 24 or 48 hours, depending on the venom type and rodent species used. This quantal data is then analysed using a specific statistical method called Probit analysis, or the Spearman-Karber method.

Results of potency assessment are generally presented as mg of venom neutralised per mL of antivenom. However, when comparing products with different protein concentrations, results may be more appropriately expressed as mg of venom neutralised per gram protein contained in the antivenom.

World antivenom situation

There is currently a worldwide shortage in the supply of antivenom to certain geographical regions of the world. The World Health Organisation recognises this shortage (WHO 2007). The WHO estimates that there are five million envenomings each year, resulting in 125,000 human deaths and three times that number of people who suffer permanent disability from the bite. The incidence of snake bite mortality is particularly high in Africa, Asia, Latin America and New Guinea. In India alone, there may be as many as 50,000 snake bite deaths each year.

The current annual need for antisera for post-exposure rabies prophylaxis, and for the treatment of snake bite and scorpion sting envenomings, amounts to nine million vials of rabies immunoglobulin and ten million vials of antivenoms (WHO 2007). Unfortunately, the present worldwide production capacity is well below these needs (WHO 2007).

The situation is complex and not merely a result of under production. Lack of health care facilities and workers in areas of the world that experience significant snakebite cases confounds the problem. The high price of antivenoms often leads to corruption, theft and counterfeit products.

Objectives

This project was undertaken as a proof-of-concept study. The main objective was to determine if alpaca could be used to make antivenom of similar potency to that of existing commercial products. Since there was no other work on alpaca and antivenom product, this project was breaking new ground.

Antivenom was chosen as the immunoglobulin product to develop because of a number of characteristics. Firstly, antivenom production requires the hyper-immunisation of an animal against toxic substances. In the case of snake venoms, these toxic substances are usually low molecular weight molecules. This aspect provided and insight into how alpaca respond to minute (ug) doses of the substance that antibodies are required to neutralise. Development of monitoring methods is also linked to this. The final antivenom product is relatively easy to test for its effectiveness using standard in vitro and in vivo assays. These assay systems for determining potency are well defined. The production of the final antivenom product would also require the successful application of methods for concentrating immunoglobulins. All of these steps are required for production of antibodies, regardless of the application of the end product.

Thus, the demonstration that antivenom production was possible, and that a potent product could be made in alpaca, would likely test multiple issues related to production of immunoglobulins in alpaca.

A number of specific objectives were defined in the research contract and are detailed below.

Specific objectives as defined in the RIRDC research agreement

1. Literature review of camelid immunology a. Unique aspects of camelid immunoglobulin molecules b. Tolerance and responses to various vaccine adjuvants c. Antibody extraction methods

2. Field work: Immunisation of alpaca a. Evaluation of different immunisation protocols b. Determination of suitable snake venom formulation and vaccine doses c. Assessment of most suitable adjuvant d. Assessment of injection site reactions post-vaccination e. Assessment of impact of immunisations on the animal’s health

3. Assessment of immune response by ELISA a. Development of Enzyme Linked Immunoassay (ELISA) to measure antibody response b. Optimisation of ELISA for alpaca antibody responses to venom

4. Assessment of serum/plasma protein analysis post-immunisation a. Electrophoresis of serum proteins to monitor immune response b. Monitoring of other protein (for example, ceruloplasmin) and IgG measurements

5. Caprylic acid fractionation of plasma a. Adaptation of existing equine protocol for alpaca b. Development of simplified method suitable for up scaling

6. Assessment of fractionated plasma (protein electrophoresis, chemistry) a. Routine chemical analysis of purified product b. Assess stability of final product

7. Assessment of potency of final product a. Assessment of potency via mouse lethality studies (ED50 studies) b. Determine volume of antivenom that neutralises 1mg of venom

Methodology

Animal ethics approval

Approval for experimental work involving animals was obtained from the Wildlife and Small Institutions Animal Ethics Committee, Bureau of Animal Welfare, Victorian Department of Primary Industries (DPI Vic), Mickleham Road, Attwood. The Wildlife and Small Institutions Animal Ethics Committee approved the project in two stages. The first approval obtained (16.11) from DPI Vic was for the hyper-immunisation and blood collection. A second application was approved (06.12) for mouse bioassay testing, to determine the potency of the antivenom. The mouse bioassay testing protocol was modified from WHO (Standardization 2010) recommendations to improve the animal welfare outcomes, but still obtain meaningful results.

Alpaca sourcing and general management

Alpaca were sourced through the Australian Alpaca Association. The initial group of animals consisted of 20 alpaca. Within this group, there was an equal mix of Suri and Huacaya alpaca types. The majority of alpaca were male castrates. Females that were used were generally infertile, or had significant conformational abnormalities, reducing their potential commercial value. Alpaca ranged in age from two years to eight years old. Alpaca originated from herds located in NSW and Victoria.

Alpaca were located in a five acre paddock near Bairnsdale, Victoria. Alpaca grazed pasture and were supplemented with lucerne hay, oats, lupins and alpaca pellets. Alpaca defecate in communal dung piles. These piles were collected from time to time to minimise build up in the paddock.

Alpaca were monitored by faecal egg counts, and treated regularly for internal parasites using injectable ivermectin (Bomectin™).

Alpaca were supplemented in early winter with a single injection of Vitamin A, D & E (Vitamec ADE Injection, AgVantage Pty Ltd, Australia).

Figure 11. Alpaca research herd.

Handling of alpaca

Alpaca are gentle animals by nature, and only minimal facilities were used to handle them. All treatments were performed on the property. Cattle handling facilities were located on the property and these were used for all animal handlings. Alpaca became accustomed to being fed, and were easily enticed into the yards. Alpaca do not ‘flow’ readily like sheep down a long race or chute. They have a tendency to stop and sit down, requiring human physical intervention to move them along.

To administer venom immunisation injections and collect 10mL blood samples, alpaca were manually restrained individually each time. One person held the alpaca around the neck, whilst a second person collected the blood sample and administered the subcutaneous injections. Physical restraint was generally minimal. Alpaca were lightly gripped around the upper neck. In most cases, there was no resistance applied by the animal. Occasionally an alpaca would spit or vocalise. Three people were generally required for each animal treatment. The third person prepared the injections and recorded all data.

Harvesting of serum was performed in the same facilities, using the same personnel. Whole blood was collected into sterile 800mL fluid collection bags.

Figure 12. Manual restraint of alpaca.

Health monitoring post-venom immunisation

To ensure that the animal health implications of immunising alpaca with snake venom were understood, a comprehensive health monitoring program was used for the first three immunisations. It was considered that the first three immunisations would be most likely to elicit adverse signs in the alpaca. To detect abnormalities, blood samples were collected immediately prior to immunisation, and again 24 hours after each of the first three immunisations. Animals were monitored clinically, and had rectal temperatures recorded. Blood samples were submitted to a commercial veterinary pathology laboratory for complete biochemistry and haematology profiles, including fibrinogen. The blood profiles were completed for each animal for the first three immunisations. Data was stratified by treatment group.

Snake venom preparation for immunisation

Venoms were diluted to stock solutions in sterile 0.9 % sodium chloride. Venoms were sterile filtered using a 0.22 µm syringe filter. Once prepared to the appropriate concentration, venoms were stored frozen at -20°C until use. Tubes containing venom were labelled with species name, volume and concentration.

Treatment groups

Alpaca were randomly allocated the one of five treatment groups. Treatment groups comprised of a combination of monovalent and polyvalent groups. The groups were chosen based on current commercial antivenoms used in Australia and Papua New Guinea. All venoms were obtained from a commercial venom supplier, Venom Supplies Pty Ltd, Tanunda, South Australia. Venoms were stored

frozen until used. The venoms of the Tiger Snake and Brown Snake were prepared as a geographically representative pool, as regional differences in venom has been observed (Flight et al. 2006).

Papuan Taipan venom originated from a supplier in Merauke, Bali, and was supplied through Venom Supplies Pty Ltd. Papuan Taipan was chosen over the Australian Coastal Taipan because of its higher toxicity. In general, it is preferable to use the most toxic of the species for immunisation. Pseudechis australis was chosen as the Black Snake venom. This is commonly called the King Brown Snake, but is rightfully a member of the Black Snake family.

Table 17. Treatment groups used for alpaca immunisation.

Antivenom Type N Common Name Species Name

Monovalent Papuan Taipan 4 Papuan Taipan Oxyuranus scutellatus canni

Monovalent Brown Snake 4 Eastern Brown Snake Pseudonaja textilis

Monovalent Tiger Snake 4 Tiger Snake Notechis scutatus

Polyvalent 4 Papuan Taipan Oxyuranus scutellatus canni

Eastern Brown Snake Pseudonaja textilis

Tiger Snake Notechis scutatus

Death Adder Acanthophis antarcticus

King Brown Pseudechis australis Bivalent (Tiger/Brown) 4 Tiger Snake (Notechis scutatus)

Brown Snake (Pseudonaja textilis)

Immunisation program

Alpaca were immunised using a low-dose, multi-site protocol (Chotwiwatthanakun et al. 2001). A combination of Complete Freund’s Adjuvant and Incomplete Freund’s Adjuvant was used. Emulsions were formed with the adjuvant and venom solution in a 1:1 ratio. The stability of the oily emulsion was tested by ensuring that it did not disperse when a droplet was placed onto the surface of distilled water. Injections were given at monthly intervals based on protocols used in sheep (Landon and Smith 2003). All injections were given subcutaneously on each side of the neck, lateral thorax and lumbar region. The injections were sited to be anatomically close to local draining lymph nodes, to maximise exposure to the immune system (Agba et al. 1996, Agba et al. 1996). The aim of the low-dose multi- site program was to expose as many lymph nodes as possible to the venom, and therefore create the highest immune response.

ELISA for venom antibody measurement

An Enzyme Linked Immunosorbent Assay (ELISA) was developed to measure circulating antibodies to the various snake venoms. The ELISA was based on well-established venom-antibody ELISAs

previously published in peer-review journals (Theakston et al. 1977, Theakston and Reid 1979, O'Leary and Isbister 2009). A brief description of the methodology is provided here.

For each assay, 96-well polystyrene microtitre plates were coated with 100 µL of crude venom at 1 µg/mL. The plates were washed three to five times after 24 hours, to remove any unbound venom. A blocking buffer was added to block any unbound sites on the plates. The blocking buffer was washed out, and diluted test alpaca (1:100) serum was added to each well. Plates were again washed after one hour. A commercial goat anti-llama Horse Radish Peroxidase (AbCam Pty Ltd, USA) conjugated secondary antibody conjugate was added. TMB was used as the chromogen. The peroxide substrate was added and colour change quantified using an automated ELISA plate reader. Data is presented as Optical Density (OD) readings. The OD of the blank wells was subtracted from the reading. Control wells containing serum from alpaca that had not been exposed to any venom were used on every plate as negative controls.

Figure 13. The author and assistant Allan Quirke harvesting whole blood from an alpaca under field conditions.

Packed cell volume

Alpaca packed cell volume measurements (PCV) were made, using a microcapillary tubes loaded with whole blood collected into EDTA vacuum tubes. Tubes were centrifuged for three minutes, and read using a manual calliper unit.

Venom toxicity determination

To determine the activity of the snake venom in the mouse bioassay, a Median Lethal Dose study was undertaken for each venom type. The assay used eight levels of venom concentrations spaced at intervals of 1.5X dilutions, with five mice used per level. Data was analysed using Spearman-Karber, and presented as Median Lethal Dose (LD50) with a 95% CI.

Neutralisation of lethality in mice

All mouse bioassay work was approved by DPI Victoria (License No. WSIAEC 06.12). To meet animal ethics requirements a modified eight hour mouse protection assay was developed. Mice were monitored continuously from the time of receiving venom. The condition of the mouse was graded from zero to four, and mice were euthanised once they were deemed to have entered Stage Four. All mice were euthanised after eight hours. If a mouse was euthanised or had entered Stage Four, it was classified as ‘dead’ for the purposes of statistical analysis. This approach was considered valid, as mice readily succumb to the neurotoxins in Australian elapid snakes.

The Gold Standard test for determining potency of antivenoms is the mouse bioassay. The principle of the test is to determine the dose of antivenom that prevents death in 50% of mice that are exposed to a constant multi-lethal dose of venom (Standardization 2010). A range of concentrations of antivenom were used to ensure groups that represented 100% death and 100% survival result. Swiss mice of a defined weight range were used, typically 16-20g. The quantal survival data was then analysed, using either Probit methods or Spearman-Karber method (Solano et al. 2010). Results were expressed as ED50 (Effective Dose 50%), and the units were mg of venom neutralised per mL of antivenom and mg or protein.

Serum protein electrophoresis

Samples of serum, both unprocessed and processed, were submitted to a commercial veterinary pathology (Gribbles Veterinary Patholgy, Melbourne) laboratory for serum protein electrophoresis.

Processing of serum

Centrifugation and filtration

After clotting was complete, freshly collected serum was centrifuged at 4,200 rpm for 15 minutes, and serum extracted. The serum was then filtered through a 3µm depth filter and then 0.22 µm sterile filter, to remove any residual red cells and cellular debris. Serum was stored in sterile bottles and either frozen at -20°C, or stored in the fridge at 4°C until used.

Figure 14. Freshly collected bag of alpaca blood ready for clotting and serum separation.

Figure 15. Alpaca serum after centrifugation and prior to filtration (LEFT), and immediately after filtration through 0.22um depth filter (RIGHT). The serum on the left still contains a small percentage of red blood cells which are removed by the filter.

Caprylic acid fractionation

The method described by (Rojas et al. 1994) was used as the basis for experimentation with caprylic acid fractionation of serum. Raw serum was acidified to pH 5.7, using glacial acetic acid. Caprylic acid was added at 6% concentration, and the mixture stirred vigorously for one hour. However, this

protocol failed to yield a suitable product. There was incomplete precipitation of the non-IgG protein, and this quickly clogged any filter.

An alternative method was tried based on the original published work on caprylic acid (Steinbuch and Audran 1969). Briefly, serum was diluted two volumes in a weakly acid sodium acetate buffer with pH adjusted to 4.5 using glacial acetic acid. Caprylic acid was added at 6% of the serum volume. The mixture was stirred for two hours and then centrifuged. However, the product was considered unacceptable as it was difficult to filter.

Due to difficulties with filtration, the caprylic acid method was abandoned in favour of the more reliable ammonium sulphate precipitation (below).

Ammonium sulphate fractionation

Fractionation of serum was successfully performed using a modified ammonium sulphate precipitation protocol, as previously described (Wingfield 1998). Briefly, solid ammonium sulphate was added to distilled water to reach 100% saturation. Serum was diluted 1:1 with the 100% saturated solution such that a final saturation of 50% was obtained. The mixture was stirred for 30 minutes and allowed to cool to 4°C overnight. The mixture was then centrifuged, the supernatant discarded, and pellet resuspended in PBS. The solution was then dialysed against PBS to remove any remaining ammonium sulphate. It was then filtered using a 0.22 µm filter, and stored in sterile vials at 4°C.

Figure 16. Precipitation of immunoglobulin proteins using 50% ammonium sulphate.

Study trip to Instituto Clodomiro Picado, San Jose, Costa Rica.

During March 2011, a visit was made to the Instituto Clodomiro Picado in Costa Rica. The visit was arranged with Prof. Jose Maria Gutierrez from the Department of Microbiology, University of Costa Rica. Time was spent in the commercial antivenom plant, research facility and serpentarium.

Results

Summary of project key achievements

1. Successful production of antivenom, that prevented laboratory mice from death against a challenge with a multi-lethal dose of venom.

2. Demonstration that alpaca can be used for antivenom production without impacting on the health or welfare the animals.

3. Development of an ELISA to measure antibodies in alpaca serum to five elapid venom types.

4. Development of practical techniques and consumables suitable for repeated large volume blood collection in alpaca.

5. Adaptation of equine plasma fractionation protocols for use with alpaca serum, including ammonium sulphate, sodium sulphate and caprylic acid.

6. Animal ethics approval for using venoms for immunisation and repeated harvesting of alpaca serum.

7. Animal ethics approval for antivenom potency testing and toxicity testing in mice.

8. Demonstration that Freund’s Complete and Freund’s Incomplete adjuvants are tolerated by alpaca, and generate strong immune responses.

9. Networking with Australian and International researchers in the field of Toxinology.

10. Establishment of a platform of techniques to undertake further research in antibody production, and specifically antibody production and harvesting for snake antivenoms.

Effects on alpaca health

Of great initial concern was how the alpaca would respond to immunisation with the various snake venoms. To monitor the health of the animals during the immunisation process, serial blood haematology and biochemistry profiles were performed at Time 0 and Time 24 hours after every immunisation, for the first three immunisations. See Figures 17 to 28 for results.

Serum biochemical changes

In general, there were relatively minimal changes in serum biochemistry and haematology parameters in the alpaca during the first three months of venom immunisation. There were no significant changes to liver enzymes or renal function. Red cell counts remained unchanged throughout.

As expected, there were changes in muscle enzymes and markers of inflammation such as fibrinogen and total white cell counts.

Serum muscle enzymes

As expected, the venoms that contained significant muscle damaging toxins resulted in an increase in the muscle enzyme Creatinine Phosphokinase (CK). Serum levels of CK rose markedly at 24 hours after the first immunisation with Tiger, Tiger and Brown, Polyvalent, and only slightly with Taipan. No increase in muscle enzymes at 24 hours was noted from Brown Snake Venom on its own. Interestingly, as the dose of venom increased at each immunisation, there was a corresponding decrease in the 24 hours post immunisation serum CK level. This probably reflected the animals developing neutralising antibodies which quickly absorbed any circulating toxin, and prevented muscle damage. Some of the CK elevation could also be due to mechanical trauma from intramuscular injections.

Serum levels of the muscle enzyme Alanine aminotransferase (AST) were also monitored. AST also originates in muscle and can be used to monitor muscle damage (Foster et al. 2009, Tornquist 2009). AST is slower to rise in serum following muscle damage, and lower levels are measured compared to CK.

Serum inflammation markers

Blood levels of fibrinogen can be used as a marker of inflammation in various species, including camelids (Foster et al. 2009, Tornquist 2009). Serum fibrinogen levels rose 24 hours after each dose of venom, but had returned to pre-treatment values at the time of the next immunisation.

White blood cell counts also rise moderately at 24 hours post-venom injection. This was not unexpected as the oily Freund’s adjuvant is designed to attract white cells and stimulate an immune response to the venom.

PCV response to whole blood collection

The response of the alpaca to collection of 800mL of whole blood was recorded by measuring the PCV at 21 days after blood collection. There was a noticeable decrease in PCV (see Table 18) following blood collection. There was also an increase in the variation between animals. Some animals appeared to be unchanged in their PCV whilst others had decreased to below what would be considered a healthy level.

Table 18. Mean packed cell volume (PCV) of alpaca prior to starting venom immunisation and at 21 days after collection of 800mL of whole blood.

Pre-Treatment (n=20) 21d Post-Bleeding (n=20)

Packed Cell Volume 0.31 (range 0.28 to 0.34) 0.26 (range 0.19 to 0.30)

Std. Deviation 0.02 0.03

Figure 17. Effect of Tiger Snake Venom on serum the muscle enzyme Creatinine Kinase (CK) at time 0 and 24 hrs after immunisation for the first three months. Results presented as mean for the Tiger Snake venom treatment group.

Figure 18. Serum CK levels in alpaca receiving the polyvalent venom mixture.

Figure 19. Serum CK levels in alpaca receiving Brown Snake venom only.

Figure 20. Serum CK levels in alpaca receiving both Tiger Snake and Brown Snake venom.

Figure 21. Serum CK levels in alpaca receiving Papuan Taipan venom.

Figure 22. Serum AST levels in alpaca receiving Tiger Snake venom.

Figure 23. Serum AST levels in alpaca receiving polyvalent venom.

Figure 24. Serum AST levels in alpaca receiving Papuan Taipan venom.

Figure 25. Serum AST levels in alpaca receiving Brown Snake venom.

Figure 26. Serum AST levels in alpaca receiving Tiger Snake and Brown Snake venom.

Figure 27. Serum fibrinogen levels in alpaca at Time 0 and Time 24 hrs after the first immunisation.

Figure 28. Total white blood cell counts in alpaca at Time 0 and Time 24 hours after the first venom immunisation.

Antibody responses to venom - ELISA results

Analysis of serum antibody levels for each venom type showed a similar pattern across all treatment groups. Alpaca that received five different types of venom in the polyvalent treatment group responded well to all five venom types, with rapid and high levels of antibodies. Antibody levels tended to increase markedly after the initial immunisations, and trend towards plateauing. See Figures 29 to 40 for charts of these responses.

It is not possible to compare the magnitude of antibody responses, because the different assay batches lead to slightly different optical density readings. However within the polyvalent group, all samples were run under the same conditions, using the same antibody dilutions, antigen concentrations and conjugate concentration. The polyvalent group tended to result in more uniform antibody concentrations after 126 days compared to the monovalent.

When the data was analysed out to 300 days for the Tiger Snake antigen in the polyvalent treatment group (see Fig 39, 40), only small increments could be seen after around 180 days of immunisation. If individual animals are observed, it is possible to see differences between animals in the maximum optical density readings (valid as same conditions). This may reflect different neutralising potencies of the serum.

Figure 29. Antibody responses of alpaca to immunisation with Brown Snake venom (P. textilis) in the polyvalent treatment group.

Figure 30. Antibody responses to King Brown (Black) snake (Pseudechis australis) venom in the polyvalent treatment group.

Figure 31. Antibody responses of alpaca to immunisation with Papuan Taipan venom (O. scutellatus canni) in the polyvalent treatment group.

Figure 32. Antibody responses of alpaca to immunisation with Common Death Adder venom (Acanthophis antarcticus) in the polyvalent treatment group.

Figure 33. Antibody responses of alpaca to immunisation with Tiger Snake venom (Notechis scutatus) in the polyvalent treatment group.

Figure 34. Antibody responses of alpaca to immunisation with Papuan Taipan venom (O. scutellatus canni) in the monovalent Taipan treatment group.

Figure 35. Antibody responses of alpaca to immunisation with Brown Snake venom (P. textilis) in the monovalent treatment group.

Figure 36. Antibody responses of alpaca to immunisation with Tiger Snake venom (Notechis scutatus) in the monovalent treatment group.

Figure 37. Antibody responses of alpaca to immunisation with Brown Snake venom (P. textilis) in the bivalent Tiger/Brown treatment group.

Figure 38. Antibody responses of alpaca to immunisation with Tiger Snake venom (N. scutatus) in the bivalent Tiger/Brown treatment group.

Tiger Snake Venom Antibodies (Polyvalent Group)

1.4

1.2

1.0

0.8

0.6

ELISA OD at 405nm at OD ELISA

0.4

0.2

0.0 0 50 100 150 200 250 300 350 Days

Day vs Alpaca 5 Day vs Alpaca 6 Day vs Alpaca 7 Day vs Alpaca 9 Day vs Mean

Figure 39. Antibodies to Tiger Snake Venom in animals receiving the polyvalent venom mixture. Data shown out to 300 days.

Average of Tiger Snake Venom antibody levels in Polyvalent group.

1.2

1.0

0.8

0.6

ELISA OD at 405nm at OD ELISA

0.4

0.2 0 50 100 150 200 250 300 350 Days

Day vs Mean

Figure 40. Average antibody levels to Tiger Snake Venom in animals receiving the polyvalent venom mixture. Data shown out to 300 days.

Serum processing

Caprylic acid

The processing of alpaca serum to remove non-IgG proteins was not possible using undiluted alpaca serum. A range of concentrations of caprylic acid from 1% to 30% were tried under different pH conditions. Undiluted serum tended to form a milky substance and clear separation of the non-IgG proteins failed to occur.

However, when alpaca serum was diluted with a 0.1M sodium acetate buffer and the volume expanded to two to four times the original serum volume, caprylic acid was able to precipitate out non-IgG protein.

Serum protein electrophoresis on agarose gels revealed the pattern of protein precipitation that occurred with caprylic acid (Figures 43-45). When alpaca serum was diluted within a 0.1M sodium acetate buffer, it readily separated. However, electrophoresis revealed that albumin was still present in significant amounts and accounted for 20% of the total protein present (Figure 44). Based on the high content of albumin and other non-IgG protein, further processing would be required to make a refined product.

Production of a highly purified IgG fraction was attempted. Further purification using a Protein G column (see Figure 45) completely removed the remaining non-IgG proteins and provided a very pure alpaca IgG protein.

Figure 41. Appearance of alpaca serum after processing with caprylic acid with serum diluted in a 0.1M sodium acetate buffer. Note the distinct separation of the non-IgG pellet and clear supernatant containing the IgG.

Figure 42. Undiluted alpaca serum mixed with caprylic acid at 3% (left) and 6% (right) caprylic acid concentration. Note the incomplete separation of the IgG and non-IgG proteins compared to the diluted serum.

Figure 43. Serum protein electrophoresis pattern of alpaca serum prior to processing serum (scanned by Frank Madaras, Venom Science Pty Ltd).

Figure 44. Serum protein electrophoresis pattern of caprylic acid fractionated alpaca serum (scanned by Frank Madaras, Venom Science Pty Ltd). Note the reduction in the albumin peak.

Figure 45. Further purification of the caprylic acid precipitated serum with Protein G. Note the single large immunoglobulin peak on the serum protein electrophoresis gel scan (scanned by Frank Madaras, Venom Science Pty Ltd).

Ammonium sulphate precipitation

Precipitation of serum IgG using ammonium sulphate was successful. It was found that dilution of the raw serum with distilled water in a ratio of one volume of serum and two volumes of water worked well. An ammonium sulphate saturation of around 40% appeared to be close to ideal, without contaminating the pellet with non-IgG proteins. Extensive dialysis was required to remove the remaining ammonium sulphate from the pellet. Electrophoresis of the purified product showed that it was 96% purified immunoglobulin protein (Figure 46).

Sodium sulphate precipitation

Sodium sulphate precipitation of IgG protein in alpaca serum appeared to work well and was successful with either use of solid powder, or a saturated solution. Temperature is critical with sodium sulphate, as the solubility is highly temperature dependent, which may limit its commercial usefulness. All processing was carried out at 22° to 25°C.

Heat treatment of serum

It was discovered that heat treatment of the raw alpaca serum at 56°C for 60 minutes improved the clarity of the final product, and did not affect potency. Without heat treatment, microscopic particulate matter tended to accumulate in the final product when stored at 4°C for a few weeks.

Figure 47. Effect of heat treatment of raw serum to 56°C prior to processing. Vial on the left has been heat treated while that on the right received no heat treatment. Note the ‘murky’ appearance to the vial on the right which has higher turbidity.

Figure 48. Heat treatment of raw serum. Bottle on left has been heated to 56°C for 60 minutes whilst bottle on right has not been heated. Note the colour change.

Median Lethal Dose determination of venoms

The lethality in mice of each of the venoms used for immunisation was tested in a Median Lethal Dose (LD50) assay.

Brown snake

Venom from the Eastern Brown Snake (Pseudonaja textilis) was used to determine its lethality in the previously described mouse bioassay. The venom used had a Median Lethal Dose of 0.025 mg/kg in Swiss mice weighing 14-18g.

Tiger snake

Venom from the Mainland Tiger Snake (Notechis scutatus) was used to determine its lethality. The venom used had a Median Lethal Dose of 0.144 mg/kg in Swiss mice weighing 14-18g.

Papuan Taipan

Venom from the Papuan Taipan (Oxyuranus scutellatus canni) was used to determine its lethality. The venom was supplied by Venom Supplies Pty Ltd, Tanunda, South Australia. The venom was collected from snakes located in Merauke, Indonesia. The venom used had a Median Lethal Dose of 0.039 mg/kg in Swiss mice weighing 14-18g.

Neutralisation of lethality and potency determination

The potency of the alpaca polyvalent antivenom product was determined using a Median Effective Dose (ED50) assay in mice. As described in the Methods section of this report, mice were given a fixed dose of venom, calculated as five lethal doses, and mixed with varying dilutions of antivenom which was pre-incubated with the venom. These mixtures were then injected i.p. into Swiss mice of weight range 18-22 g. Data was analysed using Spearman-Karber analysis.

The final formulation of polyvalent alpaca antivenom used for potency testing was prepared using heat treatment of the serum to 56°C for 1 hour, precipitation using ammonium sulphate, dialysis against PBS and phenol (0.22%). The final product was concentrated to 89.7 g/L using 20mL VivaSpin™ (Viva Products, USA) columns. The final product was sterile filtered using 0.22 µm syringe filters and placed in sterile glass vials.

Total protein was measured by Biuret method, and albumin using the Bromocresol Green Method. Serum protein electrophoresis was performed on agarose gel and the results are shown in Figure 46. The product was placed into glass vials and was stable at 4°C, with no precipitate in the vials.

To compare the alpaca polyvalent antivenom under the same assay conditions, a reference antivenom was used for each assay. The reference antivenom used was a Commonwealth Serum Laboratories monovalent antivenom. Due to difficulties in obtaining sufficient numbers of mice, only three venom types were tested. The four animals in the polyvalent treatment group contributed serum in equal amounts to the final pooled polyvalent product.

A summary of the results in shown in Table 19.

Perhaps the best measure to use when comparing antivenom products is to assess the venom neutralising capacity per gram of total protein. Alpaca polyvalent antivenom is equally as potent for Brown snake 19mg of venom neutralised per gram of total protein. However, for Tiger Snake and Papuan Taipan, it is less potent.

Table 19. Final results of Median Lethal Dose (LD50) and Media Effective Dose (ED50) for the experimental alpaca polyvalent antivenom compared to commercial monovalent antivenom products from the Commonwealth Serum Laboratory.

CSL Monovalent Antivenom1 Alpaca Polyvalent Antivenom2

Venom LD50 Snake Species (mg/kg) Potency Potency Potency Potency (mg Total Protein Total Protein ( 95% CI) (mg/mL) (mg venom (mg/mL) venom neutralised (g/L) neutralised /g (g/L) (95% C.I.) (95% C.I.) /g total protein) total protein)

Brown Snake 0.025 3.05 1.649 160.2 19.0 89.7 18.4 (P. textilis) (0.018, 0.035) (2.346, 3.965) (1.576, 2.451)

Tiger Snake 0.144 5.72 1.042 166.2 34.4 89.7 11.6 (N. scutatus) (0.059, 0.349) (4.77, 6.85) (0.830, 1.300)

Papuan Taipan 0.039 3.812 0.817 159.2 23.9 89.7 9.1 (O. scutellatus canni) (0.0.27, 0.058) (3.208, 4.528) (0.668, 0.999)

1 CSL Brown Snake Antivenom. Batch #11101. Expiry 09/12. 8.49mL. 1,000 Units. CSL Tiger Snake Antivenom. Batch #10701. Expiry 02/11. 8.11mL. 3,000 Units. CSL Taipan Antivenom. Batch # 0548-06401. Expiry 10/12. 34.4mL. 12,000 Units.

2 Batch # ALP-POLY-231012.

Discussion of results

This work has proved the concept that alpaca can be used successfully to make a polyvalent snake antivenom product in alpaca that is capable of protecting mice from the lethal effects of snake venom. The potency of the alpaca product was lower than commercial equine monovalent antivenom products.

Animal health

There appeared to be no detrimental effect of snake venom on the alpaca used in this study. Changes in muscle enzymes were as expected based on known effects of the venoms. These changes declined over time, and appeared to show the animals immune system had already produced neutralising antibodies. Of concern was the development of lumps from the oil-based adjuvant injections. It appeared that some animals were more vulnerable to these lumps than others. Of the twenty animals in the group, two animals were classed as having very large lumps that might limit their future use for antivenom production. These lumps cause problems with shearing. In sheep, the immunisations are usually given inside the legs to make shearing easier. Accidental trauma to the lumps leads to chronic discharging wounds.

The handling of the animals is important. Alpaca are flighty and stress relatively easily when handled. However, they are relatively easy to handle. We did experience three separate collapsing episodes during handling. Each of these was attributed to positioning of the head backwards over the shoulders. This seems to result in a balance mechanism disturbance and if held in this position for a few minutes the animals collapsed, and showed signs of inner ear disturbance with nystagmus. In each case, the animals recovered fully within a few minutes. We performed blood tests (biochemistry and haematology and faecal egg counts) on the affected animals each time but no abnormalities were detected.

Alpaca serum antibody levels

This work has shown that alpaca generate a strong immune response to Australian elapid snake venoms. The immune response is readily detected in the ELISA assay system developed specifically for alpaca. Alpaca generate a rapid initial response to venom, which plateaus after 250-300 days of immunisation. The true plateau is not known as animals were only followed for 300 days. Based on this work, it is suspected that further incremental increases in antibody levels would continue to occur for perhaps another six to 18 months beyond where they were monitored for in this project. This would increase the potency of the antiserum.

Assessment of antibody responses to specific toxin fragments (for example, Notexin, Acanthoxin and so on) would be an improvement of the monitoring procedure. It is these low MW toxins that are the most difficult to generate antibodies against, and a method of monitoring process is desirable. Whole venom was used as the microplate coating antigen in this project, but has limitations. Venom is a complex soup of proteins, some of which are easy for the immune system to recognise and make antibodies against, and some not so easy. Specific toxin fragments can be purchased from a commercial venom supplier (Venom Supplies Pty Ltd, Tanunda, SA). The monitoring of the immune response to specific toxin fragments has been useful for preparing antivenom to the Thai Cobra (Pratanaphon et al. 1997). Such an approach for the monitoring the alpaca has merit.

Serum processing

This work has shown that alpaca serum can be processed into a stable final product, with no visible protein precipitate, when stored at 4°C. Product stability is a major issue with serum products. Based on the work performed in this study, a combination of heat treatment of the serum to 56°C for 60

minutes, with subsequent processing using ammonium sulphate and dialysis, will result in a stable final product. Further clarity of the product could be achieved using tangential flow filtration (TFF) and lipoprotein removal using filtration.

Ammonium sulphate precipitation would appear to be the most reliable method. However, caprylic acid and sodium sulphate offer advantages as well. In terms of product safey, caprylic acid processed serum has been shown to be superior to ammonium sulphate when horse serum was used (Rojas et al. 1999, Leon et al. 2005). Caprylic acid appears to more effectively remove unwanted proteins that could trigger immune responses. It also offers advantages in terms of speed of processing and antiviral activity of the caprylic acid. Sodium sulphate has been used for decades in serum processing and is reliable, and results in a high purity final product (Grasset and Christensen 1947). The main problem with sodium sulphate is that processing must be conducted at 22-25°C in order for the salt to dissolve adequately. This may increase bacterial load of the serum, which may not be acceptable under commercial conditions.

The use of caprylic acid to precipitate non-IgG proteins was only successful when the serum was diluted in a weak acetate buffer. The problem with this is that a very large volume of fluid must be processed and then subsequently concentrated. The method has merit, but the practical issues with dilution may not be suitable for commercial scale production methods.

The final product produced using ammonium sulphate still contained a slight cloudiness compared to the commercial equine antivenoms produced by CSL. This may be due to intrinsic factors within the alpaca serum, or to other as yet unknown processing issues. A better understanding of this could only be obtained by processing the alpaca serum under commercial conditions. The product appeared stable when stored at 4°C but was not tested for extended periods (>12mths).

Immunisation protocol

The immunisation regime used in this project was effective for generating protective antibodies in alpaca serum. The protocol we used was based on what is often used in sheep (Landon and Smith 2003). Repeated injections of small amounts of the antigen over time lead to a better immune response than large amounts. Venom is expensive on a per mg basis, and anything to minimise the doses used would be of benefit in a commercial program.

Injection site reactions to the oil based adjuvants we used could be problematic over many years. Freunds Complete and Incomplete adjuvants are considered to be the best adjuvants for stimulating immunity. Our work here has shown that to be the case, with 100% of animals responding to venom injections. Freund’s inevitably leads to some local swelling, and possible abscess formation, at the injection site. This can be minimised, but not eliminated, by using small injection volumes (<0.1mL), but this means many sites must be used. Injecting such small volumes is difficult and time consuming. Sheep are typically injected in either four or six sites with Freunds (Landon and Smith 2003), and this appears to produce an adequate immune response. There is no doubt that low-dose multi-site immunisation regimes do work (Sriprapat et al. 2003). Newer adjuvants are promising in terms of safety and efficacy. Of interest are nanoparticles which have been used for Cobra venom immunisation (Waghmare et al. 2009), and the response obtained was greater than the traditional Freund’s adjuvants.

From a commercial perspective, a cyclical production scheme has great benefit for flock management. For alpaca, this may consist of immunising at two monthly intervals and blood collection 14 days after each booster immunisation. A rest period of six weeks between immunisations would allow the immune response to be enhanced, and allow the animals to produce more RBC.

To improve immune responses to low MW venom toxins, it may be possible to fractionate the venom and give booster immunisations with this toxin fraction. For example, Tiger Snake venom contains the toxin Notexin, a 13kD toxic enzyme. With some filtration of the venom, this toxin could be recovered

in quantities sufficient to be used for immunisation. Care must be taken with using specific toxins, as previous work in other snake species has shown this to not necessarily lead to protective immunity.

One animal that had received Tiger and Brown Snake venom developed neck weakness that persisted for ten days after its first immunisation. Alpaca have a very long neck, and the neck muscles are required to raised and lower their head. The affected animal was found five days after immunisation to be unable to raise its head. With supportive care the animal made a full recovery. It was presumed that this animal had developed pre-synaptic neurotoxicity, which is commonly described with Tiger Snake venom. This was the only animal that showed signs of toxicity from snake venom. Localised damage to nerves has been described in experimental animals injected close to nerves in rats (Harris and Maltin 1981).

Sheep can be used for five to eight years for serum production. Each sheep is capable of producing four litres of serum per year (Landon and Smith 2003). In contrast, horses can be used for 20 years or more, if they are managed correctly. Sheep seem to suffer from renal amyloidosis if immunised with complex antigens, as found in the Crotalus (rattlesnake) family. A gradual infiltration of the kidney with amyloid protein leads to progressive renal failure (Mensua et al. 2003). Amyloid protein arises from aberrant immune stimulation. Joint inflammation is also possible. Horses do not seem to suffer from this problem, and the problem also seems related to Freund’s adjuvant.

Practicalities of blood collection

This project has shown that a large volume of alpaca blood can be harvested under relatively basic conditions, using commercially available veterinary consumables. The method we used for restraining alpaca for whole blood collection, although satisfactory for this experiment, would not be suitable for commercial production. Alpaca are flighty animals and adequate restraint is essential for both alpaca and operator safety. Some alpaca tend to want to jump up vertically when held around the neck. When a large bore needle is being inserted into the jugular, this is dangerous. Tracheal and carotid lacerations could occur, along with injury to the human operator. Using our restraint method, with a team of three people, it would be possible to bleed around 60-70 alpaca per eight hours. This falls short of what can be achieved with sheep (Landon and Smith 2003), where a team of three can comfortably manage 100 sheep. With some improvement to the alpaca handling, and using a head restraint device, it may be possible to bleed 100 alpaca per day.

The collection of 800 mL of whole blood would appear to have some impact on the physiology of the alpaca. Our data showed that by three weeks post collection, the PCV still had not returned to normal. Some animals also had PCVs that would be considered anaemic. In a commercial setting, sheep can be collected every 28 days for five to eight years without affecting their PCV. The implications for this with alpaca are that whole blood collection would perhaps be better performed at two month intervals, to minimise the impact on the animal. Collecting whole blood each month is not commonly done with horses, as it affects their health negatively. Collecting at two monthly intervals is preferred unless plasmapharesis (return of RBC) is performed. Plasmapharesis would be slow in alpaca, and not commercially viable for larger numbers of animals. However, it is very good for the health of the animal, and allows for at least two-fold recovery of serum from the animal. An intravenous catheter must be put in place, and the animal restrained for 30-minutes for each collection.

The close anatomical location of the jugular vein and carotid artery is a problem with alpaca. Collection of blood from the carotid artery is easy to do by accident. Accidental puncture of the carotid causes persistent bleeding from the needle injection site. When using large bore cannula (10G), this bleeding can take five minutes or more to stop flowing, and slows down the bleeding process. As our experience increased, accidental carotid puncture became less likely such that it could be avoided almost entirely. Cold weather conditions leading to venoconstriction seemed to make it more likely that the cannula entered the artery. We had greater success in hitting the jugular by

inserting the cannula around 7-8cm above the lateral process of the fifth cervical vertebrae, which is where blood samples would normally be collected from.

We preferred to only collect blood from the right side of the neck. The reason for this is that the oesophagus runs down the left side, and we wished to avoid its accidental puncture. However this leads to repetitive puncture of the skin on the right side. Despite this, there were only minimal signs that the vein had been repeatedly punctured. Presumably healing occurs quite quickly after venepuncture.

Mouse testing process

The project leader was successful in obtaining the necessary animal ethics permits to conduct mouse testing studies on the antivenom. The testing of the antivenom final product in mice is an essential requirement. We have successfully obtained the necessary permits to perform this testing. We have also developed protocols and methods for performing these bioassays. The eight hour time limit imposed by the animal ethics committee is probably not long enough to monitor the mice. This eight hour cut-off would likely underestimate the final result. Some mice can recover from our defined Stage 2 of toxicity. Further work may be required to allow an extension of this mouse test to 24 or 48 hours. CSL use a seven day test period when testing potency using Guinea Pigs. The mouse potency test we used required approximately 40 mice per antivenom. Eight levels were used, which was necessary because we had no data to show what dose range we should be limited to. Using less than four mice per level will result in insufficient confidence in the result. Using more than 8 mice per level is probably excessive, but would allow for more precision and a narrower 95% confidence interval in the final result. A more flexible animal ethics license using death-as-endpoint would be far better.

The Median Lethal Doses obtained for the venoms tested were in close agreement with data published by CSL in the late 1970s (Broad et al. 1979). However, the potency of the Papuan Taipan venom (0.039 mg/kg) was slightly lower than that obtained by Sutherland (0.055 mg/kg) (Sutherland and Tibballs 2001). Papuan Taipan venom tested in Costa Rica was 20 times more toxic than the venom used in the alpaca (Vargas et al. 2011). This is difficult to explain. The venom used in the alpaca originated from Merauke in Indonesia, yet the venom used in the Costa Rican paper was from PNG. If the PNG venom is much more toxic then it would be beneficial to use it for immunising the alpaca.

Potency

The potency of antivenom is usually expressed as the amount of venom that can be neutralised per mL of antivenom. This can also be expressed as mg of venom neutralised per gram of total protein in the antivenom. The results of the antivenom comparisons are shown in Table 19.

Brown Snake venom was neutralised by the alpaca polyvalent antivenom with equal potency to the CSL monovalent Brown Snake Antivenom product. This finding is very encouraging. Brown Snake venom is perhaps the most toxic venom of any snake; our Median Lethal Dose study found it to be the most toxic of the venoms studied. The typical bite from a Brown Snake contains, on average, 10mg of venom. To neutralise a 10mg venom dose would require only 6.0 mL of the alpaca polyvalent product formulated to 8.9g/L. Brown Snake bites are the most common type of envenomation in humans in Australia, and possibly animals as well. With further purification using affinity chromatography, the Brown Snake antivenom could be concentrated to a therapeutic volume of less than 1 mL per treatment vial. Enzymatic treatment of the whole IgG antivenom to yield F(ab)2 fragments or even Fab fragments could result in a product that may be given intra-muscular, with low risk of allergic reaction.

Tiger Snake venom was less efficiently neutralised by the alpaca polyvalent antivenom than the CSL monovalent antivenom. Tiger snake antivenom was released commercially by CSL in 1930, and was the first antivenom product available in Australia. An average Tiger Snake bite injects approximately

30mg of venom, and to neutralise this would require 28.8mL of the alpaca polyvalent antivenom at 8.9g/L total protein, compared to around 10mL of 160g/L total protein for the CSL monovalent product. This would still be an acceptable volume to administer to a dog or cat, and could be concentrated further by ultrafiltration methods.

Antivenom produced against the Papuan Taipan in the polyvalent alpaca product was effective at neutralising the venom. The potency of the alpaca polyvalent product was lower than that of the CSL monovalent product. Taipan produce large quantities of venom, and a typical bite may release 120mg of venom. To neutralise this with the alpaca polyvalent product would require 146.9mL of antivenom. At this potency, this would be a considerable protein load to administer to a human or animal, containing 1.3g of total protein per treatment. However, with improved protein concentration of the antivenom this volume could easily be reduced to less than 100mL. To concentrate it even further would require the use of affinity chromatography methods to extract out the specific antibodies. This process could improve the concentration by a factor of five to 25 times that of the original serum.

Taipan venom and Tiger snake venom share some similarity in terms of toxins. The lower potency antivenom produced against these snakes is likely a result of the low molecular weight toxins being less immunogenic in the alpaca. Tiger Snake antivenom was first produced in a horse in the early 1900s. The horse was immunised with increasing doses of venom, and it was not until 3½ years later that a satisfactory titre was obtained. It seems that animals need to be exposed to the venom for very extended time periods to develop high levels of specific antibodies.

We need to keep in mind that the alpaca product tested was a polyvalent product, and not a monovalent product. In general, polyvalent products are considered to have lower potency, although this is a contentious issue. The total protein concentration can be varied. CSL typically use protein concentrations of 160g/L. However, other manufacturers limit this to 50g/L.

The potency of the alpaca antivenom can be improved through a number of methods. The polyvalent alpaca antivenom product was derived from only four animals. If a larger group of animals (say 100) were given a primary immunisation course, evaluated for potency, and 50 animals then selected, potentially a higher potency product could be obtained. The ELISA would appear to be the best tool for screening animals on their antibody response. It is possible to create a standard curve based on high responding alpaca serum, and use dilutions of this to determine a cut-off point. Animals could be defined as high responders, medium and low. A pool of serum from high ELISA responders could then be tested in a mouse bioassay, and true potency determined. The downside to this approach is that a primary immunisation phase may take six months to complete.

Another method to improve potency would involve processing to remove more of the unwanted protein in the final product. Sequential processing, with both caprylic acid followed by ammonium sulphate, has been shown to yield a superior product in sheep used for tetanus antibody production (Redwan el et al. 2005). Increased in yield by a further 5-10% could be achieved.

Increasing the time period between starting an immunisation program, and first serum harvest, would increase potency. Tiger Snake venom contains poorly immunogenic low MW substances. In sheep used to generate antibodies against the Desert Black Cobra, it took 18 months to generate sufficiently potent antibodies (al-Asmari et al. 1997). In the early work on Tiger Snake Antivenom, it took 3½ years to generate antiserum in a horse that was able to neutralise 1mg of venom per mL of horse serum (Tidswell 1902). Thus, we may be able to increase the alpaca serum potency for these venom types by a longer primary immunisation phase.

For greater increases in final product purity and potency, chromatographic methods could be applied. This could include Protein A or Protein G columns. Affinity chromatography to extract the specific antibodies required would be the method of choice. This is the method used to produce the ovine derived Crotalus (Rattlesnake) antivenom used extensively in North America (Ruha et al. 2002). This product has developed a reputation as an extremely safe and effective antibody product.

Other methods of assessing potency could be applied. We used antibody ELISA and mouse testing, but other methods of assessing the inhibition of toxic phospholipase enzymes show promise. Many methods of assessing indirect hemolysis have been tested (da Silva and Bier 1982, Gomez-Leiva and Pazos-Sanou 1983, de Araújo and Radvanyi 1987, Gutierrez et al. 1988, al-Abdulla et al. 1991). These methods rely on measuring indirect hemolysis of RBC in the presence of the phospholipase enzyme substrate egg yolk lecitihin. A kinetic enzyme assay for assessing activity has also been used (Isbister et al. 2010). Other methods include an in vitro chick biventer cervis assay system for neurotoxins (Kornhauser et al. 2010).

Assays that measure blood clotting are most relevant to assessing antivenoms. These can be manual or automated (Madaras et al. 2005). Brown snake venom may be better assessed by examining inhibition of clotting than lethality in mice (Tibballs et al. 1991).

Required snake antivenom products in Australia and PNG

In this project we have examined the medically important snakes in the Australian and Papua New Guinea region. The needs of the medical and veterinary profession in respect of snake antivenoms vary geographically. In Tasmania, the only venomous snake found is the Tiger Snake. In Victoria, the main venomous snakes are the Tiger Snake and Brown Snake. Red-Bellied Black snakes are common but are treated with Tiger-Brown antivenom. In South Australia, Tiger Snake, Brown Snake, Death Adder and King Brown (black snake family) occur. Bites from these snakes require either the specific monovalent product or a polyvalent antivenom product where snake identity is unknown. In Queensland, all five snake families occur, including the Taipan. In Western Australia, the situation is similar to South Australia. The Brown snake species differ but are treated by the Eastern Brown Snake (P. textilis) antivenom. In PNG, the Papuan Taipan is responsible for many envenomations along with Brown Snake and other snake species, treated by a five snake family polyvalent product.

Therefore in Australia, two main markets exist: (i) bivalent Tiger and Brown snake antivenom; and, (iii) polyvalent Tiger/Brown/Black/Death Adder/Taipan antivenom. This has been endorsed for the human market (O'Leary and Isbister 2009). Isbister has recommended only two products being on the market for human use: a low-volume bivalent Tiger-Brown antivenom, and a higher volume Polyvalent. He suggests that having only two products available “… would simplify the treatment of snakebite and reduce the risk of the incorrect monovalent antivenom being administered”.

It would appear that CSL monovalent antivenoms are really a polyvalent product (O'Leary and Isbister 2009, O'Leary et al. 2009). This is likely due to the immunisation program that CSL uses for its horses, in which each horse gets exposed to all venom types during its working life.

A single polyvalent or bivalent product would also simplify manufacturing, with only one set of processing required per batch.

In countries such as Costa Rica, only a single antivenom product is used. The product is polyvalent but the number of vials given depends upon the amount of information known about the snake. For example when no information on the identity of the snake is available 10 x 10mL vials of the polyvalent product are administered, which covers against snakes from the Bothrops, Lachecis and Crotalus families.

Implications

Commercial antivenom products

Based on the work in this project, alpaca derived antivenoms can compete with existing antivenoms in terms of final product potency and stability. The polyvalent alpaca antivenom was effective in preventing death in mice, with an equivalent potency to commercial antivenom products (CSL) against Brown Snake venom. However, Tiger Snake and Taipan were less potent than commercial monovalent products. Although potency was lower, this can be improved by further immunisations and more time. Like the Indian Cobra family of snakes, the most toxic elements of venom are the weakly immunogenic and low MW toxins. Further work needs to be done to improve the potency of the alpaca products, and refine a number of aspects of immunisation and serum processing before commercialisation is possible.

Alpaca serum industry development

The results of the work in this project show that Alpaca do have a place in the serum production industry, and that Australia is an ideal location for this.

A model for the global serum industry already exists in Australia. The company BTG International Pty Ltd2 is located in South Australia uses sheep for commercial serum production purposes. BTG maintains a large flock of merino wethers (10,000 head) that are used for serum production. BTG produce serum that is used for the North American Rattlesnake antivenom CroFab™. BTG also produce serum for DigiFab™, which is used for treating digitalis toxicity, and Voraxaze™, for treating toxicity induced by the chemotherapy drug methotrexate. BTG has a current share price of 350 pence per share. The company employs a number of people in South Australia and has a market capitalisation of $100m. The company was first listed in 1995 and has undergone various restructures throughout that time.

A serum production industry model based around alpaca could involve multiple serum producing herds supplying to a single bioprocessing plant. A specification for the type of alpaca would need to be developed for breeders to meet. Included in this specification could be animal bodyweight, sex, farm biosecurity status and possibly a blood test to measure circulating immunoglobulin levels. Once the animal specification has been met, this animal would then move to the immunisation farm and enter a quarantine period along with the appropriate disease tests. The animal would then receive a primary immunisation course, product potency determined, and then booster immunisations. Blood collection could take place every six to eight weeks. The whole blood could then be shipped immediately to the processing plant for product preparation or long term storage. Geographically, these farms would most likely need to be located south of the current Arbovirus line, which is presently in northern NSW. Arboviruses are insect spread viruses and could theoretically be transferred by serum. The main Arboviruses are Bluetongue, Akabane and Bovien Ephemeral Fever.

Horses, and to a far lesser extent sheep, have dominated the serum production industries around the world. There is no doubt that the use of animals for serum production is under challenge from the biotechnology industry. Non-animal methods for producing therapeutic antibodies are attracting considerable research. Recombinant DNA technology now allows small peptides to be created in non- mammalian species such as yeast. The use of chickens and egg yolk derived antibodies has had much research in the last decade. Chicken antibodies are useful and have been used for snake antivenom production. The problems with chickens are their poor ability to make antibodies to low MW

2 BTG Australasia Pty Ltd , RSD Turretfield RC, Holland Road, Rosedale, South Australia, Australia. Tel: +61 (0)8 8524 9700 . Fax: +61 (0)8 8524 9113http://www.btgplc.com/

substances and inherent allergic problems in humans to egg proteins. Despite this, animals have an enormous role to play going forward in the global serum production industry.

Figure 49. Current map produced by the National Arbovirus Monitoring Program managed by Animal Health Australia. Note the FREE zone located in the southern half of Australia. This would be the preferred location for any serum producing herds.

Demonstrating specific benefits of alpaca immunoglobulins

The benefits of using alpaca for serum production need to be strongly demonstrated. Alpaca may be capable of producing high quality antiserum products that have the unique camelid properties. However, more work must be done to demonstrate these properties. The animals as a production unit also have a lot to offer in terms of hardiness and ease of management.

Knowledge gaps

The current project was undertaken as a proof of concept work. It has proved the concept that alpaca can make potent antivenoms similar to horses. However there are a number of knowledge gaps that need to be addressed before progressing onto commercialisation is attempted. These knowledge gaps are outlined in the Recommendations section of this report.

Recommendations

Further work is required on the alpaca immunoglobulin project to bridge the gap between this simple proof of concept study and commercial reality.

This project has demonstrated that alpaca can be used to make effective antibodies that neutralise snake venoms, with potency comparable to the best commercially available equine products from the Australian Commonwealth Serum Laboratories.

The polyvalent alpaca antivenom product is effective in preventing lethality in mice, and has now demonstrated that the entire technological chain required for successful immunoglobulin production is possible in the alpaca. However, more work is required to improve the product potency, upscale the serum processing method, understand the intellectual property landscape, assess and manage biosecurity risks, understand regulatory compliance issues, and demonstrate that alpaca do indeed have similar physiochemical and immunological properties to camel immunoglobulins.

A number of projects are required for Phase 2, to fill the knowledge gaps and bring alpaca derived immunoglobulins closer to commercial reality. Perhaps the most important of these new projects is to understand the patent landscape relating to camelid antibodies. If there is not freedom to operate, then commercialisation would be difficult.

Other key pieces of work required are the disease risk assessment, and development of biosecurity protocols for alpaca. Disease transmission risks must be identified and managed for commercialisation and marketplace confidence in any alpaca derived serum products.

There are almost no studies specifically relating to alpaca immunoglobulins, despite the fact that alpaca belong to the camelid family. Work must be done to show that alpaca immungolobulins do indeed possess the unique properties described of camel immunoglobulins, such as lower allergenic potential and increased thermal stability. Once these properties are better defined within alpaca, this sets out the advantages of the alpaca serum as a production system for immunoglobulins, whatever the therapeutic application.

There are many applications other than snake antivenoms for alpaca immunoglobulins. It is necessary to start exploring these to better understand the market for such products. This will involve some travel to explore the various market opportunities. To bring a product to market requires meeting all the necessary regulatory compliance issues. These need to be understood, and strategies developed to meet these.

Alpaca serum also appears to require different processing conditions than what is typically used for horse immunoglobulins. A laboratory processing method was developed in this project which resulted in a satisfactorily concentrated immunoglobulin product. Work needs to be done to translate this bench-top method into a commercial scale system. This needs to be done on a contract basis, with a suitable bioprocessing partner company that has the necessary regulatory and technical know-how.

The potential commercial benefit to the alpaca industry is the development of an entirely new market. These animals could attract a premium price if produced to the specification required for immunoglobulin production. Based on commercial sheep serum production, the numbers of alpaca required would be in the order of thousands, and possibly up to ten thousand or more. An entrepreneurial investor would be required to provide funds required for start-up of a large scale alpaca serum farm, and the associated bioprocessing plant. Based on similar production models in sheep, this could be composed of multiple ‘farm’ sites and serum harvesting centres supplying to a single bioprocessing plant. The work described in this Phase 2 project plan is not specific to the production of snake antivenoms. This work is largely generic and equally applies to the production of

other antibody preparations. The suite of projects will be managed by the project leader Dr Andrew Padula.

Phase 2 projects

Improving the efficacy and safety of immunisation and bleeding

Alpaca react to oil based adjuvants with the formation of large subcutaneous lumps, which can develop into draining abscesses. Initial tolerance (<6 months) is good with oil adjuvants, but repeated doses at monthly intervals will ultimately be harmful to the animals. New adjuvants are available that have a much better safety profile and appear to be very well tolerated by animals. A system also needs to be developed to improve the efficiency of blood collection and ensure that this is performed to the standard required for GMP.

Optimising final product quality, clarity and stability

Alpaca serum is unique, and cannot be processed using existing protocols used for horses and sheep. Alpaca serum tends to be cloudy due to unwanted protein and lipid materials. Alpaca serum tends to form precipitates in the vial, and undesirably high turbidity, unless efforts are made to reduce these effects. The unwanted material must be removed to improve the product clarity and stability. These methods must be tested under commercial bioprocessing conditions. Product stability is essential, and this must be valuated under commercial conditions.

Infectious disease risk assessment and biosecurity plan

The injection of serum derived from an animal into either a human or another animal presents immediate disease transfer risks. These risks need to be identified and a system in place to manage these risks. An infectious disease risk assessment needs to be completed for alpaca serum immunoglobulins.

Regulatory compliance issues management

To register a therapeutic product for human or veterinary use, every step in the manufacturing process needs to be documented and performed to a minimum standard. These standards need to be developed and documented for each step of the production process.

Patent landscape search

Many patents have already been granted to various aspects of antibody production. Some patents have even been granted for specific aspects of camelid immunoglobulins. It is necessary to understand the existing patents to avoid potential infringement.

The method proposed to address the patent problem is to hire a patent attorney to perform a patent landscape search and assess the results. A firm in Melbourne that is recognized as an expert in antibodies has provided a cost estimate for this work.

Development of commercial and international linkages

A greater demand for therapeutic polyclonal antibody products exists outside of Australia than within Australia. The most common products include rabies antiserum, snake antivenoms, tetanus antitoxins and other infectious disease antibodies. Opportunities for product development need to be explored, and collaborative links developed.

Documentation of camelid immunoglobulin properties in alpaca

The unique properties of camelid immunoglobulins have only been described in any detail for the camel. Although alpaca are of the camelid family, there is no information available confirming that they possess all or some of the unique properties of camel serum.

A project could be contracted out to a group that has previously performed exactly this type of comparative work on horse, sheep, chicken and camel serum. The University of Costa Rica has performed this work in the past, and has the expertise and facilities to do this for alpaca serum. The unique properties described for camel immunoglobulins include increased thermal stability over equine immunoglobulins; reduced immunogenic potential in humans and other animals; and, enhanced enzyme inhibition properties. The project leader visited the University of Costa Rica during 2012 as part of PRJ-0777.

References

Agba, K., S. Gouro and M. Saley (1996). Lymph nodes in dromedary (Camelus dromedarius) 2. Pelvic limb, pelvis and abdomen lymph nodes. Revue vet Med 147(10): 721-730.

Agba, K., S. Gouro and M. Saley (1996). Lymph nodes in dromedary (Camelus dromedarius). 1. Cranial lymph nodes (head, neck, forelimb). Revue vet Med 147(8-9): 587.

al-Abdulla, I. H., A. M. Sidki and J. Landon (1991). An indirect haemolytic assay for assessing antivenoms. Toxicon 29(8): 1043-1046.

al-Asmari, A. K., I. H. al-Abdulla, R. G. Crouch, D. C. Smith and L. Sjostrom (1997). Assessment of an ovine antivenom raised against venom from the desert black cobra (Walterinnesia aegyptia). Toxicon 35(1): 141-145.

Anon (1957). Anti-venenes for snakebite - History and development. The Producers Review.

Barber, C. M., G. K. Isbister and W. C. Hodgson (2012). Solving the 'Brown snake paradox': in vitro characterisation of Australasian snake presynaptic neurotoxin activity. Toxicol Lett 210(3): 318-323.

Best, P. (1983). Snake envenomation. Proceedings No. 62 - Anaesthesia and Intensive Care. Sydney Postgraduate Foundation in Veterinary Science, Centre for Veterinary Education.

Best, P. (1998). Snake Envenomation of Companion Animals. Sydney Postgraduate Foundation in Veterinary Science.

Best, P. and S. K. Sutherland (1991). Snake envenomation. Proceedings No. 149 - Emergency Medicine and Critical Care, Sydney University, Centre for Veterinary Education.

Broad, A. J., S. K. Sutherland and A. R. Coulter (1979). The lethality in mice of dangerous Australian and other snake venom. Toxicon 17(6): 661-664.

Burnouf, T., E. Griffiths, A. Padilla, S. Seddik, M. A. Stephano and J. M. Gutierrez (2004). Assessment of the viral safety of antivenoms fractionated from equine plasma. Biologicals 32(3): 115- 128.

Cameron-Smith, R., L. Miloradovic, I. Cheyne and K. Healy (2000). The removal of viruses during the purification of equine antisera using filtration aids Hyflo Super-Cel and Fulmon Super A. Biologicals 28(3): 169-174.

Chotwiwatthanakun, C., R. Pratanaphon, S. Akesowan, S. Sriprapat and K. Ratanabanangkoon (2001). Production of potent polyvalent antivenom against three elapid venoms using a low dose, low volume, multi-site immunization protocol. Toxicon 39(10): 1487-1494.

Cook, D. A. (2010). Assessment of the potential of camelid antibodies to improve the treatment of snake envenoming. Doctor of Philosophy, University of Liverpool.

Cook, D. A., T. Owen, S. C. Wagstaff, J. Kinne, U. Wernery and R. A. Harrison (2010). Analysis of camelid antibodies for antivenom development: Neutralisation of venom-induced pathology. Toxicon 56(3): 373-380.

Cook, D. A., C. L. Samarasekara, S. C. Wagstaff, J. Kinne, U. Wernery and R. A. Harrison (2010). Analysis of camelid IgG for antivenom development: Immunoreactivity and preclinical neutralisation of venom-induced pathology by IgG subclasses, and the effect of heat treatment. Toxicon 56(4): 596- 603.

Cook, D. a. N., T. Owen, S. C. Wagstaff, J. Kinne, U. Wernery and R. a. Harrison (2010). Analysis of camelid IgG for antivenom development: Serological responses of venom-immunised camels to prepare either monospecific or polyspecific antivenoms for West Africa. Toxicon 56(April): 1-10.

da Silva, M. H. and O. G. Bier (1982). Titration of antiserum to South American rattlesnake (Crotalus durissus terrificus) venom by measuring inhibition of phospholipase A2 activity. Toxicon 20(3): 563- 569.

de Araújo, A. L. and F. Radvanyi (1987). Determination of phospholipase A2 activity by a colorimetric assay using a pH indicator. Toxicon 25(11): 1181-1188.

De Genst, E., D. Saerens, S. Muyldermans and K. Conrath (2006). Antibody repertoire development in camelids. Dev Comp Immunol 30(1-2): 187-198.

Fernandez, G. P., A. Segura, M. Herrera, W. Velasco, G. Solano, J. M. Gutierrez and G. Leon (2010). Neutralization of Bothrops mattogrossensis snake venom from Bolivia: experimental evaluation of llama and donkey antivenoms produced by caprylic acid precipitation. Toxicon 55(2-3): 642-645.

Ferrari, A., M. M. Rodriguez, P. Power, F. S. Weill, E. A. De Simone, G. Gutkind and J. Leoni (2007). Immunobiological role of llama heavy-chain antibodies against a bacterial beta-lactamase. Vet Immunol Immunopathol 117(3-4): 173-182.

Flight, S., P. Mirtschin and P. P. Masci (2006). Comparison of active venom components between Eastern brown snakes collected from South Australia and Queensland. Ecotoxicology 15(2): 133-141.

Foster, A., C. Bidewell, J. Barnett and R. Sayers (2009). Haematology and biochemistry in alpaca and llamas. In Practice: 276-281.

Fowler, M. E. (2011). Medicine and Surgery of South American Camelids. Third Edition. Singapore, Wiley-Blackwell.

Goldman, E. R., G. P. Anderson, J. Conway, L. J. Sherwood, M. Fech, B. Vo, J. L. Liu and A. Hayhurst (2008). Thermostable llama single domain antibodies for detection of botulinum A neurotoxin complex. Anal Chem 80(22): 8583-8591.

Gomez-Leiva, M. A. and L. Pazos-Sanou (1983). A simple method for the measurement of the hemolytic power of crude snake venoms. Rev Biol Trop 31(2): 341-342.

Grasset, E. (1957). Survey of assay methds of antivenins. Immunological factors influencing antivenin standardization. Bull World Health Organ 16: 79-122.

Grasset, E. and P. A. Christensen (1947). Enzyme purification of polyvalent antivenene against southern and equatorial African Colubrine and Viperine venoms. Trans R Soc Trop Med Hyg 41(2).

Gronski, P., F. R. Seiler and H. G. Schwick (1991). Discovery of antitoxins and development of antibody preparations for clinical uses from 1890 to 1990. Mol Immunol 28(12): 1321-1332.

Gutierrez, J. M., C. Avila, E. Rojas and L. Cerdas (1988). An alternative in vitro method for testing the potency of the polyvalent antivenom produced in Costa Rica. Toxicon 26(4): 411-413.

Hamers-Casterman, C., T. Atarhouch, S. Muyldermans, G. Robinson, C. Hamers, E. B. Songa, N. Bendahman and R. Hamers (1993). Naturally occurring antibodies devoid of light chains. Nature 363(6428): 446-448.

Harris, J. B. and C. A. Maltin (1981). The effects of the subcutaneous injection of the crude venom of the Australian common brown snake, Pseudonaja textilis on the skeletal neuromuscular system. Br J Pharmacol 73(1): 157-163.

Harrison, R. A., S. S. Hasson, M. Harmsen, G. D. Laing, K. Conrath and R. D. Theakston (2006). Neutralisation of venom-induced haemorrhage by IgG from camels and llamas immunised with viper venom and also by endogenous, non-IgG components in camelid sera. Toxicon 47(3): 364-368.

Harrison, R. A. and U. Wernery (2007). The unique properties of camelid IgG have potential to improve the treatment of snake bite. Camel Practice and Research 14(1): 15-16.

Hawgood, B. J. (1999). Doctor Albert Calmette 1863-1933: founder of antivenomous serotherapy and of antituberculous BCG vaccination. Toxicon 37(9): 1241-1258.

Herrera, M., J. Fernandez, M. Vargas, M. Villalta, A. Segura, G. Leon, Y. Angulo, O. Paiva, T. Matainaho, S. D. Jensen, K. D. Winkel, J. J. Calvete, D. J. Williams and J. M. Gutierrez (2012).

Comparative proteomic analysis of the venom of the taipan snake, Oxyuranus scutellatus, from Papua New Guinea and Australia: role of neurotoxic and procoagulant effects in venom toxicity. J Proteomics 75(7): 2128-2140.

Herrera, M., G. Leon, A. Segura, F. Meneses, B. Lomonte, J. P. Chippaux and J. M. Gutierrez (2005). Factors associated with adverse reactions induced by caprylic acid-fractionated whole IgG preparations: comparison between horse, sheep and camel IgGs. Toxicon 46(7): 775-781.

Homer, A. (1916). An Improved Method for the Concentration of Antitoxic Sera. J Hyg (Lond) 15(3): 388-400.

Isbister, G. K., M. A. O'Leary, J. Hagan, K. Nichols, T. Jacoby, K. Davern, W. C. Hodgson and J. J. Schneider (2010). Cross-neutralisation of Australian brown snake, taipan and death adder venoms by monovalent antibodies. Vaccine 28(3): 798-802.

Isbister, G. K., M. A. O'Leary, J. J. Schneider, S. G. Brown, B. J. Currie and A. S. P. Investigators (2007). Efficacy of antivenom against the procoagulant effect of Australian brown snake (Pseudonaja sp.) venom: in vivo and in vitro studies. Toxicon 49(1): 57-67.

Jones, R. G., R. L. Corteling, G. Bhogal and J. Landon (1999). A novel Fab-based antivenom for the treatment of mass bee attacks. Am J Trop Med Hyg 61(3): 361-366.

Kellaway, C. H. (1931). Observations on the certainly lethal dose of the venom of the common brown snake for the common laboratory animals. The Medical journal of Australia(2): 747-751.

Kornhauser, R., A. J. Hart, S. Reeve, A. I. Smith, B. G. Fry and W. C. Hodgson (2010). Variations in the pharmacological profile of post-synaptic neurotoxins isolated from the venoms of the Papuan (Oxyuranus scutellatus canni) and coastal (Oxyuranus scutellatus scutellatus) taipans. Neurotoxicology 31(2): 239-243.

Kurtovic, T., A. Leonardi, M. Lang Balija, M. Brgles, L. Habjanec, I. Krizaj and B. Halassy (2012). The standard mouse assay of anti-venom quality does not measure antibodies neutralising the haemorrhagic activity of Vipera ammodytes venom. Toxicon 59(7-8): 709-717.

Landon, J. and D. S. Smith (2003). Merits of Sheep Antisera for Antivenom Manufacture. Toxin Reviews 22(1): 15-22.

Lauwereys, M. and M. A. Ghahroudi (1998). Potent enzyme inhibitors derived from dromedary heavy-chain antibodies. The EMBO Journal 17(13): 3512-3520.

Leon, G., B. Lomonte and J. M. Gutierrez (2005). Anticomplementary activity of equine whole IgG antivenoms: comparison of three fractionation protocols. Toxicon 45(1): 123-128.

Maass, D. R., J. Sepulveda, A. Pernthaner and C. B. Shoemaker (2007). Alpaca (Lama pacos) as a convenient source of recombinant camelid heavy chain antibodies (VHHs). J Immunol Methods 324(1-2): 13-25.

Madaras, F., P. J. Mirtschin and T. I. M. Kuchel (2005). Antivenom Development in Australia. Toxin Reviews 24(1): 79-94.

Meddeb-Mouelhi, F., B. Bouhaouala-Zahar, Z. Benlasfar, M. Hammadi, T. Mejri, M. Moslah, H. Karoui, T. Khorchani and M. E. Ayeb (2003). Immunized camel sera and derived immunoglobulin subclasses neutralizing Androctonus australis hector scorpion toxins. Toxicon 42(7): 785-791.

Mensua, C., L. Carrasco, M. J. Bautista, E. Biescas, A. Fernandez, C. L. Murphy, D. T. Weiss, A. Solomon and L. Lujan (2003). Pathology of AA Amyloidosis in Domestic Sheep and Goats. Veterinary Pathology 40(1): 71-80.

Meyer, B. K., A. Ni, B. Hu and L. Shi (2007). Antimicrobial preservative use in parenteral products: past and present. J Pharm Sci 96(12): 3155-3167.

Mirtschin, P. J., N. Dunstan, B. Hough, E. Hamilton, S. Klein, J. Lucas, D. Millar, F. Madaras and T. Nias (2006). Venom yields from Australian and some other species of snakes. Ecotoxicology 15(6): 531-538.

Mirtschin, P. J., P. Masci, D. C. Paton and T. Kuchel (1998). Snake bites recorded by veterinary practices in Australia. Aust Vet J 76(3): 195-198.

Morgan, F. (1954). The Australian Taipan Oxyuranus scutellatus scutellatus. First International Conference on Venoms. Annual Meeting of the American Association for the Advancement of Science. E. Buckley and N. Porges. Berkeley, California.

Nguyen, V. K., A. Desmyter and S. Muyldermans (2001). Functional heavy chain antibodies in Camelidae. Advances in Immunology 79.

O'Leary, M. A. and G. K. Isbister (2009). Commercial monovalent antivenoms in Australia are polyvalent. Toxicon 54(2): 192-195.

O'Leary, M. A., R. S. Kornhauser, W. C. Hodgson and G. K. Isbister (2009). An examination of the activity of expired and mistreated commercial Australian antivenoms. Trans R Soc Trop Med Hyg 103(9): 937-942.

Paul, K., J. Manjula, E. P. Deepa, Z. E. Selvanayagam, K. A. Ganesh and P. V. Subba Rao (2007). Anti-Echis carinatus venom antibodies from chicken egg yolk: isolation, purification and neutralization efficacy. Toxicon 50(7): 893-900.

Petras, D., L. Sanz, A. Segura, M. Herrera, M. Villalta, D. Solano, M. Vargas, G. Leon, D. A. Warrell, R. D. Theakston, R. A. Harrison, N. Durfa, A. Nasidi, J. M. Gutierrez and J. J. Calvete (2011). Snake venomics of African spitting cobras: toxin composition and assessment of congeneric cross-reactivity of the pan-African EchiTAb-Plus-ICP antivenom by antivenomics and neutralization approaches. J Proteome Res 10(3): 1266-1280.

Pratanaphon, R., S. Akesowan, O. Khow, S. Sriprapat and K. Ratanabanangkoon (1997). Production of highly potent horse antivenom against the Thai cobra (Naja kaouthia). Vaccine 15(14): 1523-1528.

Redwan el, R. M., A. Khalil and Z. Z. El-Dardiri (2005). Production and purification of ovine anti- tetanus antibody. Comp Immunol Microbiol Infect Dis 28(3): 167-176.

Rojas, G., J. M. Jimenez and J. M. Gutierrez (1994). Caprylic acid fractionation of hyperimmune horse plasma: description of a simple procedure for antivenom production. Toxicon 32(3): 351-363.

Rojas, G., V. Nu, R. Otero, J. M. Gutie, A. Dõâ, E. Miranda, A. F. Uribe, J. F. Silva, J. G. Ospina, Y. Medina, M. F. Toro, M. E. Garcõâ, G. Leo, Y. Mena, N. Gonza and L. C. Arenas (1999). A randomized blinded clinical trial of two antivenoms, prepared by caprylic acid or ammonium sulphate fractionation of IgG, in Bothrops and Porthidium snake bites in Colombia: correlation between safety and biochemical characteristics of antivenoms. Toxicon 37: 895-908.

Rojas, G., M. Vargas, A. Robles and J. M. Gutierrez (1993). Turbidity of hyperimmune equine antivenom: the role of phenol and serum lipoproteins. Toxicon 31(1): 61-66.

Ruha, A. M., S. C. Curry, M. Beuhler, K. Katz, D. E. Brooks, K. A. Graeme, K. Wallace, R. Gerkin, F. Lovecchio, P. Wax and B. Selden (2002). Initial postmarketing experience with crotalidae polyvalent immune Fab for treatment of rattlesnake envenomation. Ann Emerg Med 39(6): 609-615.

Russell, F. E. (1988). Snake venom immunology: Historical and practical considerations. J Toxicol Toxin Reviews 7(1): 1-82.

Sells, P. G. (2003). Animal experimentation in snake venom research and in vitro alternatives. Toxicon 42(2): 115-133.

Solano, G., A. Segura, M. Herrera, A. Gomez, M. Villalta, J. M. Gutierrez and G. Leon (2010). Study of the design and analytical properties of the lethality neutralization assay used to estimate antivenom potency against Bothrops asper snake venom. Biologicals 38(5): 577-585.

Sriprapat, S., S. Aeksowan, S. Sapsutthipas, C. Chotwiwatthanakun, P. Suttijitpaisal, R. Pratanaphon, O. Khow, V. Sitprija and K. Ratanabanangkoon (2003). The impact of a low dose, low volume, multi- site immunization on the production of therapeutic antivenoms in Thailand. Toxicon 41(1): 57-64.

Standardization, W. E. C. o. B. (2010). WHO Guidelines for the production control and regulation of snake antivenom immunoglobulins. Geneva, Switzerland, World Health Organisation.

Steinbuch, M. and R. Audran (1969). The isolation of IgG from mammalian sera with the aid of caprylic acid. Arch Biochem Biophys 134(2): 279-284.

Sutherland, S. (1979). Australian venoms and care of the envenomed patient. Doctor of Medicine, University of Melbourne.

Sutherland, S. K. (1992). Deaths from snakebite in Australia. The Medical journal of Australia 157(7): 740.

Sutherland, S. K. and J. Tibballs (2001). Australian animal toxins : the creatures, their toxins, and care of the poisoned patient. Melbourne, Oxford University Press.

Theakston, R. D., M. J. Lloyd-Jones and H. A. Reid (1977). Micro-ELISA for detecting and assaying snake venom and venom-antibody. Lancet 2(8039): 639-641.

Theakston, R. D. and H. A. Reid (1979). Enzyme-linked immunosorbent assay (ELISA) in assessing antivenom potency. Toxicon 17(5): 511-515.

Tibballs, J. and S. K. Sutherland (1991). The efficacy of antivenom in prevention of cardiovascular depression and coagulopathy induced by Brown Snake (Pseudonaja) species venom. Anaesth Intensive Care 19: 530-534.

Tibballs, J., S. K. Sutherland and S. Kerr (1991). Studies on Australian snake venoms, Part II: The haematological effects of brown snake (Pseudonaja) species in the dog. Anaesth Intensive Care 19(3): 338-342.

Tidswell, F. (1902). A preliminary note on the serum-therapy of snake-bite. The Australiasian Medical Gazette.

Tornquist, S. J. (2009). Clinical pathology of llamas and alpaca. Vet Clin North Am Food Anim Pract 25(2): 311-322.

Trevett, A. J., D. G. Lalloo, N. C. Nwokolo, S. Naraqi, I. H. Kevau, R. D. Theakston and D. A. Warrell (1995). The efficacy of antivenom in the treatment of bites by the Papuan taipan (Oxyuranus scutellatus canni). Trans R Soc Trop Med Hyg 89(3): 322-325.

van der Linden, R. H. J., L. G. J. Frenken, B. de Geus, M. M. Harmsen, R. C. Ruuls, W. Stok, L. de Ron, Wilson, S., P. Davis and C. T. Verrips (1999). Comparison of physical chemical properties of llama V-HH antibody fragments and mouse monoclonal antibodies. Biochimica Et Bio-physica Acta- Protein Structure and Molecular Enzymology 1431 (1): 37-e46.

Vargas, M., A. Segura, M. Herrera, M. Villalta, R. Estrada, M. Cerdas, O. Paiva, T. Matainaho, S. D. Jensen, K. D. Winkel, G. Leon, J. M. Gutierrez and D. J. Williams (2011). Preclinical evaluation of caprylic acid-fractionated IgG antivenom for the treatment of Taipan (Oxyuranus scutellatus) envenoming in Papua New Guinea. PLoS Negl Trop Dis 5(5): e1144.

Waghmare, A., R. L. Deopurkar, N. Salvi, M. Khadilkar, M. Kalolikar and S. K. Gade (2009). Comparison of Montanide adjuvants, IMS 3012 (Nanoparticle), ISA 206 and ISA 35 (Emulsion based) along with incomplete Freund's adjuvant for hyperimmunization of equines used for production of polyvalent snake antivenom. Vaccine 27(7): 1067-1072.

White, J. (1991). Snakebite: an Australian perspective. Journal of Wilderness Medicine 2: 219-244.

White, J. (1998). Envenoming and antivenom use in australia. Toxicon 36(11): 1483-1492.

WHO (2007). Report of a consultative meeting. Rabies and envenomings: a neglected public health issue: Report of a consultative meeting. Geneva, Switzerland, World Health Organisation.

Wingfield, P. (1998). Protein precipitation using ammonium sulfate. Current Protocols in Protein Science Supplement 13: Appendix 3F.

Ziska, S. (2012). Effects of serial harvest of plasma on total plasma protein and total immunoglobulin G concentrations in donor horses involved in a plasmapharesis program. Am J Vet Res 73(6).

92 Alpaca Immunoglobulins By Andrew Padula

This report documents the findings on the use of alpaca for producing a therapeutic immunoglobulin product (snake antivenom), for treating snake bite envenomation.

This report is targeted at those interested in alternative uses to fibre production for alpaca in Australia. The report has been prepared to be read by someone with a basic understanding of science.

RIRDC is a partnership between government and industry to invest in R&D for more productive and sustainable rural industries. We invest in new and emerging rural industries, a suite of established rural industries and national rural issues.

Most of the information we produce can be downloaded for free or purchased from our website .

RIRDC books can also be purchased by phoning 1300 634 313 for a local call fee.

Phone: 02 6271 4100 Fax: 02 6271 4199 Bookshop: 1300 634 313 Email: [email protected] Postal Address: PO Box 4776, Kingston ACT 2604 Street Address: Level 2, 15 National Circuit, Barton ACT 2600