||||||||||||||III USOO51961.93A United States Patent (19) (11 Patent Number: 5,196,193 Carroll (45) Date of Patent: Mar. 23, 1993

(54) ANTIVENOMS AND METHODS FOR W. G. Martin and W. H. Cook, Can. J. Biochem. MAKING ANTIVENOMS Physiol. 36153 (1958). Wingert and Wainschel, 5. Med. J. 68, 1015 (1975). (75) Inventor: Sean B. Carroll, Cottage Grove, Wis. Christopher and Rodning, 5. Med. J. 79, 159 (1986). 73) Assignee: Ophidian Pharmaceuticals, Inc., Ellenhorn and Barceloux, Medical Toxicology, Ch. 39, Madison, Wis. Elsevier Press (1988). Parrish, Public Health fit. 81, 269 (1986). (21) Appl. No.: 429,791 Russell, JAMA 233, 341 (1975). 22 Filed: Oct. 31, 1989 Ayeb and DeLori, In: Handbook of Natural Toxins, vol. 2, insect Poisons, Allergens, and Other Inverte 51) int. Cl...... A61K 39/395 brate Venoms (Anthony T. Tu, Ed.) (Marcel Dekker 52) U.S. Cl...... 424/85.8; 530/387.1; 1984) Ch. 18, pp. 607-638. 530/389.1; 530/856; 530/858 Hassan, In: Handbook of Natural Toxins, vol. 2, Insect 58) Field of Search ...... 530/387, 387.1, 389.1; Poisons, Allergens, and Other Invertebrate Venoms, 424/85.8 (Anthony T. Tu, Ed.) (Marcel Dekker 1984) Ch. 17, pp. (56) References Cited 577-605. Lumley et al., Med. J. Aust. 148, 527 (1988). U.S. PATENT DOCUMENTS Endean, Toxicon 25, 483 (1987). 3,415,804 12/1968 Polson ...... 260/112 Olson et al., Toxicon 32,733 (1984). 4,165,370 7/1979 Coval ...... 424/85 Baxter and Marr, Toxicon 7, 195 (1969). 4,357,272 1/1982 Polson ...... 260/12 Habermehl, Venomous Animals and Their Toxins, 4,554,019 10/1985 Polson ...... 424/85 4,661,346 4/1987 New et al... 424/85.8 Springer-Verlag, Berlin (1981). 4,748,08 5/1988 Stolle et al...... 424/87 MacDonald et al., Am. J. Epidemiology 124, 794 4,806,346 2/1989 Hum ...... 424/85.8 (1986). 4,849,352 7/1989 Sullivan et al...... 435/69 Tacket et al., Am. J. Med. 76, 794 (1984). Thorne and Gorbach, Pharmacology of Bacterial Tox OTHER PUBLICATIONS ins, In: International Encyclopedia of Pharmacology B. S. Thalley and S. B. Carroll, Bio/Technology 8:934 and Therapeutics, Dorner and Drews (Eds.), Pergamon (1990). Press, Oxford (1986), pp. 5-16. T. H. Jukes et al., J. Immunol. 26:353 (1934). Russell, JAMA 215, 1994 (1971). A. Buxton, J. Gen. Microbiol. 7:268 (1952). (List continued on next page.) E. Orlans, Immunology 12:27 (1967). E. D. Heller, Res. Vet. Sci. 18:117 (1975). Primary Examiner-Robert A. Wax C. G. Aulisio and A. Shelokov, Proc. Soc. Exp. Biol. Assistant Examiner-R. Baker Med. 131:1 150 (1969). Attorney, Agent, or Firm-Haverstock, Medlen & R. E. Faith and L. W. Clem, Immunology 25:151 Carroll (1973). 57) ABSTRACT T. T. Kramer and H. C. Cho, Immunology 19:157 The production of antivenoms in non-mammals and (1970). improvements in the effectiveness of both non-mam G. Ramon, Comptes. Rendus des Seances de la Societe malian antivenoms and mammalian antivenoms so that de Biologie 99:1476 (1928). they are more suitable for treatment of humans and S. Schmidt et al., Det. Kgl. Danske videnskabernes animals as well as for analytical use. Selskab. Biologiske Meddelelser 12:1 (1936). R. A. Stedman et al., J. of Comp. Pathology 79:4 (1969). H. Yamamoto et al., Japan J. Vet. Res. 23:4 (1975). 31 Claims, 16 Drawing Sheets 5,196, 193 Page 2

OTHER PUBLICATIONS Eds.), Am. Assoc. Advan. Sci., Washington, D.C. Audibert et al., Proc. Natl. Acad. Sci., USA 79, 5042 (1956), pp. 373-380. (1982). Gingrich and Hohenadel, In: Venoms (E. E. Buckley Alouf, Ann. Inst. Pasteur/Microbiol. 136B, 309 (1985). and N. Porges, Eds.) Am. Assoc. Advan. Sci. Washing New et al., New Engl. J. Med. 311, 56 (1984). ton, D.C. (1956), pp. 381-385. Freitas et al., Toxicon 27, 341 (1989). Shulov et al., Harefuah. 3. Med. Assn. Israel 56, 55 Perez et al., Toxicon 22,967 (1984). (1959). Iddon et al., Toxicon 26, 167 (1988). Bahraoui et al., 3. Immunol. 136, 337 (1986). Martinez et al., Toxicon 27, 239 (1989). Kornalik and Táborská, Toxicon 10, 1135 (1989). Russell et al., Toxicon 8, 63 (1970). Ownby et al., Toxicon 21, 849 (1983). Curry et al., J. Toxicol.-Clin. Toxicol. 21, 417 Grasset, Trans. Roy. So. Trop. Med. Hyg. 26, 267 (1983-1984). (1932). World Health Organization Publication No. 58, Geneva Greval, Ind. Jour. Med. Res. 22, 2 (1934). (1981). Pope, Brit. 3. Exp. Path. 20, 132 (1939). Sutherland, Med. J. Aust. 1, 613 (1977). Brit. 3. Exp. Path 20, 201 (1939). Christensen, In: Snake Venoms, (Springer-Verlag, Ber Mebs et al., Toxicon 26, 453 (1988). lin, 1979) Ch. 20, pp. 825-846. Dos Santos et al., Toxicon 27, 297 (1989). Parrish and Hayes, Clin. Tox. 3, 501 (1970). Coulter et al., Med. 3. Aust. 1, 433 (1980). Yang et al., Toxicon 15, 51 (1977). Ho et al., Am. 3. Trop. Med. Hyg. 35, 579 (1986). Kukongviriyapan et al., J. Immunol. Meth. 49, 97 Ho et al., Toxicon 24, 211 (1986). (1982). Theakston and Reid, Toxicon 17, 511 (1979). Karlsson et al., Eur. 3, Biochem. 21, 1 (1971). Theakston, Toxicon 21, 341 (1983). Lomonte et al., Toxicon 23, 807 (1985). Chandler and Hurrell Clinica Chemica Acta 121, 225 Sullivan et al., 3, Vet. Hum. Toxicol. 24, 192 (suppl.) (1982). (1982). Minton et al., Clin. Toxicol. 22, 303 (1984). Sullivan and Russell, Proc. Western Pharmacol. Soc. Minton, Ann. Energ. Med. 16,932 (1987). 25, 185 (1982). Sullivan and Russell, Toxicon Suppl. 3, 429 (1983). Weinstein et al., Toxicon 23, 825 (1985). Jeter et al., Toxicon 21, 729 (1983). Schaeffer et al., Toxicon 26, 67 (1988). Bar-Or et al., Clin. Tox. 22, 1 (1984). Rose et al., Eur. 3. Immunol. 4, 52 (1974). Russell et al., Am. 3, Trop. Med. Hyg. 34, 141 (1985). Jensenius et al., 3. Immunol. Methods 46, 63 (1981). Sullivan, Ann. Emerg. Med. 16,938 (1987). Poison et al., Immunol. Commun. 9, 475 (1980). Johnstone and Thorpe, Immunochemistry in Practice Poison et al., Immunol. Invest. 14, 323 (1985). 2d Edition (Blackwell Scientific Publications 1987) p. Stuart et al., Anal. Biochem. 173, 142 (1988). 209. Song, et al., 3. Immunol. 135, 3354 (1985). Polson et al., Immunol. Comm. 9, 495 (1980). Accurate Chemical and Scientific Corporation (West Benson et al., 3, Immunol. 87, 610 (1961). bury, N.Y.), 1989 product catalogue. Benedict and Yamaga, In: Comparative Immunology Bauwens et al., Clin. Chim. Acta 170, 37 (1987). (3.3, Marchaloni Ed.), Ch. 13, Immunoglobulins and Zrein et al., Arch. Virol. 90, 197 (1986). Antibody Production in Avian Species (Blackwell, Ricke et al., Appl. Environ. Microbiol. 54,596 (1988). Oxford 1966), pp. 335-375. Vieira et al., 3. Immunoassay 7, 57 (1986). Patterson et al., 3. Immunol. 89, 272 (1962). Burger et al., Thrombosis Research 40, 283 (1985). Carroll and Stollar, 3. Biol. Chen. 258, 24 (1983). Van Regenmortel and Burckard, 3. Virol. Methods 11, Phillpot, 3r. et al., Toxicon 16, 603 (1978). 217 (1985). Weber and Osborn, In: The Proteins, 3rd Edition, (H. Bar-Joseph and Malkinson (1980) 3. Virol Methods 1, Neurah and R. L. Hill, Eds.) (Academic Press, NY 179 (1980). 1975) pp. 179-223. Altschuh et al., 3. Immunol. Methods 69, 1 (1984). Carroll and Laughon, In: DNA Cloning: A Practical Approach, vol. III, D. Glover, Ed., (IRL Press, Oxford Bartz et al., 3. Infect. Dis. 142, 439 (1980). 1987), pp. 89-111. -- Piela et al., Avian Dis. 28,877 (1984). Theakston, In: Natural Toxins. Animal, Plant and Mi Piela et al., Avian Dis. 29, 457 (1985). crobial, (3.B. Harris, Ed.), Clarenden Press, Oxford Mohammed et al., Avian Dis. 30, 398 (1986). 1986, pp. 287-303. Mohammed et al., Avian Dis. 30, 389 (1986). Ohsaka, In: Snake Venoms, (C. Y. Lee, Ed.) Handbook of Experimental Pharmacology, vol. 52, Springer Ver Motha, Aust. Vet. 3.. 64, 259 (1987). lag, Berlin 1979, pp. 480-546. Yolken et al., Pediatrics 81, 291 (1987). Martin et al., 3. Biol. Chem. 262, 4452 (1987). Bernfeld and Wan, Science 142, 678 (1963). Kabat, Structural Concepts in Immunology and Immu Carrel et al., Nature 221, 386 (1969). nochemistry (Holt, Rinehart, and Winston, NY, 1968). Physician's Desk Reference, 43rd edition, pp. Martinez-Hernandez et al., 3. Histochen. Cytochem. 2277-2278, Medical Economics Company 1989. 23, 146 (1975). Kunkel et al., 3. Toxicol.-Clin. Toxicol. 21, 503 McGuire et al., Molecular Immunology 16, 787 (1979). (1983-84). Sterogene Biochemicals, Brochure, "Actisep TM In Criley, In: Venoms (E. E. Buckley and N. Porges, munoaffinity Purification System'. U.S. Patent Mar. 23, 1993 Sheet 1 of 16 5,196,193

FIG. 1 LAYING HENS J. COCKTAL OF ENACTIVATED VENOMS

MMUNIZATION BOOST WITH INCREASING DOSES OF NATIVE VENOM COLLECTEGGS -> STORE AT 4°C J. YOLKS AND PEG

- TOTALYOLK MMUNOGLOBULIN / l COVALENTLY - LINKED VENOMANTIGEN MATRIX

WASH

UNBOUND PROTEIN DISCARDED SECONDELUTION WITH DENATURANT RECYCLES MATRIX FIRSTELUTION, NON-DENATURING

AFFINITY PURIFIED ANTIVENOM

--> IN VIVOTHERAPY U.S. Patent Mar. 23, 1993 Sheet 2 of 16 5,196,193

FIG. 2

VENOMA (J-A-O ) VENOMB (O-A-O ) l MMUNIZE J MMUNIZE ANTIVENOMA ANTIVENOMB

MATRIX A'

FINAL FINAL FLOW-THROUGH FLOW-THROUGH

ANT-VENSEE ANT-VENOMB-SPECIFIC MONOVALENT MONOVALENT SUBPOPULATION SUBPOPULATION (, O) (O,O)

ANTI-VENOM A- AND B-SPECIFIC CROSS-REACTIVE SUBPOPULATION (A )

U.S. Patent Mar. 23, 1993 Sheet 4 of 16 5,196,193

FIG. 4

VENOMA ( I-A-O ) VENOMB (D-A-O ) J IMMUNIZE TOGETHER

ANTIVENOM AB

ELUATE

N ABCOCKTAI MATRIX ANTI-VENOMA, B-SPECIFIC ALL SUBPOPULATIONS 4) (, A, O, O, O.) U.S. Patent Mar. 23, 1993 Sheet 5 of 16 5,196,193

SENVT U.S. Patent Mar. 23, 1993 Sheet 6 of 16 5,196,193

: s : g C X

Ovw U.S. Patent Mar. 23, 1993 Sheet 7 of 16 5,196,193

C?EZINTIWIWINT) TOHLNOO

A61HONOLLITTIG #-0!

U.S. Patent Mar. 23, 1993 Sheet 9 of 16 5,196,193

CO O'8 Hod -- OH - BNCNWS) W7 -- S8.

SH 9'. O. CN cHis OddOS WO

-- BSDOW

-- S8.

-- HSWWNBW LS8. -H HSWW S8

HOWO

e g g 08aw U.S. Patent Mar. 23, 1993 Sheet 10 of 16 5,196,193

FIG. 10

2 3 4 5 6 7 8 9 1 O

6 -

6

U.S. Patent Mar. 23, 1993 Sheet 11 of 16 5,196,193

2.6 O ON C. durissus terrificus 22 A ON C. atrox 18 X NON - IMMUNE HORSElgG 4. C. durissus terrificus - SPECIFIC 1.0 MONOVALENT SUBPOPULATION 0.6 0.2 FIG. A 10.0 5.0 1.0 0.1 Ab CONCENTRATION (ug/ml)

2.6 O ON C. duriSSUS terrificus 22 A ON C. atrox 1.8 X NON - IMMUNE HORSE IgG 1.4 C. atrOx-SPECIFIC MONCVALENT SUBPOPULATION 1.0 0.6 FIG. 11B 0.2 100 5.0 1.0 0.1 Ab CONCENTRATION (ug/ml)

2.6 O ON Courissus terrificus 22 A ON C. atrox 1.8 X NON - IMMUNE HORSElgG O 1.4 C. atrox/ C C. durissus terrificus - SPECIFIC 1.0 POLYVALENT SUBPOPULATION 0.6 FIG. 11C 0.2 100 5.0 1.0 0.1 Ab CONCENTRATION (ug/ml)

U.S. Patent Mar. 23, 1993 Sheet 13 of 16 5,196,193

(?Ed)TVIHELVW?NILHVLSvTOOd (dBS)BLwnTBdESILOVVTOOd (OHL)BLwnTEENIGINVnлTOOd

Ov W

U.S. Patent Mar. 23, 1993 Sheet 15 of 16 5,196, 193

F.G. 15

U.S. Patent Mar. 23, 1993 Sheet 16 of 16 5,196,193

F.G. 16

2 3 4 5 6 7 8 9 10 1 1 12 13 14

16 97 -

66 -

. 45 -

36 - 29 24

20

14 - 2 5, 196, 193 1 2 (Order Scorpiones) produce the most significant ven ANTIVENOMS AND METHODS FOR MAKING oms. While scorpion venoms are also complex mixtures, ANTIVENOMS there has been some success identifying their active agents. Approximately thirty different protein neuro DESCRIPTION toxins, each having a molecular weight of about 7000 Background of the Invention daltons, have been isolated. M. E. Ayeb and P. Delori, In: Handbook of Natural Toxins, Vol.2, Insect Poisons, The present invention relates to antivenoms suitable Allergens, and Other Invertebrate Venoms, (Anthony for treatment of humans and animals as well as for ana T. Tu, Ed.)(Marcel Dekker 1984, Chapter 18 (pp. lytical use. 10 607-638). Of the approximately 650 scorpion species, I. VENOMS the most dangerous belong to the Buthidae family and A toxin is a single protein or peptide that has deleteri the genuses Tityus (North and South America), Centru ous effects in man or animals. A venom comprises a roides (U.S. and Mexico), Centrurus (Mexico), Androc plurality of toxins; they are relatively complex mixtures tonus (Mediterranean/North Africa), Buthacus (Medi of proteins and peptides that can cause considerable 15 terranean/North Africa), Leiurus (Mediterranean/ morbidity and mortality in humans and animals. North Africa), Buthotus (Mediterranean/North Af The chemical actions of and biological reactions to rica), Buthus (Mediterranean/North Africa), and venoms are as diverse as their sources. Depending on Parabuthus (South Africa). F. Hassan, In: Handbook of the nature of the venoms, their toxic effects may be Natural Toxins, Vol.2, Insect Poisons, Allergens, and evident in the cardiovascular, hematologic, nervous, 20 Other Invertebrate Venoms, (Anthony T. Tu, Ed.)(- and/or respiratory systems. Marcel Dekker 1984), Chapter 17 (pp. 577-605). iii) Each region of the world has its own particularly Coelenterata. In the Coelenterata phylum, jelly fish are troublesome venomous species. Within Eukaryota (see an important venomous species; the venom from Chiro Table 1), some specific venom sources from the Ani 25 nex fleckeri is among the most potent and medically malia kingdom are most notable. significant. In the waters off Northern Australia, about A. EUKARYOTA one fatality occurs each year. J. Lumley et al., Med. J. i) Chordata. A number of Chordata classes are Aust. 148,527 (1988). Several toxic fractions have been sources of venoms (e.g. Amphibians, Fish, Reptiles). characterized from C. fleckeri venom including two Among Reptiles, the most significant order is snakes. 30 high molecular weight myotoxins (R. Endean, Toxicon Snake venom is a relatively complex mixture of en 25, 483 (1987)) and several low molecular weight toxins zymes, non-enzymatic proteins and peptides, and as yet having hemolytic or dermonecrotic properties (C. E. Olson et al., Toxicon 22,733 (1984); E. H. Baxter and A. unidentified compounds. W. A. Wingert and J. G. M. Marr, Toxicon 7, 195 (1969)). iv) Mollusca. In the Wainschel, S.Med. J. 68:1015 (1975). D. C. Christopher 35 Mollusca phylum, the most significant venomous mem and C. B. Rodning, S. Med. J. 79:159 (1986). bers are the coneshells (Conidae) which produce potent TABLE 1. myotoxins that can be fatal. G. G. Habermehl, Venom Phylogeny of Toxin- and Venom-Producing Organisms ous Animals and Their Toxins (Springer-Verlag, Berlin SUPERKINGDOM: PROKARYOTA 1981). Little is known about the structure of the mollus KINGDOM: MONERA 40 can myotoxins. - DIVISION: BACTERIA SUPERKINGDOM: EUKARYOTA B. PROKARYOTA KINGOOM: FUNGI KINGDOM: PLANTAE Prokaryotes are an important source of toxins. Most KINGDOM: ANIMALA PHYLUM: CHORDATA 45 bacterial toxins, for example, are well known. Among CLASSES: Amphibia species of bacteria, the most notorious toxin sources are Reptilia certainly Clostridum botulinum and Clostridium parabot Pisces PHYLUM: ARTHROPODA ulinum. The species produce the neurogenic toxin CLASSES: Arachnida known as botulinus toxin. While a relatively rare occur Insecta rence in the United States, involving only 355 cases Myriapoda 50 between 1976 and 1984 (K. L. MacDonald et al., Am J. PHYLUM: COELENTERATA Epidemiology 124,794 (1986)), the death rate due to the PHYLUM: MOLLUSCA botulism toxin is 12% and can be higher in particular risk groups. C. O. Tacket et al., Am. J. Med. 76, 794 While there are some chemical similarities, the venom (1984). of each species exhibits its own characteristic toxicity. 55 Many other bacteria produce protein toxins of signifi M. J. Ellenhorn and D. G. Barceloux, Medical Toxicol cance to humans, including Bacillus anthracis, Bordetella ogy, Ch.39 (Elsevier Press (1988). pertussis (diptheria), Pasteurella pestis, Pseudomonas ae Of the over 100 species of snakes in the United States, ruginosa, Streptococcos pyrogenes, Bacillus cereus, E. coli, approximatelylic Health Rpt. 10%81.269 are (1966). poisonous. The majorityH. M. Parrish, of these Pub are 60 Shigella, Staphylococcus aureus, ibrio cholerae, and Clo. from the family Crotalidae. The venomous species in stridium tetani. Thorne and Gorbach, Pharmacology of clude the rattlesnakes (Crotalus), cottonmouths and Bacterial Toxins, In: International Encyclopedia of copperheads (Agkistrodon), and pigmy and massas Pharmacology and Therapeutics, F. Dorner and J. sauga rattlesnakes (Sistrurus). There are also poisonous Drews (eds.), Pergamon Press, Oxford (1986), pp. 5-16. members of the Elapidae family, the coral snakes (Mi 65 II. TREATMENT cruroides). F. E. Russell et al., JAMA 233:341 (1975), ii) Arthropoda. In the Arthropoda phylum, an important As noted above, a toxin is defined as a single protein class is Arachnida. Among Arachnida, scorpions or peptide and a venom is defined as comprising a plu 5,196, 193 3. 4. rality of toxins. Both toxin and venom have been used as i. Raising Antivenoms. The first step in treatment by antigen for treatment. passive immunization involves raising an antibody with Exposure to most venoms in humans does not result reactivity that is specific for the venom. Such an anti in protective immunity. Furthermore, all attempts to body is referred to as an antivenom. As noted above, create protective immunity against venoms with vac venoms pose unique problems for immunization. They cines have failed. F. E. Russell, JAMA 215:1994 (1971) are often expensive and available in only small amounts. (rattlesnake venom). By contrast, there has been success Furthermore, because they are toxic, they can do great creating protective immunity against individual toxins, damage before, and in some cases without, generating including diptheria (F. Audibert et al., Proc. Natl. an immune response. Acad. Sci USA 79:5042 (1982) and tetanus vaccines. J. 10 Usually the problem of a toxicity is approached by E. Alouf, Ann Inst. Pasteur/Microbiol. 136B, 309 modifying the venom in some manner. Modification of (1985). venoms, however, creates new problems. On the one hand, the modification may have so damaged the A. ACTIVE IMMUNIZATION venom that it is largely non-immunogenic. On the other Tetanus toxoid injections provide an effective protec 15 hand, while not rendered non-immunogenic, the modifi tion because they elicit a low level of circulating anti cation may have so altered the venom that a new antige body and establish immunological memory. When ex nicity is created. That is, antibody raised to the modified posed to a low dose of the tetanus organism and toxin, venom is directed to the modification as part of the the immunized animal can neutralize the organism and antigenic site. In this case, the antibody raised to the toxin before the infection develops. 20 modified venom may not react with the unmodified In the case of animal venoms, such prophylactic mea venom (as it will be found in its natural state). Finally, sures have not been feasible. First, many animal venoms the modification may itself be toxic or cause unexpected are too difficult or too expensive to obtain to immunize side effects. a population where a relatively small percentage of that Immunization with venoms is also complicated by population will be exposed to the animal venom. Sec 25 their complex composition. Venoms are remarkably ond, even if they can be obtained, animal venoms, unless heterogeneous. Furthermore, the various components detoxified, may cause more morbidity when adminis of venoms are present in different amounts. There is tered to a large population than would be caused by the some concern that immunization with whole venom venomous animals themselves. Third, even if the venom will not result in antibody reactive with all venom con is affordable, obtained in sufficient quantity, and detoxi 30 ponents. fied, it is extremely difficult to achieve the titer of circu ii. Administration. The second step in treatment by lating antibody necessary to neutralize the infusion of passive immunization (assuming, of course, the prob what can be a large amount of venom (up to one gram lems with the first step have been dealt with), involves of animal venom as compared with nanogram or pico the administering of antivenom to the host. The first gram amounts of tetanus toxin). Finally, even with suc 35 concernis whether the host will tolerate the administra cessful immunization, immunological memory is too tion of "foreign' antibody. In other words, will the slow to respond to the immediate crisis of envenoma host's immune system recognize the administered anti tion. body as antigen and mount an adverse response? Although active immunization with venoms has the Adverse host responses are typically of two types, above-named problems, some investigators have chosen 40 immediate and delayed. Immediate reactions are also of to pursue research in this area rather than in the area of two types: 1) anaphylaxis, and 2) Arthus reaction. Ana passive immunization, arguing that passive immuniza phylaxis is IgE mediated and requires sensitization to tion is too long and expensive. These investigators have antigen. The Arthus reaction is complement dependent made some progress in the method of immunization by and requires only antibody-antigen complexes. Both using liposomes. R. R. C. New et al., New Eng. J. Med. 45 immediate types of reactions are referred to as hyper 311 56 (1984). T. V. Freitas et al., Toxicon 27:341 sensitivity reactions; the host responds as if primed by a (1989). first exposure. Such immediate reactions can be acute. Indeed, anaphylaxis, if untreated, can lead to respira B. PASSIVE IMMUNIZATION tory failure and death. Because the problems with active immunization have 50 Delayed reactions are caused by a host primary in not been overcome, the only treatment available for mune response to the foreign proteins of the antivenom. venoms is passive immunization. Passive immunization, The reaction, called "serum sickness,' is characterized like active immunization, relies on antibodies binding to by fever, enlarged lymph glands, and joint pain. These antigens. For our purposes here, antitoxin refers to symptoms are apparent a number of days after passive antibody raised against a single toxin. Antivenom refers 55 immunization and gradually subside. to antibody raised against whole venom. The next concern about administering antivenoms is In the case of passive immunization, the antibody the dose. Without knowing the amount of venom in the used to bind the venom (antigen) is not made in the host it is difficult to know the amount of antivenom animal afflicted with the venom. Generally, an immune needed to treat the host. Furthermore, even if the response is generated in a first animal. The serum of the amount of venom can be estimated, how is the amount first animal is then administered to the afflicted animal of antivenom to be measured? Some approaches mea (the "host”) to supply a source of specific and reactive sure antivenom in units of volume. Such an approach antibody. The administered antibody functions to some does not account for different antivenom antibody con extent as though it were endogenous antibody, binding centrations within the same volume of serum. the venom toxins and reducing their toxicity. (It is not 65 iii. Commercial Antivenoms. Antivenoms have been known whether the antibody directly blocks the action raised in a number of mammals. See J. C. Perez et al., of venon toxins or merely carries venom toxins out of Toxicon 22:967 (1984) (mice). D. Iddon et al., Toxicon the blood stream.) 26:167 (1988) (mice). R. A. Martinez et al., Toxicon 5,196, 193 5 6 27:239 (1989) (mice). M. E. Ayeb and P. Delori, In: celoux, Medical Toxicology, Ch.39 (Elsevier Press Handbook of Natural Toxins, Vol.2, Insect Poisons, 1988). Allergens, and Other Invertebrate Venoms, (Anthony One of the most difficult aspects of clinical manage T. Tu, Ed.) (Marcel Dekker 1984), Chapter 18 (pp. ment of envenomation is the lack of standardization of 607-638) (rabbits). F. E. Russell et al., Toxicon 8:63 5 antivenoms. The recommended dosages of therapeutic (1970) (goats). S. C. Curry et al., J. Toxico).-Clin. horse-derived antivenoms is usually given in units of Toxicol. 21417 (1983-1984) (goats). F. Hassan, In: volume. For example, treatment with the Wyeth anti Handbook of Natural Toxins, Vol. 2, Insect Poisons, venom is measured in terms of vials of antivenom; each Allergens, and Other Invertebrate Venoms, (Anthony vial represents approximately 10 mls of antivenom in T. Tu, Ed.) (Marcel Dekker 1984), Chapter 17 (pp. 10 solution. D. C. Christopher and C. B. Rodning, S. Med. 577-605) (cows). Horses, however, are the animal of J.79:159 (1986). M. J. Ellenhorn and D. G. Barceloux, choice by an overwhelming number of investigators Medical Toxicology, Ch.39 (Elsevier Press 1988). H. and commercial antivenom producers. World Health M. Parrish and R. H. Hayes, Clin. Tox. 3:501 (1970). F. Organization Publication No. 58 (Geneva 1981). E. Russell et al., JAMA 233:341 (1975). Horses are sturdy and tolerant to the antibody-raising 5 process. Most importantly, they yield large volumes of The potency of individual lots of antivenoms will blood (as much as ten liters per bleeding for large ani vary because of two principal factors. First, because mals). whole antisera or immunoglobulin fractions are used There are significant disadvantages, however, when and the specific antibody titer per unit volume will vary using horses for antivenom production. First, for large 20 from animal to animal and from day to day, the amount production of antivenoms, horses more than 5 years old of venom-reactive antibodies will differ from prepara and usually less than 8 years old are required. Second, tion to preparation. Second, refinement procedures because new horses are easily killed or injured, produc such as ammonium sulfate precipitation and pepsin di tion should be under veterinary care and supervision. gestion can reduce the yield of active antibody, causing Third, tetanus is known to be a common disease among 25 variations in the titer of active ingredient per unit vol horses; animals must be immunized as soon as they are ume. These difficulties are exacerbated when anti introduced to the farm. F. Hassan, In: Handbook of venom is raised against a set of venoms in order to treat Natural Toxins, Vol. 2, Insect Poisons, Allergens, and a range of species. That is, when certain species are Other Invertebrate Venoms, (Anthony T. Tu, Ed.) more diverged from the immunizing group, it is more (Marcel Dekker 1984), Chapter 17 (pp. 577-605). 30 difficult to determine how much antivenom will be Fourth, large amounts of venom (antigen) are required required. for immunization in order to generate a satisfactory Because of the array of common and serious side immune response in horses. Fifth, horse antibody binds effects of unpurified antivenoms the physician must and activates human and other mammalian complement exercise caution not to give excessive amounts of horse pathways, leading (at the very least) to complement 35 product. Patients who receive seven or more vials of depletion and (at worst) to a more acute reaction by the the Wyeth preparation are reported to invariably de host. Most commercial antivenoms contain anticomple velop serum sickness; approximately 80% of patients mentary activity. S. K. Sutherland, Med J. Australia 1: overall who receive the preparation develop serum 613 (1977). Sixth, some humans are hypersensitive to sickness within three weeks. M. J. Ellenhorn and D. G. horse serum proteins and may react acutely to even 40 Barceloux, Medical Toxicology, Ch.39 (Elsevier Press very small amounts of horse protein. P. A. Christensen, 1988). In: Snake Venoms (Springer-Verlag 1979), Chapter 20 iv. Avoidino Side Effects. Because the commercial (pp. 825-846). antivenoms presently available can cause their own In spite of these problems, horse antivenom is the adverse reactions, the risk of possible death or serious only specific treatment of most venom poisonings 45 injury from the venom must be weighed against the risk known at the present time. It is considered vital for of a hypersensitivity reaction to horse serum. Before treating severe cases of snake envenomation. H. M. administration of horse serum, good medical practice Parrish and R. H. Hayes, Clin. Tox. 3:501 (1970). Simi requires that serum sensitivity tests be performed. H. M. larly, horse serum containing antivenoms is considered Parrish and R. H. Hayes, Clin. Tox. 3:501 (1970). life-saving in the treatment of scorpion stings. F. Has 50 san, In: Handbook of Natural Toxins, Vol. 2, Insect Serum sensitivity is typically performed by subcuta Poisons, Allergens, and Other Invertebrate Venoms, neously injecting a small amount of diluted serum in the (Anthony T. Tu, Ed.) (Marcel Dekker 1984), Chapter arm of the patient. A salt solution is injected in the other 17 (pp. 577-605). arm as a control. Normally, a positive hypersensitivity In the United States, the primary commercial pro 55 test is indicated by no more than formation of a welt on ducer of antivenom to snake venoms is Wyeth Labora the skin surface with surrounding swelling. Some pa tories (Marietta, Pennsylvania). To make a useful anti tients, however, develop anaphylactic shock, i.e. a full venom to members of the Crotalidae family, horses are hypersensitivity reaction. It is recommended in the immunized with a mixture of venon from four distinct medical literature that adrenalin be available for these species. To reduce their toxicity, the venoms are modi CSS. fied by treatment with formalin. To prolong their ab While sensitivity testing has its advantages, it is gen sorption, the modified venoms are mixed with alumi erally acknowledged that it has no predictive value for num hydroxide gel. H. M. Parrish and R. H. Hayes, serum sickness and reactions due to complement activa Clin. Tox. 3:501 (1970). Serum is collected and total tion. World Health Organization Publication No. 58 antibody is precipitated. During the collection process, 65 (Geneva 1981). Thus, all patients must be regarded as it is reported that the ammonium sulfate precipitation potential "reactors' and all drugs and equipment re destroys up to one half of the neutralizing antibodies of quired for dealing with reactions must be available be the crude antivenom. M. J. Ellenhorn and D. G. Bar fore antivenoms are administered. 5,196, 193 7 8 J. B. Sullivan's research group examined inmunoaf V. PURIFICATION finity purification with whole venoms. See J. B. Sulli One approach to avoiding side effects deserves spe van et al., J. Vet. Hum. Toxicol. 24:192 (Suppl.) (1982). cial note. It has been theorized that the high incidence J. B. Sullivan and F. E. Russell, Proc. Western Phar of side effects with current commercial horse antiven 5 macol. Soc. 25:185982. J. B. Sullivan and F.E. Rus oms is due to the bulk of irrelevant protein in these sell, Toxicon Suppl. 3:429 (1983). W. S. Jeter et al., preparations. (Protein other than specific antibody is Toxicon 21:729 (1983). D. Bar-Oretal, Clin. Tox. 22:1 considered to be irrelevant protein.) Under this theory, (1984). F. E. Russell et al., Am. J. Trop. Med. Hyg. the removal of irrelevant protein would reduce the 34:141 (1985). J. B. Sullivan, Ann. Emerg. Med. 16:938 burden of foreign protein and, thereby, reduce the inci 10 (1987). All of this work was performed with a poly dence of adverse immune responses. acrylamide resin and trapping as the means for associat F. Hassan, In: Handbook of Natural Toxins, Vol. 2, ing the venom with the resin. Insect Poisons, Allergens, and Other Invertebrate Ven Trapping involves suspending molecules in a gel. : oms, (Anthony T. Tu, Ed.) (Marcel Dekker 1984), Trapping does not involve attachment (covalent or Chapter 17 (pp. 577-605) attempted a crude purification 15 non-covalent) of the venom via a reactive group on the of horse antivenom. First, the horse serun was sub resin; without such an attachment, venom can find its jected to a mild pepsin digestion followed by ammo way through the matrix and end up in the eluate. Fur nium sulfate precipitation. Then, the precipitate was thermore, as venom from the antigen matrix finds its heat denatured; the heat-labile fraction was removed. way out of the suspension, there is a progressive reduc 20 tion in the antibody binding capacity of the antigen Unfortunately, approximately one-third of the initial matrix. Loss of binding capacity renders the matrix antivenom activity was reported to be lost by this non-recyclable, i.e. one cannot recover the same method. amount of purified antibody in subsequent purifications. A handful of antivenom investigators have consid Polyacrylamide has several drawbacks. First, poly ered immunoaffinity purification. However, most stud 25 acrylamide has low porosity and, hence, can sterically ies have only examined antibodies to a single toxin. C. hinder some antibody-antigen interactions, thereby re C. Yang et al., Toxicon 15, 51 (1977) attempted im ducing the antibody binding capacity of the polyacryla munoaffinity purification of antibody to a toxin in a mide-antigen matrix. A. Johnstone and R. Thorpe, Im snake venom. These investigators used cobrotoxin, a munochemistry in Practice, 2d Edition (Blackwell Sci neurotoxic crystalline protein isolated from the venom 30 entific Publications 1987), p. 209. Second, polyacryl of Taiwan cobra (Najanaja atra); whole venom was not amide itself is a neurotoxin; there is a concern that poly used. Cobratoxin attached to Sepharose (CNBr acrylamide may leech from the polyacrylamideantigen activated Sepharose 4B) was used as an antigen matrix matrix into the eluate and contaminate purified anti and formic acid was used to elute the toxin-specific body. antibodies. The immunoaffinity purified neutralizing 35 vi. Non-mammalian Sources of Antivenoms. As men capability than the unpurified antiserum. tioned above, most antivenoms are made in mammals V. Kukongviriyapan et al., J. Immunol. Meth. 49.97 and the overwhelming majority have been made in (1982) followed with a similar purification scheme. horses. There have been only a few attempts made at Again, whole venom was not used. These investigators raising antivenoms in non-mammals. A. Polson et al used Naja naja siamensis toxin 3, purified according to Immunol. Comm. 9:495 (1980), attempted to raise anti the method of E. Karlsson et al., Eur. J. Biochem. 21, 1 venoms against snake venoms in chickens. Their work (1971). A number of antigen matrices were studied, was unsuccessful; the chicken immunoglobulin showed including toxin-Sepharose (CNBr-activated Sepharose no protective activity against the venom in an assay 4B), toxin-succinylaminoethyl Sepharose, toxin-albu performed in mice. It was speculated that chicken anti min Sepharose, and toxin-succinylaminoethyl Biogel. 45 body interactions with venom are inherently weaker Horse antibody was used. Unfortunately, only approxi and less stable than those of horse antibody. mately 5% of the applied protein was reportedly bound and the destruction of antigenic sites on the immobilized SUMMARY OF THE INVENTION toxin occurred extensively. Most importantly, the toxin The present invention relates to antivenoms suitable neutralizing capacity recovered in the purified antibody for treatment of humans and animals as well as for ana represented only approximately one-third that of the lytical use. unpurified globulin. In one embodiment, the present invention contem M. E. Ayeb and P. Delori, In: Handbook of Natural plates a composition comprising polyvalent antivenom, Toxins, Vol. 2, Insect Poisons, Allergens, and Other comprised of immunoglobulin of which greater than Invertebrate Venoms, (Anthony T. Tu, Ed.) (Marcel 55 fifty percent is venom-reactive, and having two or more Dekker 1984), Chapter 18 (pp. 607-638) also followed monovalent subpopulations. The composition is prefer the Yang et al. procedure and applied it to purifying ably in an aqueous solution in therapeutic amounts and antibodies against individual scorpion neurotoxins. intravenously injectable. The polyvalent antivenom Again, whole venom was not used. These investigators preferably has reactivity to C. atrox, B. atrox. C. ada used toxin II of A. australis Hector. While these investi manteus and C. durissus terrificus venom. Preferably, gators did not report yields, they noted that formic acid one of the monovalent subpopulations comprises anti caused denaturation of the antibody. body with reactivity to C. durissus terrificus venom. The B. Lomonte et al., Toxicon 23:807 (1985), purified polyvalent antivenom preferably comprises horse anti antibodies against B. Asper myotoxin coupled to CNBr body. activated Sepharose 4B. The antimyotoxin was only 65 In another embodiment, the present invention con 0.5-1.0% of the antivenom protein and was found to be templates a composition comprising polyvalent anti less effective than crude antivenom in neutralizing the venom, comprised of immunoglobulin of which greater lethal effects of the venom. than fifty percent is venom-reactive, and having two or 5,196, 193 10 more monovalent subpopulations, wherein the polyva The present invention also contemplates a method for lent antivenom is derived from a first polyvalent anti purifying antivenom. In one embodiment, the purifying venom comprised of immunoglobulin of which less than method comprises a) providing, in any order: i) at least fifty percent is venom-reactive, and has substantially the one antivenom comprising immunoglobulin of which same spectrum of reactivity as said first polyvalent less than 50% is venom-reactive, ii) at least one antigen antivenom. matrix comprising at least one venom immobilized on In still another embodiment, the present invention an insoluble support, iii) at least one first and one second contemplates a composition comprising, polyvalent eluent; b) applying antivenom to the antigen matrix so antivenom, derived from a first polyvalent antivenom that greater than ninety percent of the venom-reactive comprised of immunoglobulin of which less than fifty O immunoglobulin binds the venom immobilized on the percent is venom-reactive, and having substantially the insoluble support of the antigen matrix; c) dissociating same spectrum of reactivity as the first polyvalent anti at least fifty percent of the bound venom-reactive in venom. The composition is preferably in an aqueous munoglobulin from the venom with the first eluent; and solution in therapeutic amounts and intravenously injec d) stripping the antigen matrix of substantially all of the table. The polyvalent antivenom preferably has reactiv 15 venom-reactive immunoglobulin with the second eluent ity to C. atrox. B. atrox, C. adamanteus and C. durissus so that the antigen matrix is recyclable. It is desirable terrificus venom. The polyvalent antivenom preferably that the antivenom comprises horse antivenom. Prefera comprises horse antibody. The composition preferrably bly, the antivenom comprises chicken antivenom and comprises polyvalent antivenom comprising two or the chicken antivenom comprises yolk immunoglobulin. more monovalent subpopulations. 20 The present invention also contemplates a method of In a preferred embodiment, the present invention analyzing antivenom. In one embodiment, the analyzing contemplates a composition comprising venom-neutral method comprises a) providing, in any order, i) a first izing avian antivenom. The composition is preferably in and a second immunizing venoms, ii) a solution com an aqueous solution in therapeutic amounts and intrave prising antivenom comprising immunoglobulin subpop nously injectable. Preferrably, the avian antivenom is 25 ulations having reactivity with the immunizing venoms, chicken antivenom and the chicken antivenom is con iii) a first antigen matrix comprised of the first immuniz prised of yolk immunoglobulin. It is desirable that the ing venom immobilized on an insoluble support, iv) a avian antivenom is comprised of protein comprised of second antigen matrix comprised of a second immuniz greater than 90% immunoglobulin and greater than ing venom immobilized on an insoluble support, and v) 50% venom-reactive immunoglobulin. Preferably, the 30 one or more eluents; b) applying the solution of anti avian antivenon is comprised of protein comprised of venom to the first antigen matrix so that greater than greater than 90% immunoglobulin and greater than 90% of the immunoglobulin reactive with the first in 99% venom-reactive immunoglobulin. It is desirable munizing venom is bound to the first antigen matrix and that the avian antivenom is polyvalent. Preferably, the so that the solution passes through the first antigen avian antivenom is high avidity chicken antivenom. 35 matrix to create a first flow-through; c) applying the The present invention also contemplates an antigen eluent to the first antigen matrix so that at least fifty matrix useful for purification of antivenom. In one em percent of the bound immunoglobulin reactive with the bodiment, the antigen matrix comprises C. durissus ter first immunizing venom is dissociated to create a first rificus venom attached to an insoluble support. In one eluate; and d) applying, in any order, i) the first flow embodiment, the antigen matrix comprises a plurality of 40 through to the second antigen matrix, followed by the venoms attached to an insoluble support. Preferably, eluent to create a second eluate, ii) the first eluate to the the attachment is covalent attachment. It is desirable second antigen matrix so that the solution passes that the insoluble support prior to attachment to the through the second antigen matrix to create a second plurality of venoms comprise a resin having aldehyde flow-through, followed by the eluent to create a third groups. Preferably, the plurality of venoms comprises 45 eluate. In one embodiment, the method further con C. atrox, B. atrox, C adamanteus and C. durissus ter prises, after step d), comparing the venom-reactivity of rificus venom. the first flow-through, the first eluate, the second flow The present invention also contemplates a method for through, the second eluate, and the third eluate, with immobilizing whole venom. In one embodiment, the the venom reactivity of the antivenom. It is desirable method comprises a) providing an insoluble support; b) 50 that the antivenom comprises horse antibody. Prefera providing two or more whole venoms; and c) attaching bly, the antivenom comprises chicken antibody and the two or more whole venoms to the insoluble support. chicken antibody comprises yolk immunoglobulin. It is Preferably, the attaching is covalent attaching. It is desirable that, after applying the eluent, the first and desirable that the insoluble support comprises a resin second antigen matrices are rendered recyclable. A comprising aldehyde-activated agarose. 55 desirable recycling eluent is guanidine. In another embodiment, the method for immobilizing The present invention also contemplates a method of whole venom comprises a) providing an insoluble sup treatment. In one embodiment, the treatment comprises: port; b) providing a single whole venom; and c) attach a) providing i) avian antivenom in an aqueous solution ing the single whole venom to the insoluble support by in therapeutic amounts that is intravenously injectable, covalent binding. 60 ii) at least one envenomed subject b) intravenously in The present invention also contemplates a method of jecting the avian antivenom into the subject. Preferably, producing antivenom. In one embodiment, the produc the avian antivenom is chicken antivenom and the ing method comprises a) providing one or more immu chicken antivenom comprises yolk immunoglobulin. It nizing venoms; b) providing at least one avian species; is desirable that the avian antivenom comprises immu and c) immunizing the avian species with one or more 65 noglobulin of which greater than fifty percent is venom immunizing venoms, so that a neutralizing antivenom is reactive. Preferably, the subject is a mammal. produced. Preferably, the avian species comprises In an alternative embodiment, the present invention chickens. contemplates a method of treatment, comprising: a) 5,196, 193 11 12 providing i) polyvalent antivenom in an aqueous solu FIG. 9 shows an immunoaffinity purification profile tion in therapeutic amounts that is intravenously injec for a preferred embodiment of the method of the pres table, comprising immunoglobulin of which greater ent invention. than fifty percent is venom-reactive, and having two or FIG. 10 shows SDS-PAGE analysis of chicken and more monovalent subpopulations, ii) at least one enve horse antivenoms. nomed subject; b) intravenously injecting the polyva FIGS. 11(A)-11(C) show the reactivity by ELISA of lent antivenom into the subject. It is desirable that the mammalian antivenom purified using one embodiment polyvalent antivenom is horse antivenom. of the method of the present invention. It is not intended that the present invention be limited FIG. 12 shows the reactivity by ELISA of avian by the source of the venom used for immunizing, purify 10 antivenom purified using one embodiment of the ing or analyzing. Similarly, it is not intended that the method of the present invention. present invention be limited by the source of the venom FIG. 13 compares the reactivity by ELISA of two for which the antivenom compositions of the present antivenom preparations, purified using two different invention are reactive. For example, the present inven embodiments of the method of the present invention. tion contemplates venoms selected from the group con 15 FIG. 14 compares the reactivity by ELISA of two sisting of Chordata, Arthropoda and Coelenterata ven antivenom preparations, purified using two different ons. Venoms selected from the group consisting of embodiments of the method of the present invention. snake venoms, spider venoms, scorpion venoms or jelly FIGS, 15(A)-15(C) show the spectrum of reactivity fish venoms are specifically contemplated. The present by Western Blot of antivenoms before and after in invention also contemplates venom selected from the 20 munoaffinity purification. group consisting of Crotalus scutulatus, Notechis scuta FIG.16 shows the spectrum of reactivity by Western tus, Acanthophis antarcticus, Oxyuranus scutellatus, Blot of immunoaffinity purified antivenoms with immu Pseudonaja textilis, Pseudechis australis, Enhydrina schis nizing and non-immunizing venoms. tosa, Ophiophaous hannah, Vipera ammodytes, Vipera 25 DESCRIPTION OF THE INVENTION aspis, Vipera berus, Vipera xanthina palestinae, Vipera The present invention is directed to antivenoms and lebetina, Cerastes cerastes. Cerastes vipera, Bitis arietans, method for making antivenoms. The properties of anti Bitisgabonica, Vipera russelli, Echis carinatus, Trimeresu venom of the present invention make the antivenoms rus flavoviridis, Agkistrodon halys, A. piscivorus, A. contor multi-purpose; antivenoms prepared according to the trix, Naja naja, Naja n. haje, Naja n, kaouthia, Naja n. 30 present invention are useful for analytical studies in oxiana, Najan, sputatrix, Najan. atra, Naja nivea, Naja vitro and useful as therapeutic agents. nigrocollis, Hemachatus hennachatus, Dendroaspis angus The present invention contemplates I) producing ticeos, Dendroaspis jamesoni, Dendroaspis polylepis, Den antivenoms in non-mammals, and II) increasing the droaspis viridis, Bungarus caerulus, Bungarus fasciatus, effectiveness of both non-mammalian antivenoms and Bungarus multicinctus, Agkistrodon rhodostoma, Agkis 35 mammalian antivenoms, however they might have been trodon acutus, Bothrops atrox, Bothrops jararaca, Bothrops produced. The present invention further contemplates jararacussu, Bothrops alternatus, Lachesis muta, Micrurus III) treating humans and animals by in vivo administra corralus, Micrurus fulvius, Micrurus frontalis, Micrurus tion of antivenoms. A preferred embodiment of the niagrocinctus, Leiurus quinouestriatus, Tityus serrulatus, method of the present invention is shown in FIG. 1 Centruroides suffusus, Centruroides noxius, Centruroides illustrating the temporal relationship of the method sculpturatus, Androctonus australis, Buthotus judaicus, steps. The individual steps are described separately Buthus tamalus, Latrodectus mactans, Latrodectus hes below. perus, Loxosceles reclusa, and Chironex fleckeri venom. Preferably, the antivenoms of the present invention I. Obtaining Antivenoms in Non-Mammals react with C. atrox, B. atrox, C adamanteus and C. 45 A preferred embodiment of the method of the present durissus terrificus venoms. invention for obtaining antivenoms involves immuniza DESCRIPTION OF THE ORAWINGS tion. However, it is also contemplated that antivenoms could be obtained from non-mammals without immuni FIG. 1 is a schematic, showing a preferred embodi zation. In the case where no immunization is contem ment of the method of the present invention. plated, the present invention may use non-mammals FIG. 2 is a schematic, showing one approach to with preexisting antibodies to toxins and/or venoms as venom epitope determinations of the present invention. well as non-mammals that have antibodies to toxins FIG. 3 is a schematic, showing a preferred approach and/or venons by virtue of reactions with the adminis to venom epitope determinations of the present inven tered (non-venom) antigen. An example of the latter tion. 55 involves immunization with synthetic peptides or re FIG. 4 is a schematic, showing a preferred approach combinant proteins sharing epitopes with venom com to antivenom immunoaffinity purification of the present ponents. invention. In a preferred embodiment, the method of the present FIG. 5 is a Western Blot, showing the reactivity of invention contemplates immunizing non-mammals with antivenoms raised against modified venoms. 60 whole venom(s). It is not intended that the present in FIG. 6 shows the reactivity by ELISA of antivenom vention be limited to any particular venom. Venom raised in different chickens against snake venom. from all venomous sources (see Table 1) are contem FIG. 7 shows the reactivity by ELISA of two prepa plated as immunogens. rations of antivenom raised in one chicken over three When immunization is used, the preferred non-man hundred days apart. 65 mal is from the class Aves. All birds are contemplated FIG. 8 shows the increase in elution efficiency ob (e.g. duck, ostrich, emu, turkey, etc.). A preferred bird served with increased residence time of a non-denatur is a chicken. Importantly, chicken antibody does not fix ing eluent. mammalian complement. See H. N. Benson et al., J. 5,196, 193 13 14 Immunol. 87:610 (1961). Thus, chicken antibody will (native) at approximately two week intervals up to normally not cause a complement dependent reaction. approximately one hundred days. The preferred collec A. A. Benedict and K. Yamaga, In: Comparative Immu tion time is sometime after day 100. This preferred in nology (J. J. Marchaloni, Ed.), Ch. 13, Immunoglobu munization schedule results in the production of high lins and Antibody Production in Avian Species quantities of reactive chicken antibody (i.e. reactive (pp.335-375) (Blackwell, Oxford 1966). Thus, the pre with the components of the immunized venom(s)) per ferred antivenoms of the present invention will not ml of egg yolk (i.e. "high titers"). Furthermore, this exhibit complement-related side effects observed with preferred immunization schedule results in the produc antivenoms known presently. tion of antivenoms with "high avidity.” High avidity is When birds are used, it is contemplated that the anti 10 defined as antibody reactivity with multiple epitopes on venom will be obtained from either the bird serum or individual venom components as measured by the for the egg. A preferred embodiment involves collection of mation of precipitin lines in Ouchterlony immunodiffu the antivenom from the egg. Laying hens export immu sion gels at salt concentrations less than 1.5M NaCl. noglobulin to the egg yolk ("IgY") in concentrations Antivenoms requiring salt concentrations of 1.5M NaCl equal to or exceeding that found in serum. See. R. Pat 15 or greater to form precipitin lines are "low avidity" terson et al., J. Immunol. 89:272 (1962). S. B. Carroll antivenoms. While not limited to any precise mecha and B. D. Stollar, J. Biol. Chem. 258:24 (1983). In addi nism, high avidity antivenoms have a greater probabil tion, the large volume of egg yolk produced vastly ity of neutralizing venom components in vivo. exceeds the volume of serum that can be safely obtained Where birds are used and collection of antivenom is from the bird over any given time period. Finally, the 20 performed by collecting eggs, the eggs may be stored antibody from eggs is purer and more homogeneous; prior to processing for antibody. It is preferred that there is far less non-immunogobulin protein (as com storage of the eggs be performed at 4' C. for less than pared to serum) and only one class of immunoglobulin is one year. transported to the yolk. It is contemplated that chicken antibody produced in It has been noted above that, when considering im 25 this manner can be buffer-extracted and used analyti munization with venoms, one may consider modifica cally. While unpurified, this preparation can serve as a tion of the venon to reduce its toxicity. In this regard, reference for activity of the antibody prior to further it is not intended that the present invention be limited by manipulations (e.g. immunoaffinity purification). immunization with modified venom. Unmodified ("na II. INCREASING THE EFFECTIVENESS OF tive') venom is also contemplated as an immunogen. 30 It is also not intended that the present invention be ANTIVENOMS limited by the type of modification--if modification is When purification is used, the present invention con used. The present invention contemplates all types of templates purifying to increase the effectiveness of both venom modification, including chemical and heat treat non-mammalian antivenoms and mammalian antiven ment of the venom. The preferred modification, how 35 oms. Specifically, the present invention contemplates ever, is heat-inactivation. increasing the percent of venom-reactive immunoglob It is not intended that the present invention be limited ulin. When evaluated for immunoglobulin content, pu to a particular mode of immunization; the present inven rity and reactivity, at different stages of purification, tion contemplates all modes of immunization, including preferred antivenoms of the present invention have the subcutaneous, intramuscular, intraperitoneal, and intra 40 following relationship: less than 50% of the immuno vascular injection. globulin of the polyvalent antivenom prior to purifica The present invention further contemplates immuni tion (i.e. of the "first polyvalent antivenom'') is venom zation with or without adjuvant. (Adjuvant is defined as reactive; greater than 50% of the immunoglobulin of a substance known to increase the immune response to the polyvalent antivenom after purification is venom other antigens when administered with other antigens.) 45 reactive. If adjuvant is used, it is not intended that the present While all types of purification (e.g. purification based invention be limited to any particular type of adjuvan on size, charge, solubility, etc.) may be used, the pre t-or that the same adjuvant, once used, be used all the ferred purification approach for mammalian antibody is time. While the present invention contemplates all types immunoaffinity purification. The preferred purification of adjuvant, whether used separately or in combina 50 approaches for avian antibody are: A) Polyethylene tions, the preferred use of adjuvant is the use of Com Glycol (PEG) separation, and B) Immunoaffinity purifi plete Freund's Adjuvant followed sometime later with cation. Incomplete Freund's Adjuvant. When immunization is used, the present invention A. PEG PURIFICATION contemplates a wide variety of immunization schedules. 55 The present invention contemplates that avian anti In one embodiment, a chicken is administered venom(s) venom be initially purified using simple, inexpensive on day zero and subsequently receives venom(s) in procedures. In one embodiment, chicken antibody from intervals thereafter. It is not intended that the present eggs is purified by extraction and precipitation with invention be limited by the particular invervals or polyethylene glycol (PEG). PEG purification exploits doses. Similarly, it is not intended that the present in 60 the differential solubility of lipids (which are abundant vention be limited to any particular schedule for col in egg yolks) and yolk proteins in high concentrations of lecting antibody. However, a preferred schedule for polyethylene glycol 8000. Polson et al., Immunol. immunization of the present invention is the administra Comm.9:495 (1980). The technique is rapid, simple, and tion of a cocktail of (heat-inactivated) venoms on day relatively inexpensive and yields an immunoglobulin zero at 1 mg/ml for each venom, with subsequent ad 65 fraction that is significantly purer in terms of contami ministrations of the same cocktail (heat-inactivated or nating non-immunoglobulin proteins than the compara native) at the same dose on days 14 and 21, and with ble ammonium sulfate fractions of mammalian sera and gradually increasing doses ("boosts') up to 10 mg/ml horse antivenoms. Indeed, PEG-purified antibody is 5,196, 193 15 16 sufficiently pure that the present invention contern parative information is available on the forces affecting plates the use of PEG-purified antivenoms in the pas the stability of chicken antibody-antigen complexes. sive immunization of envenomed humans and animals. Indeed, very little structural or chemical data has been compiled on the properties of chicken immunoglobulins B. IMMUNOAFFINITY PURIFICATION (e.g., amino acid sequence of heavy and light chains, As noted, immunoaffinity purification is the preferred disulfide bond structures, three-dimensional structure, purification approach for both mammalian and avian freeze-thaw and thermal stability, etc.). antivenom. Immunoaffinity purification is separation The majority of immunoaffinity purification of anti based on the affinity of antibody for specific antigen(s); venom that has been done has only used individual antibody that binds to specific antigens) is separated 10 toxins and not whole venom. No comparative informa from antibody that does not bind (under the conditions tion is available on the forces affecting the stability of used). The present invention contemplates the use of the whole venom antigen matrix. immunoaffinity purification to dramatically reduce the The immunoaffinity purification of the present inven foreign protein burden of antivenoms by elimination of tion involves consideration of the venom(s) used to raise irrelevant protein (non-immunoglobulin and non-anti 15 the antivenom that is to be immunoaffinity purified. In gen-binding immunoglobulin) when the antivenom is this regard, the present invention provides a method for used therapeutically. While not limited to any specific evaluating the immunochemical similarity or dissimilar theory, it is contemplated that a reduction in the protein ity of venoms that allows for i) means of designing burden will be accompanied by a reduction in side ef. cost-effective immunization cocktails for new anti fects associated with passive immunization of foreign 20 venom formulas, ii) means of designing cost-effective protein. antigen matrices for purifying new or existing antiven The present invention contemplates immunoaffinity oms, iii) means of identifying the monovalent and poly purification by use of an "antigen matrix' comprised of valent antibody subpopulations of an existing anti venom(s) attached to an insoluble support. Antibody to venom, and iv) means of determining its spectrum of be purified is applied in solution to the antigen matrix. 25 reactivity of antivenom for the further design of anti The solution passes through the antigen matrix and venoms for treatment. Additional considerations in comprises the "flow through." Antibody that does not clude the nature of the resin, method of binding the bind, if present, passes with the solution through the venom(s) to the resin, method of applying the unpuri antigen matrix into the flow through. To eliminate all fied antivenoms, and method of recovering purified non-binding antibody, the matrix is "washed' with one 30 antivenom. or more wash solutions which, after passing through the matrix, comprise one or more "effluents." "Eluent' is a i. Venoms Used To Immunize chemical solution capable of dissociating antibody Every geographic area has its own unique collection bound to the antigen matrix (if any) that passes through of venomous species. The greater the immunochemical the antigen matrix and comprises an "eluate.' Antibody 35 similarity of the venoms produced by these species, the that is dissociated (if any) is freed from the antigen greater the likelihood hat antivenoms raised against matrix and passes by elution with the eluent into the one species will react with and neutralize other species. eluate. The more dissimilar the venoms, the greater the need is In one embodiment, the material for the insoluble for antivenoms that react with a number of species so support (hereinafter "resin') takes the form of spherical that individual species need not be identified in the case beads. In a preferred embodiment, the resin is a syn of an emergency. thetic polymer capable of forming a gel in aqueous Immunochemical similarity is better understood in media (e.g. agarose). terms of epitopes. An epitope is defined as an antibody The immunoaffinity purification of the present inven combining site on an antigen. Where venom is the anti tion provides a number of benefits. First, the immunoaf 45 gen, an epitope is a discrete (typically measured in ang finity purification of the present invention provides for stroms) region of a venom component where antibody maximum attachment of the antigen (e.g. venom) to the binds to the venom component. Where an epitope is not resin, i.e. high attachment efficiency. Second, the im present in the venom of other species, it is referred to as munoaffinity purification of the present invention pro a "species-unique epitopes." Where an epitope is com vides for the recovery of as much of the reactive anti mon to venom of different species it is referred to as a body of the unpurified antibody (the preferred unpuri "species-shared epitope.” Where there is a single spe fied antibody is PEG-purified whole yolklgY) as possi cies-unique epitope in a venom, there is said to be immu ble, i.e. the quantity of antibody purified is optimized. nochemical dissimilarity between the venom and any Third, the immunoaffinity purification of the present other venom. The greater the number of species-unique invention allows for the recovery of the antibody in an 55 epitopes the greater the immunochemical dissimilarity. active state, i.e. the quality of reactivity is preserved. Because of the practicality of treatment within one Fourth, the immunoaffinity purification of the present geographic area, the present invention contemplates invention provides that the bound antibody be eluted raising antivenoms with a mixture of venoms (a "cock quantitatively; there is no significant (less than two tail') as an immunogen. Using cocktails, the antivenoms percent) retained antibody to progressively decrease of the present invention have reactivity with more than column capacity after successive cycles of use, i.e. the one venom. Furthermore, using the immunoaffinity antigen matrix is recyclable. Fifth, and most impor purification of the present invention, the immunoaf. tantly, the immunoaffinity purification of the present finity purified antivenoms of the present invention re invention allows for the retention in the purified anti tain this spectrum of reactivity. This is in contrast to venom of the spectrum of reactivity of the unpurified 65 existing antivenoms. Previously, the purification of antivenom. antivenoms raised using mixtures of venoms has not Most previous studies of polyclonal antibody purifi preserved the spectrum of reactivity of the unpurified cation have involved mammalian antibodies. No com antivenom. 5,196, 193 17 18 The present invention contemplates that antivenom is venom antigen matrices are not capable of binding and a "population' of antibodies. The population may be purifying the spectrum of antivenom antibodies present composed of one or more "subpopulations." Where in the polyvalent commercial antivenom investigated. more than one whole venom is used as an immunogen Thus, purification in the manner described by these (and assuming the venoms are immunogenic), the result researchers necessarily resulted in antivenom with a ing antivenom is "polyvalent." A polyvalent antivenom more limited reactivity than the unpurified antivenom. is herein defined as a population of antibodies having To retain the spectrum of reactivity, the present in reactivity with all of the immunizing venoms. Where vention gives consideration to the venom(s) used to only one whole venom is used as an immunogen, the immunize when determining the appropriate venoms resulting antivenom is "monovalent.' O for immunoaffinity purification. In a preferred embodi A polyvalent antivenom may have "monovalent sub ment, the immunoaffinity purification of the present populations," "polyvalent subpopulations,' and/or invention contemplates using the same venom or cock "crossreactive subpopulations.' In all cases, valency is tail of venoms for purification as was used for immuni defined by venom reactivity with immunizing venom(s) zation. By using the same venom or cocktail of venoms, (and not by antibody class or subclass). Monovalent 15 the purified antivenom derived from unpurified anti subpopulations of a polyvalent antivenom are herein venom has substantially the same spectrum of reactivity characterized as subpopulations exhibiting reactivity as the unpurified antivenom, where "substantially" is with some but not all immunizing venoms. Polyvalent defined as greater than 50% of the antigen-binding subpopulations of a polyvalent antivenom are herein reactivity of the unpurified antivenom with respect to characterized as subpopulations exhibiting reactivity 20 each immunizing venom. As a relative value, the mea with all of the immunizing venoms. Crossreactive sub surement of "50% reactivity' need not be made by any populations of a polyvalent antivenom are herein char particular assay. Noetheless, conventional direct anti acterized as subpopulations exhibiting reactivity with gen-binding ELISA techniques are preferred. In other non-immunizing venom(s). embodiments, the present invention contemplates using Monovalent antivenoms may have "monovalent sub 25 different venom or cocktails for purification as was used populations' and "crossreactive subpopulations.” for immunization. Monovalent subpopulations of a monovalent antivenom are herein characterized as subpopulations exhibiting ii. Nature of the Resin reactivity with the immunizing venom. Crossreactive The present invention contemplates immobilization subpopulations of a monovalent antivenom are herein 30 of venoms with an insoluble support. There are essen characterized as subpopulations exhibiting reactivity tially three ways this can be achieved. First, venom with non-immunizing venom(s). components can be physically trapped in a gel; this Subpopulation characterization is best understood by approach does not rely upon any particular chemical example and by ignoring reactivity with non-immuniz reactivities of the venom components. Second, venom ing venoms. Where a first and a second whole venom 35 components can be covalently coupled to an "acti are used together to immunize, the resulting polyvalent vated' matrix; this approach relies on the existence of antivenom can in theory have i) a monovalent subpopu functional groups on the venom components that can lation of antibody reactive only with the first venom, ii) covalently bond with the matrix. Third, venom compo a monovalent subpopulation of antibody reactive only nents can be coupled to insoluble supports using bifunc with the second venom, and iii) a polyvalent subpopula tional reagents as linking groups that react with a) side tion of antibody (i.e. antibody reactive with both ven groups on venom components and b) groups on the oms). On the other hand, the resulting polyvalent anti insoluble support. venom could also comprise two monovalent subpopula As noted earlier, only individual venom components tions without any polyvalent subpopulation, or a poly have thus far been covalently attached to a matrix; valent subpopulation without any monovalent subpopu 45 whole venom has not been immobilized in this manner. lations. - With single venom components, a number of factors, Whether in fact the resulting polyvalent antivenom including mode of attachment, the type of resin, and the does have monovalent subpopulations depends on distance between the resin and the attached antigen, can whether the venoms used as immunogen have species influence the success of the immobilization approach. unique epitopes. Where there are no species-unique 50 See e.g., V. Kukongviriyapan et al., J. Immunol. Meth. epitopes, there will be no monovalent subpopulations. 49.97 (1982). With whole venoms these factors are mul Polyvalent antivenom can be made either by a) in tiplied and result in greater uncertainty. munizing with a venom cocktail or b) immunizing with One important issue in immobilizing whole venom is single venoms and mixing two or more monovalent coupling efficiency: are the functional groups accessible antivenoms. In either case, the reactivity of the sub 55 such that the bulk of the venom components are cou population(s) determine the spectrum of reactivity of pled? Another issue involves antigenicity: does attach the population, i.e. the antivenom. Importantly, ment to the matrix preserve (or destroy) the antigenicity whether monovalent or polyvalent, the purification of of the venom components? With respect to the first the present invention allows for the quantitative reten issue, little is known about the chemical structure of tion in the purified antivenom of the spectrum of reac 60 most venom components. While the existence of reac tivity of the unpurified antivenom. tive primary amines should allow for attachment to In nearly all previous studies, antitoxin and anti most activated matrices, there is no assurance that this venom antibodies were purified using only a single will occur at a sufficient level to be useful. With regard toxin. Antibodies purified in this manner do not have to the second issue, it is possible that the functional the spectrum of reactivity required to neutralize the 65 groups involved in covalent attachment are part of, or plurality of distinct toxins present in whole venoms. In near, important epitopes. Random alteration or steric the handful of studies using whole venom, the investiga hindrance of epitopes through covalent bonding of tors used only single venom antigen matrices. Single functional groups could dramatically influence the anti 5,196, 193 19 20 body binding capacity of the matrix. Non-random alter By using sequential elution steps, the purification of the ation or steric hindrance of epitopes could be expected present invention overcomes the problem of poor re to significantly impact the ability to recover immunoaf. cyclability that is associated with traditional purifica finity purified antivenom with the spectrum of reactiv tion methods. ity of the unpurified antivenom. In the preferred embodiment, the purification of the In a preferred embodiment, the present invention present invention allows for quantitative purification of provides a covalent attachment method for whole the venom-specific antibodies present initially in the egg venom that allows for high coupling efficiency and high yolk IgY, retention of antibody antigen-binding activ antibody binding capacity. It is preferred because of the ity, and recyclability of the matrix, by a first elution ease and efficiency of antigen attachment, the stability O with a non-denaturing eluent and a second elution with of the attachment (relative to resins involving non denaturing eluent. The second elution recovers the covalent attachment), and mechanical and chemical remaining portion of antibody and recycles the antigen strength towards denaturants that are used during chro matrix. Thus, following application of crude antivenom, matographic procedures. a preferred embodiment of the method of the present In one embodiment, the covalent attachment method 15 invention comprises the steps: 1) washing the matrix employs cyanogen bromide activated Sepharose 4B with a first buffer, 2) washing the matrix with a second (Pharmacia) as a resin for covalent attachment of whole buffer, 3) washing the matrix with a third buffer, 4) venom. The preferred resins with active groups for dissociating specific antibody from the venom with a covalent attachment are resins with aldehydes as active first eluent, 5) washing the matrix again with the third groups ("aldehyde-activated resins'), such as are de 20 buffer, 6) stripping the column of substantially all non scribed in U.S. Pat. No. 3,836,433 to Wirth et al., which venom protein with a second eluent, 7) re-equilabrating is hereby incorporated by reference. A preferred resin is the column for recyclability with the first buffer, and 8) aldehyde-activated agarose. Chicken antibody eluted removing eluent from the dissociated, specific antibody, from such an antigen matrix exhibits a consistently wherein the first buffer is a phosphate-containing, low higher venom binding activity (i.e. quality) than 25 ionic strength, non-detergent-containing, neutral pH chicken antibody eluted from other antigen matrices. buffer, the second buffer is a high ionic strength, non ionic detergent-containing, slightly alkaline buffer, the iii. Binding Venom(s) To The Resin third buffer is a non-phosphate-containing, low ionic It is not intended that the immunoaffinity purification strength, non-detergent-containing, neutral pH buffer, of the present invention be limited to any particular 30 the preferred first eluent is a non-denaturing solution venom or mixture of venoms. In the preferred embodi compatable with an extended residence time to dissoci ment, however, a cocktail is used. The preferred bind ate bound antibody, the preferred second eluent is a ing of the venom, whether single whole venoms or strongly denaturing solution that rapidly dissociates the cocktails of whole venom, to an aldehyde-activated antibody that was not released by the first eluent, and resin is via sodium cyanoborohydride reduction. 35 the removal of both eluents from the dissociated anti In a preferred embodiment, the antigen matrix is body is performed by extensive dialysis against a low "pre-stripped' with an eluent prior to any further use of ionic strength, non-detergent-containing, neutral pH the antigen matrix. By pre-stripping, the purification of buffer. the present invention avoids contamination of im While not limited to any particular theory, it is be munoaffinity purified antivenom preparations with 40 lieved that the first buffer washes the bulk of the un venon that failed to attach to the resin. bound protein from the matrix. The second buffer is of high ionic strength in order to wash non-venom-reac iv. Applying Unpurified Antivenon tive, electrostatically-bound proteins from the matrix, It was noted above that by using the same venom or and contains a non-ionic detergent to wash non-venom cocktail of venoms in the antigen matrix as was used in 45 reactive, hydrophobically-bound proteins from the ma immunization, the purified antivenom derived from trix. Neither the high ionic strength nor the non-ionic unpurified antivenom has substantially the same spec detergent should disrupt specific antibody-antigen in trum of reactivity of the unpurified antivenom. The teractions. The third buffer does not contain phosphate maximum retention of the spectrum of reactivity is because phosphate is not compatible with the preferred achieved when one, in addition using the same venom SO first eluent. After the first eluent, it is highly desirable or cocktail of venoms in the antigen matrix as was used that the matrix be equilibrated with the third buffer to in immunization, uses the venom or cocktail of venoms wash the remaining eluent from the matrix before the in the antigen matrix at a concentration that allows for second eluent is applied. After the second eluent, the the presentation of antigen in greater amounts than that phosphate-containing first buffer is used to remove the needed to bind all of the specific antibody applied to the 55 second eluent and to re-equilibrate the matrix in prepa antigen matrix. The latter is achieved by monitoring the ration for a new purification cycle. flow through for reactivity as the unpurified antibody is In a preferred embodiment, the present invention applied; where the flow through shows less than 10% of contemplates 4M guanidine-HCl pH 8.0 as a second the reactivity of the unpurified antibody that is applied, eluent to dissociate the antigen-antibody complexes the venom or cocktail of venoms in the antigen matrix 60 formed between the immobilized venom and the is viewed to be at a concentration that allows for the chicken antibody specific for the antigen, because of its presentation of antigen in greater amounts than that efficiency, solubility at 4 C and ease of removal by needed to bind all of the specific antibody applied to the dialysis. antigen matrix. It was determined that 0.1M glycine pH 2.5 as a first 65 eluent caused unsatisfactory qualitative effects (high v. Recovering Purified Antivenon background sticking to solid surfaces and aggregation In the preferred embodiment, the purification of the of renatured antibody) on chicken antibodies. Similarly, present invention contemplates sequential elution steps. 4M guanidine-HCl, when used as a first eluent, eluted 5,196, 193 21 22 material that possessed only 17-49% of the original unique and/or species-shared epitopes. In this example, activity of the unpurified chicken IgY. one antivenom is used; the antivenom was raised by immunization with Venom A together with Venom B. vi. Assessment of the Spectrum of Reactivity Again, for ease of understanding, FIG. 3 has been As noted above, whether in fact the resulting polyva drawn to show species-unique epitopes (dark squares lent antivenom does have monovalent subpopulations and circles for Venom A and open squares and circles depends on whether the venoms used as immunogen for Venom B) as well as species-shared epitopes (dark have species-unique epitopes. The present invention triangles). contemplates determining whether a species' venom has Antivenom AB is immunoaffinity purified sequen species-unique and/or species-shared epitopes (relative 10 tially over the immunizing venoms used individually as to other species' venom). The present invention con antigen matrices (Matrix A, Matrix B and Matrix B"). templates making this determination using both i) anti The flow-throughs in all cases are reapplied (dotted body raised against single venoms (i.e. monovalent anti lines) to assure quantitative recovery of specific anti venoms), and ii) antibody raised against a plurality of body. The flow-through from Matrix A (Flow venoms (i.e. polyvalent antivenoms). In this manner, the 15 antivenoms of the present invention are useful in the Through 1) is then applied to Matrix B. After washing selection of the appropriate venom and cocktails as both Matrix A and Matrix B, the bound antibody is immunogens for the production of the preferred poly eluted. The existence of antibody (if present) in each valent antibodies for treatment. eluate (Eluate I is from Matrix A; Eluate 2 is from Ma FIG. 2 shows schematically one approach to deter 20 trix B) is detected by A280. mining whether a species' venom has species-unique Eluate 1 is applied to Matrix B' (Matrix B can be the and/or species-shared epitopes. In this example, two same as Matrix B' or can be a duplicate). The flow antivenoms are used: one raised by immunization with through (Flow-Through 2) is collected for character Venom A and the other raised by immunization with ization. The existence of non-binding antibody (if pres Venom B. Depending on the source of the antivenoms, 25 ent) in the flow-through is determined by A280. After a prestep (not shown) may be desirable to eliminate washing Matrix B", the bound antibody (if any) is eluted viscosity and hydrophobicity problems during purifica and collected for characterization. The existence of tion (e.g. chicken antivenom derived from eggs should antibody (if present) in this eluate (Eluate 3) is detected be PEG-treated in a prestep procedure to eliminate by A280. lipids). For ease of understanding, FIG. 2 has been 30 Where antibody is detected in Eluate 2, this antibody drawn to show species-unique epitopes (dark squares must be non-reactive with Venom A, but reactive with and circles for Venom A and open squares and circles Venom B. This indicates there are one or more species for Venom B) as well as species-shared epitopes (dark unique epitopes and consequently one or more monova triangles). lent subpopulations. Where antibody is detected in Elu Antivenom A and Antivenon B are immunoaffinity 35 ate 3, this antibody must be reactive with both Venom purified separately over their respective immunizing A and Venom B. This indicates there are one or more venom immobilized to make an antigen matrix (Matrix species-shared epitopes and consequently one or more A and Matrix B). The flow-through in each case is polyvalent subpopulations. reapplied to assure quantitative recovery and then dis A comparison of FIG. 2 with FIG. 3 shows that the carded (dotted lines). After washing, the bound anti use of a cocktail immunogen (FIG. 3) allows for the body is eluted. The existence of antibody (if present) is same determinations with fewer "runs' (i.e. elutions detected in each eluate (Eluate 1 is from Matrix A: from antigen matrices). This advantage continues and, Eluate 2 is from Matrix B) by ultraviolet light absorp indeed, becomes more significant when epitope deter tion at 280 nm ("A280') minations are desired for greater numbers of venoms. The eluates in each case are then applied to the re 45 For example, where three venoms need to be analyzed, spective non-immunizing venom as an antigen matrix; a cocktail immunogen approach allows for epitope de Eluate 1 is applied to Matrix B and Eluate 2 is applied terminations with only seven runs, while a single venom to Matrix A" (Matrix A can be the same as Matrix A" or immunogen approach requires twelve runs. can be a duplicate; Matrix B can be the same as Matrix B' or can be a duplicate). Again, the flow-through in 50 Once epitope determinations are made for the ven each case is reapplied to assure quantitative recovery oms of interest, a preferred cocktail matrix can be de (dotted lines). The final flow-throughs are collected for signed for immunoaffinity purification such that the characterization The existence of non-binding antibody purified antivenom retains the spectrum of reactivity of in the flow-through is determined by A280. After wash the unpurified antivenom. While a cocktail matrix con ing, the bound antibody (if any) is eluted and collected 55 taining all of the immunizing venoms will (if used in a for characterization. The existence of antibody (if pres quantitative protocol) invariable retain the spectrum of ent) is detected in each eluate (Eluate 3 is from Matrix reactivity (see FIG. 4), elimination of venoms having no B'; Eluate 4 is from Matrix A") by A280. species-unique epitopes may be desired where large Where antibody is detected in the final flow scale (e.g. commercial) purifications are to be per throughs, this antibody must be non-crossreactive. This 60 formed with scarce and/or expensive venom(s). indicates there are one or more species-unique epitopes III. Treatment and consequently one or more monovalent subpopula tions. Where antibody is detected in the final eluates, The present invention contemplates antivenom ther this antibody must be crossreactive. This indicates there apy for envenomed humans and animals. The method of are one or more species-shared epitopes and conse 65 antivenom treatment of the present invention involves quently one or more crossreactive subpopulations. consideration of a) venom identification, b) degree of FIG. 3 shows schematically a preferred approach to envenomation, and c) type and dose of antivenom to be determining whether a species' venom has species administered. 5,196, 193 23 24 In another embodiment, the treatment of the present A. Venom Identification invention contemplates the use of PEG-purified anti Commonly, the venomous species is not seen, let venom from birds. The use of yolk-derived, PEG-puri alone captured for identification, at the time of enveno fied antibody as antivenon allows for the administra mation. The lack of reliable species identification, par tion of 1) non(mammalian)-complement-fixing, avian ticularly in emergency situations, taken together with antibody, 2) a less heterogeneous mixture of non the cost of raising antivenom, makes it preferrable that immunoglobulin proteins, and 3) only one-third as the antivenoms used in treatment not be limited in their much total protein to deliver the equivalent weight of reactivity to a single species. Thus, in a preferred en active antibody present in currently available anti bodiment, the present invention contemplates raising 10 venom. This means that approximately three times polyvalent antivenoms according to the potential for more venom-reactive antibody can be administered envenomation by venomous inhabitants of any particu before the risk of serum sickness reaches that of the lar geographical area. currently available antivenom. The non-mammalian source of the antivenom makes it useful for treating B. Degree of Envenomation 5 patients that are sensitive to horse or other mammalian Not all venoms are potentially fatal; even bites or serums. PEG-purified antivenom is useful for treating victims where the amount of antivenom required is stings from the most potent species may not be life relatively small or the cost of affinity purification is threatening if a relatively low degree of envenomation prohibitive. The amount of antivenom required may be occurs. A key clinical dilemma, however, results from relatively small (<250mg) when the extent of systemic the fact that the amount of venom delivered is highly envenomation is slight. For instance, low envenomation variable and the attending medical personnel must rely may occur in the case of certain scorpion species, very on the victim's symptoms in assessing the extent of the small or immature snakes, or snake species that typically overall threat of serious injury or death. These symp deliver small (<5 mg) of venom in a single bite. Cost is tons include, but are not limited to, local pain, hemor 25 prohibitive in certain regions of the world; such areas rhaging, numbness, edema, necrosis, nausea, vomiting, cannot sustain the financial burden of immunoaffinity blood clotting abnormalities, faintness, proteinuria, res reagents and the skilled labor necessary to produce piratory distress and paralysis. Importantly, the severity highly-purified antivenoms. of these symptoms must be considered in connection In a preferred embodiment, the treatment of the pres with the time after envenomation. Typically, symptoms 30 ent invention uses yolk-derived, immunoaffinity puri are more severe over time. Thus, less severe symptons fied antibody as antivenom, which allows for the admin early on do not ensure a low level of envenomation. istration of 1) non-complement-fixing antibody, and 2) The qualitative nature of these symptoms and the fre only 1/20th as much total protein to deliver the equiva quent difficulty in establishing a meaningful time frame lent weight of active antibody present in currently make a determination of the degree of envenomation 35 available antivenom. This means that twenty times approximate at best. more venom-reactive antibody can be administered before the risk of serum sickness reaches that of the C. Dosage of Antivenom currently available antivenom. Thus, physicians may It was noted by way of background that a balance treat victims more aggressively with far greater must be struck when administering currently available amounts of active antivenom without fear of increased antivenom; sufficient antivenom must be administered side effects. Because the initial foreign protein exposure to neutralize the venom, but not so much antivenom as is reduced, the risk of sensitizing a patient to antivenon to increase the risk of untoward side effects. These side is also reduced. This is important where the patient effects are caused by i) patient sensitivity to horse prote must undergo subsequent antivenom treatment. ins, ii) anaphylactic or immunogenic properties of non 45 immunoglobulin proteins, iii) the complement fixing EXPERIMENTAL properties of mammalian antibodies, and/or iv) the The following examples serve to illustrate certain overall burden of foreign protein administered. It is preferred embodiments and aspects of the present in extremely difficult to strike this balance when, as noted vention and are not to be construed as limiting the scope above, the degree of envenomation (and hence the level thereof. of antivenom therapy needed) can only be approxi In the disclosure which follows, the following abbre mated. viations apply: A280 (Absorbance at 280 nm); eq (equiv The present invention contemplates significantly re alents); M (Molar); M (micromolar); N (Normal); mol ducing side effects so that this balance is more easily (moles); mmol (millimoles); umol (micromoles); nmol achieved. Treatment according to the present invention 55 (nanomoles); gm (grams); Ing (milligrams); ug (micro contemplates reducing side effects by using i) in grams); L (liters); ml (milliliters); ul (microliters); 'C. munoaffinity purified antivenom from mammalian (degrees Centigrade); CFA (Complete Freund's Adju sources, ii) PEG-purified antivenom from birds, and/or vant); IFA (Incomplete Freund's Adjuvant); ELISA iii) immunoaffinity purified antivenom from birds (Enzyme-linked Immunosorbent Assay); MW (molecu In one embodiment, the treatment of the present in lar weight); OD (optical density); EDTA ethylene vention contemplates the use of immunoaffinity purified diaminetetracetic acid); PAGE (polyacrylamide gel antivenom from mammalian sources. While comple electrophoresis); Aldrich (Aldrich Chemical Co., Mil ment-fixing, immunoaffinity purification of antivenon waukee, Wis.); Beckman (Beckman Instruments, San from mammalian sources reduces the total protein bur Ramon, Calif); BRL (Bethesda Research Laboratories, den up to approximately twenty fold. This means that 65 Gaithersburg, Md.); Cappel (Cappel Laboratories, Mal approximately twenty times more venon-reactive anti vern, Pa.); Eastman (Eastman Kodak, Rochester, N.Y.); body can be administered before the risk of serum sick Fisher (Fisher Biotech, Springfield, N.J.); GIBCO ness reaches that of the currently available antivenom. (GIBCO, Grand Island, N.Y.); Gilford (Gilford, Ober 5,196, 193 25 26 lin, Ohio); IBF (IBF Biotechnics, Savage, Md.); Mal ished all protease activity and was, therefore, the treat linckrodt (Mallinckrodt, St. Louis, Mo.); Pierce (Pierce ment used to prepare heat-treated immunogen. Chemical Co., Rockford, Ill.); Sigma (Sigma Chemical b) immunization: Two, one-year old white leghorn Co., St. Louis, Mo.); Wyeth (Wyeth Laboratories, Mar hens were each immunized with 1 mg of Crotalus atrox ietta, Pa.). venom after it was treated according to one of the inac For convenience when discussing antivenoms, the tivation methods described above (total of six immu immunized animal used as the source is used as a modi nized hens). The hens were injected with the inacti fier in front of the term "antivenom' (e.g. "horse anti vated venom in CFA (GIBCO) on day zero subcutane venom' means antivenom raised in a horse and ously in multiple sites (both sides of the abdomen, both "chicken antivenom' means antivenom raised in a 10 breasts, and in both wings to involve more of the lym chicken). phatic system). The hens were subsequently injected The antivenom used as starting material and/or con with 1 mg of the same inactivated venom in IFA trol antivenom in some examples below (hereinafter (GIBCO) on Day 11 and Day 19. "unpurified horse antivenom') was obtained from c) antivenom collection: Chicken immunoglobulin Wyeth (lot #M878035) This unpurified horse anti 15 (Igy) was extracted according to a modification of the venom has been used extensively by others for the treat methods of A. Polson et al., Immunol. Comm. 9:495 ment of humans afflicted with venomous bites. D. C. (1980). A gentle stream of distilled water from a squirt Christopher and C. B. Rodning, S. Med. J. 79:159 bottle was used to separate the yolks from the whites, (1986). M. J. Ellenhorn and D. G. Barceloux, Medical and the yolks were broken by dropping them through a Toxicology, Ch.39 (Elsevier Press 1988). H. M. Parrish 20 funnel into a graduated cylinder. The broken yolks and R. H. Hayes, Clin. Tox. 3:501 (1970). F. E. Russell were blended with 7 volumes of egg extraction buffer to et al., JAMA 233:341 (1975). improve antibody yield (Egg extraction buffer =0.01M Sodium phosphate, 0.1M NaCl, pH 7.5, containing EXAMPLE 1. 0.01% NaN3), and PEG 8000 (Baker) was added to a Production of an Antivenom to a Cocktail of Modified 25 concentration of 3.5%. When all the PEG dissolved, Snake Venoms in a Non-Mammal the protein precipitate that formed was pelleted by To determine the best course for raising high titer egg centrifugation at 9000xg for 10 minutes. The superna antibodies against venoms, the effect of various meth tant was decanted and filtered through cheesecloth to ods of venom modification was examined. In order to 30 remove the lipid layer, and PEG was added to a final demonstrate that, as a result of modification, toxicity concentration of 12% (we assumed the supernatant was would be inactivated but antigenicity would be pre 3.5% PEG). The solution was centrifuged as above, the served. The example involved a) venom modification, supernatant discarded, the pellet redissolved in 2 times b) immunization, c) antivenom collection, and d) antige the original yolk volume of egg extraction buffer, and nicity assessment. 35 PEG added to 12% for a second precipitation. After a) Venom modification: Crotalus atrox venom centrifugation the supernatant was discarded and the (Sigma) was modified by formaldehyde, glutaraldehyde pellet centrifuged twice more to extrude the PEG. This or heat treatment. In order to monitor the inactivation crude IgYpellet was then dissolved in the original yolk of venom, the inhibition of total venom protease activ volume of egg extraction buffer and stored at 4 C. ity was measured according to a modified method of V. 40 d) antigenicity assessment: Eggs were collected dur B. Philpot, Jr. et al., Toxicon 16, 603 (1978). The assay ing the day 25-day 28 period (and stored intact at 4 C. consisted of mixing 10 microliters of venom or buffer for approximately six months) to assess whether the control with 25 microliters of azo-coupled cowhide venoms were sufficiently antigenic to raise an antibody. beads (Sigma) in 200 microliters of PBS for five to Eggs from the two hens in each modification group fifteen minutes. Protease activity is detected when the 45 were pooled and antibody was collected as described supernatant turns red due to hydrolysis of the azo dye. above. Antigenicity was assessed on Western Blots Results were quantitated at A525 on a spectrophotome according to the method of H. Towbinet al. Proc. Nat. ter (Gilford). Acad. Sci. USA 76:4350 (1979). 100 ug samples of three With respect to inactivation, it is not necessary that distinct venom types (Crotalus adamanteus, Crotalus complete inhibition be achieved. It is simply desirable as 50 atrox, and Agkistrodon pisciworus) were dissolved in SDS a means of minimizing the impact of immunization in reducing sample buffer (1% SDS, 0.5% 2-mercaptoe the immunized animal. thanol, 50 mM Tris pH 6.8, 10% glycerol, 0.025% w/v For formaldehyde treatment, a 10 mg/ml dilution of bromophenol blue), heated at 95" C. for 10 minutes and whole C. Atrox venom was made in various concentra separated on a 1 mm thick, 12% SDS-polyacrylamide tions of formaldehyde (0.25-8.0% w/v) and left for 1 55 gel. K. Weber and M. Osborn, In: The Proteins, 3rd hour at room temperature. Because it was observed that Edition (H. Neurath and R. L. Hill, Eds.) (Academic 8.0% formaldehyde gave complete inhibition of venom Press, NY) (pp. 179-223). Part of the gel was cut off and protease activity, these conditions were used to prepare the proteins stained in Coomassie Blue. The proteins in the formaldehyde-treated immunogen. the remainder of the gel were transferred to nitrocellu For glutaraldehyde treatment, a 10 mg/ml dilution of 60 lose using the ABN polyblot electro-blotting system whole C. Atrox venom was made in glutaraldehyde and according to the manufacturers instructions (Fisher). left for 1 hour at room temperature. Because it was The nitrocellulose was temporarily stained with 10% observed that 1% glutaraldehyde totally inactivated Ponceau S (S. B. Carroll and A. Laughion, In: DNA Venom protease activity, this concentration was used to Cloning: A Practical Approach, Vol. III, D. Glover, prepare the formaldehyde-treated immunogen. 65 Ed., IRL Press, Oxford, pp.89-111)) to visualize the For heat treatment, a 10 mg/ml dilution of whole C. lanes, then destained by running a gentle stream of Atrox was heated to 95 C. in a water bath for five distilled water over the blot for several minutes. The minutes and then plunged into ice. This treatment abol nitrocellulose was immersed in PBS containing 3% 5,196, 193 27 28 BSA overnight at 4.C to block any remaining protein inactivation was as described in Example 1) in PBS binding sites. were mixed with one volume of RibiLES-STM adju The blot was cut into strips and each strip was incu vant (Ribi ImmunoChem Research Inc., Hamilton, bated with the appropriate primary antibody. The three Mont) at 37° C. and vortexed to a milky emulsion be primary antibodies discussed above were used (along fore injection. with pre-immune chicken antibody as a control) diluted To prepare the Freund's adjuvant/venom antigen 1:250 in PBS containing 1 mg/ml BSA for 2h at room mixture, heat-inactivated venom was mixed in with temperature. The blots were washed with 2 changes CFA (GIBCO) in a relationship of 5:4 (adjuvant:antigen each of large volumes of PBS, BBS-Tween and PBS by volume) and emulsified to a firm consistency by successively (10 min/wash). Goat anti-chicken IgG O passage through an antigen mixer made from two 18 alkaline phosphatase conjugated secondary antibody gauge stainless steel hypodermic needles that had been (Fisher Biotech) was diluted 1:400 in PBS containing 1 brazed together. ng/ml BSA and incubated with the blot for 2 hours at To prepare the bentonite adjuvant/venom antigen room temperature. The blots were washed with 2 mixture, one volume of native venom was mized with changes each of large volumes of PBS and BBS-Tween, 15 followed by 1 change of PBS and 0.1M Tris-HCl, pH one volume of a sterile, 2% (w/v) bentonite (Sigma) 9.5. Blots were developed in freshly prepared alkaline suspension to adsorb the venom proteins to the particu phosphatase substrate buffer: 100 g/ml Nitro-Blue late. Tetrazolium (Sigma), 50 g/ml 5-Bromo-4-Chloro-3- b) immunization: Six, (previously unimmunized) one Indolyl Phosphate (Sigma), and 5 mM MgCl2 in 50 mM year old white leghornhens (numbered for reference as Na2CO3 pH 9.5. #337, #339, #340, #353, #354, and #355) were immu The results are shown in FIG. 5. The Coomassie Blue nized on Day zero. Two birds (#339, #354) received strip (Strip 1) illustrates the order of the venoms put in the Ribi adjuvant/antigen mixture. Two other birds all the other strips: Crotalus adamanteus was placed in (#337, #353) received the antigen with CFA. The re Lane 1; Crotalus atrox, was placed in Lane 2; Agkistro 25 maining two birds (#340, #355) received the antigen don piscivorus was placed in Lane 3. From FIG. 5, anti absorbed to bentonite. The hens were injected subcuta body reactivity is seen in Strips 2 (formaldehyde-treated neously in multiple sites (both sides of the abdomen, immunogen) and 4 (heat-treated immunogen); very both breasts, and in both wings to involve more of the little reactivity can be seen in Strips 3 (glutaraldehyde lymphatic system). treated immunogen) and 5 (no immunogen). This sug 30 All of the birds were re-injected in the same manner gests that, while the glutaraldehyde treatment was use with the same amount of antigen (prepared in the same ful to reduce protease activity, the treatment denatured way for each two bird group) on Days 14 and 21, with the venom to the point where antigenicity was also the exception of birds #337 and #353, which received severely reduced. In terms of antigenicity, it appears antigen in IFA. from FIG. 5 that there is the following relationship 35 c) antibody collection: Antibody was extracted from among the different treatments: heat-treateddfor the eggs as described in Example 1. Importantly, the maldehyde-treated) glutaraldehyde-treated. method of extracting antibody from the eggs resulted in Importantly, the chicken antibody recovered from approximately 80% recovery of initial antibody recov the eggs reacts with many protein bands in Lane 2 (the ery according to the following assay: C. atrox venom preparation) of Strips 2 and 4. The Yolks were blended with seven volumes of egg ex presence of many protein bands illustrates the complex traction buffer (0.01 MNaphophate, 0.1M NaCl, pH 7.5 ity of the various venoms. Some of these protein bands containing 0.01% NaN3). Then a small volume of di from the different venoms appear to co-migrate, sug luted yolk was further diluted seven-fold in egg extrac gesting that there might be proteins in common among tion buffer, centrifuged at 9000xg for 10 minutes and the venoms. Interestingly, while the antivenom was 45 the supernatant assayed for antibody activity by ELISA raised against a single venom (C. atrox), the results from see d) below). PEG-purified material was compared Lanes 1 (the Crotalus adamanteus preparation) and 3 with the crude yolk sample for antivenom activity. (the Agkistrodon piscivorus preparation) of Strips 2 and 4 Eggs were collected from bird #355 beyond Day 28 indicate that the chicken antibody reacts with non for later use; PEG-purified antibody from #355 eggs immunizing venoms, i.e. is crossreactive. Clearly, there collected from days 31-45 is referred to as "PEG-puri is a set of antigenically related proteins in the three fied 355” in Examples 13, 16, 25, and 27, below. different venoms. d) antigenicity assessment: The impact of the differ EXAMPLE 2. ent adjuvants on the antigenicity of the venom was assessed on Day 28 eggs (stored intact at 4' C. until they Adjuvant Effects on Antivenom Titers 55 were assayed on Day 40) by ELISA. To prepare for the To determine the best course for raising high titer egg ELISA, 96-well Nunc Immuno-Plates were coated antibodies against venoms, the impact of different types overnight at 4 C. in a humidified chamber with 200 of adjuvants was demonstrated. The example consisted ul/well of the appropriate venom (in this case C. atrox) of a) adjuvant/antigen mixture preparation, b) immuni at a concentration of 2 g/ml. The next day the wells zation, c) antibody collection, and d) antigenicity assess were blocked with PBS containing 0.1% bovine serum ent. albumin (BSA) for 2 hours at room temperature. To a) adjuvant/antigen preparation: In all cases the anti perform the ELISA, appropriately diluted antibody gen consisted of 1 mg of each of 4 snake venoms (4 mg was added in PBS containing 0.1% BSA and the plates of total antigen per bird): Agkistrodon contortrix, Agkis were incubated for 2 hours at room temperature. The trodon piscivorus, Crotalus atrox, and Crotalus adaman 65 plates were then washed three times with BBS (0.1M teus (Sigma). boric acid, 0.025M sodium borate, i.M. NaCl, pH 8.3) To prepare the Ribi adjuvant/venom antigen mix containing 0.1% Tween 20, twice with PBS containing ture, three volumes of the heat-inactivated antigen (heat 0.1% Tween, and twice with just PBS. Alkaline phos 5,196, 193 29 30 phatase-conjugated rabbit anti-chick IgG (Fisher) was diluted 1:500 in PBS containing 0.1% BSA, added to EXAMPLE 4 the plates, and incubated 2 hours at room temperature. Response of a Non-Mammal to High Doses of a The plates were washed as before, except Tris-buffered Cocktail of Native Venoms saline, pH 7.2, was substituted for PBS in the last wash, To optimize the response and increase the titer of and p-nitrophenyl phosphate (Sigma) was added at 1 antibody, further immunization was demonstrated. In mg/ml in 0.05M Na2CO3 pH 9.5, 10 mM MgCl2. The this example, bird #354 was used exclusively. As in plates were then evaluated either qualitatively by visual Example 3, the example involved a) adjuvant/antigen examination or quantitatively by reading at 410 nm on a mixture, b) immunization, c) antivenom collection, and Dynatech MR300Micro ELISA reader approximately O d) antibody titer assessment. 30 minutes after the substrate was added. a) adjuvant/antigen mixture: As in Example 3, this The ELISA results from the Day 28 eggs showed example involved the use of immunized birds (contrast Examples 1 and 2). Therefore, the venoms were not good reactivity for all the birds (data not shown). How modified and were used in their native form. The first ever, no clear difference between the three adjuvants 15 venom mixture consisted of 0.5 mg each of Crotalus was apparent when evaluated qualitatively. Nonethe atrox and Crotalus adamanteus and 0.5 mg of B. atrox less, bentonite did cause a palpable abcess in one bird, (Sigma). The second venom mixture consisted of 2.5 mg consistent with its reported tendency to do so. P. A. each of Crotalus atrox and Crotalus adamanteus (Sigma). Christensen, In: Snake Venoms (Springer-Verlag 1979), The third venom mixture consisted of 10 mg each of Chapter 20 (pp. 825-846). Bentonite also caused a de 20 Crotalus atrox and Crotalus adamanteus (Sigma). The crease in the laying frequency of this bird. first, second and third venom mixtures were mixed In view of the cost of the RIBI mixture and the side separately with IFA. The three adjuvant/antigen mix effects of the bentonite, the fact that Freund's adjuvant tures were mixed and emulsified as in Example 3. works just as well makes Freund's adjuvant a preferred b) immunization: Bird #354 was immunized on Day adjuvant. 25 72 with the first adjuvant/antigen mixture, on Day 86 with the second adjuvant/antigen mixture, and on Day EXAMPLE 3 106 with the third adjuvant/antigen mixture. In all Booster Immunizations with a Cocktail of Native cases, the injections were made subcutaneously in multi Venoms ple sites. 30 c) antibody collection: Antibody was extracted from To optimize the response and increase the titer of yolks (as described in Example 1) of Day 58-60 eggs antibody, further immunization was demonstrated. The (hereinafter "PEG-purified Pool 1"; PEG-purified Pool example involved a) adjuvant/antigen mixture, b) im 1 is used in Example 18, below), Day 74-81 eggs (here munization, c) antivenom collection, and d) antibody inafter "PEG-purified Pool 2"); PEG-purified Pool 2 is titer assessment. 35 used in Examples 16, 17, 18 and 19 below), Day 94-9- a) adjuvant/antigen mixture: since this example in 9eggs (hereinafter “PEG-purified Pool 3"; PEG-puri volved the use of immunized birds (contrast Examples 1 fied Pool 3 is used in Examples 18 and 28 below) and and 2), the venoms were not modified and were used in Day 120-126 eggs (hereinafter "PEG-purified Pool 4'; their native form. The venom mixture consisted of 0.5 PEG-purified Pool 4 is used in Examples 5, 14, 18, 23 mg each of Crotalus atrox and Crotalus adamanteus and and 26). 0.25 mg of B. atrox (Sigma). IFA was mixed with the d) antibody titer assessment: The impact of the Day venon mixture in a 5:4 volume ratio (adjuvant:antigen) 72 boost of native venom (including an increased dose and emulsified to a firm consistency by passage through of B. atrox) on antibody titer was assessed on Day 81 using Pool 2 (the eggs were stored intact at 4' C. until an antigen mixer made from two 18 guage stainless steel 45 they were extracted and assayed on Day 81) by ELISA hypodermic needles that had been brazed together. (see Example 2 for general discription of ELISA). The b) immunization: The six one-year old white leghorn results indicated a continuing increase in antibody titer. hens of Example 2 (#337, #339, .340, #353, #354, and Importantly, the previously immunized birds toler #355) were immunized on Day 49. All the birds re ated 20 mg of active native venom (3 times the dose ceived the same adjuvant/antigen mixture. As before, required to kill an adult human on a body weight basis) the hens were injected subcutaneously in multiple sites. with no apparent ill effects. This demonstrates that far c) antibody collection: Antibody was collected from greater immunizing doses (mg/kg) can be used than the eggs as described in Example 1. have been used in immunization schedules in the past. d) antibody titer assessment: The impact of the Day These higher doses allow for higher antivenom titers. 49boost of native venom (including B. atrox for the first 55 time) on antibody titer was assessed on Day 62 using EXAMPLE 5 Day 57-61 eggs (stored intact at 4' C. until they were Duration of the High Titer Response to Venoms in a assayed on Day 62) by ELISA as described in Example Non-Mammal 2. 60 To optimize the duration of the response and increase The results are shown in FIG. 6. Chicken #354 the productive period of the laying hen, further immu clearly had the highest titer as measured on Day 60 nization was demonstrated. In this example, bird #354 following the Day 49 boost. Interestingly, the bird with was used exclusively. As in Example 4, the example the lowest titer, #339, was previously immunized in the involved a) adjuvant/antigen mixture, b) immunization, same manner as #354, suggesting that other factors may 65 c) antivenom collection, and d) antibody titer assess be involved in generating high titers than immune sta ent. tus. Importantly, all the birds show a significant titer as a) adjuvant/antigen mixture: As in Example 4, this compared with the unimmunized control. example involved the use of immunized birds. There 5,196, 193 31 32 fore, the venoms were not modified and were used in 90-95%. Thus almost 10 mg of venom protein was their native form. The first venom mixture consisted of bound per ml of resin. 2.5 mg. each of Crotalus atrox and Crotalus adamanteus For later use, the antigen matrix was suspended in an (Sigma); the second venom mixture consisted of 5.0 mg. equal volume of 1M ethanolamine-10mM Tris-HCl (pH each of C. atrox and C. adamanteus: the third mixtures 8.5) for 2 hours at 4' C. to block remaining protein-reac consisted of 10 mg. C. atrox and 5 mg. C adamanteus: tive sites. The antigen matrix was then washed with the fourth mixture consisted of 10 Ing. C. atrox and 5 PBS containing 0.02% sodium azide and stored at 4 C. mg. C. adamanteus: the fifth mixture consisted of 15 mg. C. atrox and 10 mg. C adamanteus. All five venom EXAMPLE 7 mixtures were mixed separately with IFA and the ad O Covalent Attachment of Whole Snake Venom to an juvant/antigen mixtures emulsified as in Example 3. Aldehyde-Activated Polyacrylamide/Agarose Matrix b) immunization: Bird #354, which had been last immunized on day 106 (see Example 4), was immunized In this example, the coupling efficiency of the alde on day 300 with the first adjuvant/antigen mixture, on hyde-activated, polyacrylamide/agarose resin, Ultrogel day 328 with the second adjuvant/antigen mixture, on 15 AcA 22 (IBF) (hereinafter "activated resin II'), was day 356 with the third adjuvant/antigen mixture, on demonstrated. C. atrox venom was diluted in PBS (pH day 372 with the fourth adjuvant mixture, and on day 7.2) at a concentration of 10 mg/ml. Activated Resin II 407 with the fifth adjuvant/antigen mixture. In all cases, was washed with 10 volumes of distilled H2O and then the injections were made simultaneously in multiple with 2.5 volumes of 0.5MNaPO4 (pH 7.0). Resin II was sites. 20 then added in an equal volume to the venom solution. c) antivenom collection: Antivenon antibody was The mixture (hereinafter "antigen matrix') was split extracted from the eggs as described in Example 1. into two equal volumes. One was agitated for 18 hours d) antibody titer assessment: The impact of these five at 4' C. The other was agitated for 18 hours at room boosts of native venom was assessed on day 422 using temperature. Both antigen matrix solutions were antibody from day 412-415 eggs ("day 422 prep” indi 25 washed with PBS on a glass funnel. The filtrates were cates the eggs were stored intact at 4.C until antibody collected and coupling efficiency was calculated as in was PEG-purified and assayed on day 422) and com Example 6. The antigen matrix agitated at 4 C. showed pared in an ELISA (see Example 2) with PEG-purified 52% coupling yield and the antigen matrix agitated a Pool 4. room temperature showed 62% coupling yield. Thus, It can be seen (FIG. 7) that the response of the bird is 30 5-6 mg of venom protein per ml of matrix was coupled comparable with both preparations (in terms of reactive using activated Resin II. antivenom per ml). Thus, a bird, that has not been im For later use, protein reactive sites were blocked in munized for almost two hundred days, can be re-immu 1M ethanolamine-10mM Tris-HCl (pH 8.0) at 4' C. for nized with venom in a second immunization program 3 hours. The antigen matrix was then washed and stored such that the response is equivalent to the response 35 in PBS containing 0.02%. observed after the initial immunization program. EXAMPLE 8 Clearly, birds are useful for more than one year for Covalent Attachment of Whole Snake Venom to an antivenom production. Aldehyde-Activated Agarose Matrix EXAMPLE 6 In this example, the coupling efficiency of the alde Covalent Attachment of Whole Snake Venom to hyde-activated resin, ACTIGELA (Sterogene) (here Cyanogen Bromide-Activated Agarose Matrix inafter "activated resin III'), was demonstrated. C. In this example, the coupling efficiency of Sepharose atrox venom was dissolved in PBS (pH 7.2) at a concen 4B (Pharmacia) (hereinafter "resin I'), was demon 45 tration of 10 mg/ml. Activated Resin III was washed strated using snake venon as antigen. C. atrox venom with 3 volumes of PBS and added (in equal volume) to was diluted in PBS (pH 7.2) at a concentration of 10 the venom solution. Thereafter, 1/10 volume of 1M mg/ml. In a chemical hood, Resin I was washed with 5 sodium cyanoborohydride (Aldrich) was added. The volumes of chilled distilled H2O, suspended in an equal mixture (hereinafter "antigen matrix') was then split volume of 2.5M potassium phosphate buffer (pH 12.2) into two equal volumes. One was agitated for 4 hours at in a beaker immersed in an ice bath, and stirred gently. room temperature. The other was agitated overnight at In a separate vessel in the same chemical hood, 1 gram 4' C. Both antigen matrix mixtures were washed on of CNBr (Alrich) was dissolved in 1 ml of acetonitrile glass funnels with PBS. The filtrate was collected and per 10 ml of gel to be coupled. Thereafter, the CNBr coupling efficiency was calculated as in Example 6. The solution was added to the gently stirring solution of 55 results showed that coupling efficiency of activated resin I over a period of two minutes. The mixture (here Resin III is in the range of 80 to 90%. The antigen inafter "activated resin I') was continually stirred for matrix was stored in PBS containing 0.02% sodium an additional eight minutes and then washed in a scint azide at 4 C. ered glass funnel with 10 volumes of cold distilled H2O EXAMPLE 9 followed by 10 volumes of cold PBS. The venom solu tion was then added to the activated resin in the funnel Covalent Attachment of Whole Snake Venom from a and the mixture (hereinafter "antigen matrix') was agi Second Species to an Aldehyde-Activated Agarose tated overnight. The uncoupled filtrate was collect Matrix from the funnel and measured (A280) Coupling effi In this example, the coupling efficiency of the alde ciency was calculated as the amount of coupled protein hyde-activated resin, ACTIGELA (Sterogene) (here (A280 units) divided by the total starting amount of inafter "activated resin III'), was studied with another protein (A280). The results showed that the coupling venom. C. durissus terrificus venom was dissolved in efficiency of activated resin I was in the range of PBS (pH 7.2) at a concentration of 5 mg/ml. Activated 5,196, 193 33 34 Resin III was washed with 3 volumes of PBS and added this stripped antibody is collected and measured (A280), (in equal volume) to the venom solution. Thereafter, the antigen matrix is washed and the same amount of 1/10 volume of 1M sodium cyanoborohydride (Ald PEG-purified, chicken antibody (diluted in egg extrac rich) was added. The mixture (hereinafter "antigen matrix') was agitated overnight at 4 C. The antigen tion buffer) is loaded on the antigen matrix at a flow rate matrix was then washed on a glass funnel with PBS. of 1 ml per min, and washed successively with several The filtrate was collected and coupling efficiency was bed volumes of PBS, BBS-Tween (0.1M boric acid, calculated as in Example 6. The results showed that 0.025M sodium borate, 1M NaCl, 0.1% (v/v) Tween 20 coupling efficiency of activated Resin III was 95%. The pH 8.3), and PBS until the effluent is free of protein antigen matrix was stored in PBS containing 0.02% 10 (A280) sodium azide at 4 C. In the first elution, bound chicken antibody is eluted immediately with 4M Guanidine-HCl (pH 8.0) and the EXAMPLE 10 antigen matrix is washed with PBS. The eluate is col Covalent Attachment of Whole Snake Venom from a lected and measured (A280). The elution efficiency of Third Species to an Aldehyde-Activated Agarose 5 4M guanidine-HCl is functionally defined as 100%; Matrix there is no antibody remaining on the column that can In this example, the coupling efficiency of the alde be further eluted with 4M guanidine-HCl, Elution effi hyde-activated resin, ACTIGELA (Sterogene) (here ciencies calculated for other eluents (see below) are inafter "activated resin III'), was studied with another venom. C adamanteus venon was dissolved in PBS 20 relative efficiencies using 4M guanidine-HCl as 100%. (pH 7.2) at a concentration of 10 mg/ml. Activated The same antigen matrix is reacted with the same Resin III was washed with 3 volumes of PBS and added amount of PEG-purified chicken antibody and washed (in equal volume) to the venom solution. Thereafter, as discussed above. Bound chicken antibody is eluted 1/10 volume of 1M sodium cyanoborohydride (Ald immediately with 2M Guanidine-HCl (pH 8.0) and the rich) was added. The mixture (hereinafter "antigen 25 antigen matrix is washed with PBS. The eluate is col matrix') was agitated for 4 hours at room temperature. lected and measured (A280). Relative elution efficiency The antigen matrix was then washed on a glass funnel is calculated as the total A280 units collected here di with PBS. The filtrate was collected and coupling effi vided by the total A280 units collected for 4-M guani ciency was calculated as in Example 6. The results dine-HC. showed that coupling efficiency of activated Resin III 30 The same antigen matrix is reacted a third time with was 78%. The antigen matrix was stored in PBS con the same amount of PEG-purified chicken antibody and taining 0.02% sodium azide at 4 C. washed as above. Bound chicken antibody is immedi EXAMPLE 11 ately eluted with 8M Urea (pH 8.0) and the antigen Covalent Attachment of a Mixture of Whole Snake 35 matrix is washed with PBS. The eluate is collected and Venoms to an Aldehyde-Activated Agarose Matrix measured (A280). Relative elution efficiency is calcu lated as above. In this example, the coupling efficiency of the alde The same antigen matrix is reacted a fourth time with hyde-activated resin, ACTIGELA (Sterogene) (here the same amount of PEG-purified chicken antibody and inafter "activated resin III') was studied with a cocktail 40 of four venoms. C. atrox, C adamanteus, A. piscivorus washed as above. Bound chicken antibody is immedi and A. contortrix venoms were dissolved together in ately eluted with 4M Urea (pH 8.0) and the antigen PBS (pH 7.2) with each venom at a concentration of 10 matrix is washed with PBS. The eluate is collected and mg/ml. Activated resin III was washed with 6 volumes measured (A280). Relative elution efficiency is calcu of PBS and added (in equal volume) to the venom solu 45 lated as above. tion. Thereafter, 1/10 volume of 1M sodium cyanobor The same antigen matrix is reacted a fifth time with ohydride (Aldrich) was added. The mixture (hereinaf the same amount of PEG-purified chicken antibody and ter "antigen matrix') was then agitated for seven hours at room temperature and left overnight at 4 C. The washed as above. Bound chicken antibody is immedi antigen matrix was then washed on a glass funnel with ately eluted with 0.5M diethylamine (pH 11.5) and the PBS. The filtrate was collected and coupling efficiency 50 antigen matrix is washed with PBS. The eluate is col was calculated as in Example 6. The results showed that lected and measured (A280). Relative elution efficiency coupling efficiency of activated resin III was 54%. The is calculated as the percent of maximum yield (see antigen matrix was stored in PBS containing 0.02% above). The efficiency of elution for the five eluents is sodium azide at 4 C. shown in Table 2. It can be seen that 4M Urea and 0.5M 55 diethylamine are poor eluents; they fail to remove all EXAMPLE 12 the bound chicken antibody. The other three eluents Elution of Specifically-Bound Antibodies from an remove 90% or more of the bound antibody. Affinity Matrix with Different Eluents TABLE 2 In this example, the elution efficiency of various elu ents on chicken antibody is demonstrated. Chicken Efficiency of Eluents antibody is generated, collected and extracted from Eluent Efficiency eggs as described in Example 1. Resin I is used to pre 4 M Guanidine-HCl, pH 8.0 100 pare an antigen matrix as in Example 6. Five eluents are 2 M Guanidine-HCl, pH 8.0 90 studied in successive elutions. Between each elution, the 65 8 M Urea, pH 8.0 100 antigen matrix is washed with TBS, the eluate is col 4 M Urea, pH 8.0 65 lected and measured (A280), the antigen matrix is 0.5 M diethylamine, pH 11.5 38 stripped of remaining antibody with 4M guanidine-HCl, 5,196, 193 35 36 06/197,714, which is hereby incorporated by reference. EXAMPLE 13 The eluate ("actigel/actisep-purified 355") was col Elution of Non-Mammalian Antivenon Antibodies lected and measured (A280), and the antigen matrix was from an Aldehyde-Activated Agarose Venom Antigen washed with TBS. Remaining antibody on the antigen Matrix Using Guanidine matrix was eluted with 4M guanidine-HCl. This eluate In this example, chicken antibody was eluted with ("actigel guano purified 355" was collected and mea 4M guanidine-HCl (pH 8.0) from an antigen matrix sured (A280), and the antigen matrix was then washed made up with resin I. Approximately 10 mg C. atrox with PBS. Elution efficiency was calculated as the per venom was coupled per ml of CNBr-activated Se cent of the total antibody eluted by ACTISEP and was pharose 4B as in Example 6 above. 100 ml of PEG-puri 10 found to be 26% (400 pig total protein). The remaining fied 355 (see Example 2) was loaded on a 5 ml antigen 74% (1.5 mg total protein) was eluted by quanidine. matrix at a flow rate of 1 ml per minute. The flow through ("355 flow through') was collected and the EXAMPLE 6 antigen matrix was washed successively with several Further Increasing the Elution Efficiency of a bed volumes of PBS, BBS-Tween (0.1M boric acid, 15 Non-Denaturing Eluent 0.025M sodium borate, 1M NaCl, 0.1% (v/v) Tween 20 In this example, the elution efficiency of ACTISEP pH 8.3), and PBS until the effluent was free of protein on an activated aldehyde resin was optimized by a mod (A280) Bound chicken antibody was eluted immediately ified protocol; the residence time was increased to 2 with 4M guanidine-HCl (pH 8.0) and the antigen matrix hours (Assay 1) and then to 2 hours and 45 minutes was washed with PBS. The eluate ("Resin I purified 20 (Assay 2). 355') was collected and measured (A280) and found to contain 125 ug of antibody per ml of PEG-purified 355 Assay 1 applied. The cocktail matrix of Example 11 (above) was used EXAMPLE 14 to immunoaffinity purify 10 mls of PEG-purified 355 25 (see Example 2). The antibody was applied as in Exam Elution of Non-Mammalian Antivenom Antibodies ple 15 except that the residence time was increased to 90 from an Aldehyde-Activated Agarose Venom Antigen minutes (time 0 is just before the eluent is detectable in Matrix with a Non-Denaturing Eluent the effluent of the column). The eluate was collected In this example, chicken antibody was eluted from an and measured (A280), the antigen matrix was stripped of aldehyde-activated resin with 4M guanidine-HCl (pH 30 remaining antibody with 4M guanidine-HCl, and this 8.0). 5 mg C. atrox venom was coupled per ml of UL stripped antibody was collected and measured (A280). TROGEL AcA 22 as in Example 7 above. 5 mls of Elution efficiency was calculated as in Example 15 and PEG-purified Pool 4 (9 mg/ml total protein) antibody found to be 47%. was loaded on a 3 ml antigen matrix at a flow rate of 1 ml per minute. The flow through ("Pool 4/ultro flow 35 Assay 2 through") and the antigen matrix was washed succes The C. atrox matrix of Example 8 was used to in sively with several bed volumes of PBS, BBS-Tween munoaffinity purify 10 mls of PEG-purified Pool 2 (see (0.1M boric acid, 0.025M sodium borate, 1M NaCl, Example 4). The antibody was applied as in Example 15 0.1% (v/v) Tween 20 pH 8.3), and PBS until the efflu except that the residence time was increased to 2 hours ent was free of protein (A280) Bound chicken antibody and 45 minutes (time 0 is just before the eluent is detect was immediately eluted with 4M guanidine-HCl (pH able in the effluent of the column). The eluate was col 8.0) and the antigen matrix was washed with PBS. The lected and measured (A280), the antigen matrix was eluate ("ultro-purified Pool 4') was collected and mea stripped of remaining antibody with 4M guanidine-HCl, sured (A280) and found to contain 742 g of antibody and this stripped antibody was collected and measured per ml of PEG-purified Pool 4 applied. 45 (A280). Elution efficiency was calculated as in Example 15 and found to be 73%. EXAMPLE 1.5 The results of this example are shown in FIG. 8 (the Increasing the Elution Efficiency of a Non-Denaturing results of Example 15 are plotted in FIG. 8 for purposes Eluent by Increasing Column Residence Time of comparison). The efficiency of antibody elution was In this example, the elution efficiency of a recently increased to 47% with 90 minutes of residence time and developed elution medium on an aldehyde-activated to 73% with 2 hours and 45 minutes of residence time. resin was demonstrated. i0 mg C. atrox venom was EXAMPLE 17 coupled per ml of ACTIGEL as in Example 8 to make the antigen matrix. 10 mls of PEG-purified 355 (see 55 Optimization of Elution Efficiency Example 2) was loaded on a 5 ml antigen matrix at a In this example, the elution efficiency of ACTISEP flow rate of 1 ml per minute. The flow through ("355 on an activated aldehyde resin was optimized by a fur flow through') was collected and the antigen matrix ther modified protocol. 10 mg C. atrox venom was was washed successively with several bed volumes of coupled per ml of ACTIGEL to make an antigen matrix PBS, BBS-Tween (0.1M boric acid, 0.025M sodium 60 as in Example 8 above. Affinity purification was carried borate, 1M NaCl, 0.1% (v/v) Tween 20 pH 8.3), and out as in FIG. 9. 25 mls of PEG-purified Pool 2 (see PBS until the effluent was free of protein (A280). Bound Example 4) was loaded on the antigen matrix at a flow chicken antibody was eluted with ACTISEP Elution rate of 1 ml per minute. The antigen matrix was washed Medium (Sterogene) according to the manufacturer's and the bound antibody eluted with ACTISEP as in instructions, i.e. eluent was applied in a manner such 65 Example 15 except the residence time was increased by that it was in contact with the antigen matrix ("resi stopping the flow of the column at the point where the dence time') for 30 minutes. The formulation for AC peak of eluted protein concentration was reached. As TISEP is provided in U.S. patent appplication Ser. No. before, the antigen matrix was washed with TBS, the 5,196, 193 37 38 eluate was collected and measured (A280), the antigen mg C-atrox venom was coupled per ml of ACTIGEL A matrix was stripped of remaining antibody with 4M as in Example 8.2 mls of unpurified horse antivenom guanidine-HCl, this stripped antibody was collected (Wyeth; log #M878035) containing 210 mg/ml (total and measured (A280), the antigen matrix was washed 420 mg) were applied to a 5 ml antigen matrix, the with buffer and stored at 4 C. for later use. 5 flow-through was washed through the column initially The peaks in FIG. 9 are numbered to correspond to with PBS and saved for further analysis, the matrix was chicken antibody flow through (peak 1), non-specifi then washed with BBS-Tween until the effluent was cally bound antibody (peak 2), ACTISEP eluted anti substantially free of protein (A280) and then with PBS. body (peak 3) (note ACTISEP baseline, i.e. absorbance Bound antibody was eluted immediately with 4M guan attributable to the eluent alone), and 4M guanidine- 10 idine, collected and measured (A280) after complete stripped antibody (peak 4). Importantly, by stopping the dialysis. The antigen matrix was then re-equilibrated flow of the column at the point where the peak of eluted with PBS. This eluate contained 19.7A280 units of anti protein concentration was reached, the efficiency of body. antibody elution could be further increased to 89% The flow-through was then re-applied to the 5 ml (efficiency calculated as in Example 14) for an AC 15 Catrox venom antigen matrix and washed and eluted as TISEP elution of 128ug of antibody per ml of PEG described above in order to affinity purify any antibody purified antibody applied. An additional 16 g of anti not isolated in the first pass above. The guanidine-HCl body (per ml pf PEG-purified antibody applied) was eluate from the second pass contained only 3.4 A280 recovered with quanidine (for a total of 144 g/ml of units of antibody, indicating that approximately 85% of specific antibody per ml of PEG-purified antibody ap- 20 the Catrox specific antibody was purified in the first plied). pass. Importantly, the 23.1 A280 units (16.5 mg) of total EXAMPLE 1.8 antibody purified from both passes represents only 3.7% (16.5 mg/420 mg) of the total A280 units of protein The Increase in Titer of Antivenom antibody with present in the crude horse antivenom, indicating that Further Immunization 25 95% or more of the protein present in the antivenom In this example, chicken antibody was quantitatively does not react with Catrox venom. This is in contrast immunoaffinity purified from an aldehyde-activated with the pool #4 affinity purified chicken anti-Catrox resin to show increasing antibody titer with increasing described in Example 18 where 905 g/ml of a 9 mg/ml immunization. 10 mg C. atrox venom was coupled per PEG prep or 10% of the total protein was Catrox ml of ACTIGELA as in Example 8 above. The four, 30 specific antibody. Because of the higher concentration chicken #354, PEG-purified pools described in Exam of antivenom antibodies and increased protein homoge ple 4 were used. After each elution, the antigen matrix neity (see below) of the chicken IgY, we contemplate was washed with TBS, the eluate was collected and that this antivenom, without affinity purification, is measured (A280), the antigen matrix was stripped of useful for passive immunization (as well as in vitro ana remaining antibody with 4M guanidine-HCl, this 35 lytical work). stripped antibody was collected and measured A280, the To examine the composition of the crude and affinity antigen matrix was washed and a new pool of chicken purified antivenoms, analytical SDS-PAGE was per antibody was loaded to the antigen matrix. formed on antivenoms at different stages of purification. TABLE 3 A fresh 2 ml sample of crude horse antivenom (Wyeth; Antivenom Antibody Titers lot #M878035) was applied to the same 5 ml Catrox #354 Pool Titer (ug Ab/ml egg yolk) venom antigen matrix described above and the column 98 washed with PBS, BBS-Tween, and TBS until the efflu 144 ent was substantially free of protein (A280). Bound anti 600 45 body was eluted with ACTISEP using the optimized 905 protocol in Example 17 and collected and measured after complete dialysis. The antigen matrix was then Each pool (5-20 mls) was loaded on the same 5 ml washed with TBS and the remaining antibody eluted antigen matrix at a flow rate of 1 ml per min, and immediately with 4M guanidine-HCl, collected and washed successively with several bed volumes of PBS, measured (A280) after complete dialysis. The antigen BBS-Tween (0.1M boric acid, 0.025M sodium borate, matrix was then re-equilibrated by washing with PBS. 1M NaCl, 0.1% (v/v) Tween 20 pH 8.3), and PBS until Samples of the unpurified, flow-through, ACTISEP the effluent was free of protein (A280) Bound chicken and guanidine fractions were retained for further analy antibody was eluted immediately with 4M guanidine SS. HCl (pH 8.0) and the antigen matrix was washed with 55 Samples from the affinity purification of 25 mls of PBS. The eluates were collected and measured (A280) Pool 2 described in Example 17 were analyzed along for amounts of specific antibody. The results are shown with the horse antivenom samples by SDS-PAGE on a Table 3. The results demonstrate that the C. atrox 10% reducing gel (FIG. 10). Comparison of 100 g of specific antibody titer increased nearly ten-fold over a the applied crude horse antiserum (Lane 1) with 100 ug period of approimately 60 days due to further immuni of the crude horse antiserum flow through (Lane 2) zations. revealed no detectable differences in composition. However, 30 g (Lane 3) and 150 ug (Lane 5) of the EXAMPLE 19 guanidine-eluted horse antibody and 30 ug of the AC Purity of Mammalian and Non-Mammalian TISEP-eluted horse antibody (Lane 4) all exhibited a Antivenoms Immunoaffinity Purified on an 65 pattern of fewer polypeptides; the predominant poly Aldehyde-Activated Whole Venom Antigen Matrix peptide was found to be a 55,000 dalton band (size is In this example, the purity of antivenoms before and estimated from molecular weight markers in Lane 7) after immunoaffinity purification was demonstrated. 10 corresponding to the relative mobility of the heavy 5,196, 193 39 40 chain of horse immunoglobulin. None of the higher ces in succession in order to remove as much C. atrox molecular weight polypeptides present in the crude reactive antibody as possible. The total amount of C. antivenom (Lanes 1 and 2) were found in the eluates atrox reactive antibody, calculated from the A280 of all (Lanes 3, 4 and 5), indicating that these high molecular three column eluates, was 7.5 mg per ml of this lot of weight proteins are not immunoglobulin and are re- 5 Wyeth antivenom. moved during affinity purification. Densito-metric To assess the spectrum of activity of the purified C. scanning of gel lane 1 indicated that no more than 37% atrox-reactive antibody, and to compare it with the of the total protein present in the crude Wyeth anti activity of both the crude antivenom and the non-C. venom was immunoglobulin (data not shown). Since atrox reactive fraction of the antivenom that flowed 3.7% of the total protein is venom-reactive (see above), O through all three C. atrox matrices, we examined anti it follows that 10% (3.7%/37%) of the total horse im body reactivity by ELISA using the four original ven munoglobulin is venom-reactive. oms used for immunization as antigens (see above). Similarly, 30 g of the PEG-purified, chicken anti 96-well Nunc Plates were coated overnight at 4' C. in a body flow through (Lane 9) contains high molecular humidified chamber with 2001/well of the appropriate weight polypeptides that are not recovered in the guani 15 venom dissolved in PBS at 5 g/ml. The venoms used dine (Lane 10) and ACTISEP (Lane 11) eluates. The in this example were C. atrox, C adamanteus, C. du bands associated with these eluates (Lanes 10 and I1) rissus terrificus, and B. atrox (Sigma). The next day the correspond to the polypeptides of a commercial (Cap wells were blocked with PBS containing 0.1% bovine pel) sample of pure chicken immunoglobulin (Lane 6). serum albumin for 2 hours at room temperature. Anti (The proteins from unpurified chicken yolks are shown 20 bodies to be tested for binding to each of the four ven in Lane 8). These results demonstrate that: i) the PEG oms were diluted in PBS containing 2% (v/v) normal purified chicken antivenom (Lane 9) is less heteroge goat serum (GIBCO). In order to directly compare the nous than the crude horse antivenom (Lane 1), in that crude Wyeth antivenom, the antigen matrix flow far fewer non-immunoglobulin proteins are present in through fraction, and the affinity purified anti-C. atrox the PEG-purified chicken IgY preparation (greater 25 with respect to their reactivities, the antibodies were than 90% of the total PEG-purified protein is immuno diluted in such a way as to normalize i) the volume of globulin and, given that 10% of the protein is venom the flow-through fraction with the volume of the crude reactive, it follows that 11% (10%/90%) of the immu Wyeth antivenom starting material, and ii) the concen noglobulin is venom-reactive); ii) the affinity purifica tration of C. atrox-specific antibodies in the crude anti tion removes substantially all of the non-immunoglobu- 30 venom, i.e., since the crude antivenom contain 7.5 lin protein from the crude antivenoms (greater than mg/ml of C. atrox-specific antibody, to compare the 99% of the protein in these preparations is immunoglob unpurified crude antivenon to a set concentration of ulin); and iii) both the ACTISEP and guanidine eluates purified antibody dilutions were made of the crude containing the horse and chicken antivenoms are essen material such that the concentrations of C. atrox tially pure (greater than 99%) antigen-specific immuno-35 specific antibody were the same (e.g., 2.5 g/ml of globulin. Table 4 summarizes the immunoglobulin con specific antibody is 2quivalent to a 1:3000 dilution of tent, purity and reactivity of the horse and chicken crude antivenom). antivenoms at different stages of purification. Note that 200 ul/well of four 3-fold serial dilutions of each before affinity purification (see "crude horse' and antibody and a normal unimmunized horse serum con "PEG IgY"), less than 50% of the antivenom immuno- 40 trol (Sigma) were incubated for 2 hours at room temper globulin ("Ig') is venom reactive. ature. The plates were then washed three times with BBS containing 0.1% Tween 20, twice with PBS TABLE 4 Tween 20, and twice with PBS. Alkaline phosphatase Immunoglobulin Content, Purity & Reactivity conjugated goat anti-horse IgG (Fisher Biotech) was Antivenon % Ig % reactive protein % reactive Ig 45 diluted 1:500 in PBS containing 2% (v/v) normal goat Crude Horse 37 3.7 10 AP Horse >99 >99 >99 serum, added to the plates, and incubated 2 hours at PEG IgY >90 10 room temperature. The plates were washed as before, AP gY >99 >99 >99 except Tris-buffered saline was substituted for the last wash, and p-nitrophenyl phophate was added and the After affinity purification ("AP), greater than 50% of 50 hydrolysis of the substrate measured at 410 nm as de the antivenom immunoglobulin ("Ig') is venom-reac scribed in Example 2. tive. TABLES EXAMPLE 2.0 Reactivity of Purified Horse Antivenon 5 % of initial reactivity with Spectrum of Reactivity of Horse Antivenom before and 5 C. C. durissus after Affinity Purification on a C. atrox Antigen Matrix C. atrox adaranteus B. atrox terrificus Starting Material 100 00 100 100 In this example, the ability of a C. atrox antigen ma (crude Wyeth) trix to bind and purify the spectrum of antibodies pres C. atrax Flow- 11 7 21 49 ent in a crude polyvalent horse antivenom was exam through ined. 2 ml of Wyeth Polyvalent Crotalid antivenom (lot C. atrox affinity 88 97 85 40 #M878035), raised against C. atrox, C adamanteus, B. purified atrox, and C. durissus terrificus venoms, was applied to a 5 ml C. atrox Actigel A antigen matrix prepared as The ELISA results of the relative reactivity of the described in Example 8. The flow-through was col 65 crude horse antivenom, the C. atrox antigen matrix lected, the column washed, and antibody eluted as de flow-through, and the affinity purified anti-C. atrox scribed in Example 17. The flow-through fraction was antibody at a concentration of antibody that fell within applied to two more C. atrox-Actigel A antigen matri the linear range of the ELISA (the normal horse serum 5,196, 193 41 42 control values have been subtracted) on each of the four ELISA results of the relative reactivity of the different venoms are shown in Table 5. The results show that preparations at a specific antibody concentration that affinity purification of this antivenom with the C. atrox fell within the linear range of the assay are shown in antigen matrix recovers the majority of the antibody Table 6. By passing the crude Wyeth antivenom sequen activity against three of the original venoms used for tially over the C. atrox antigen matrix followed by the immunization, C. atrox, c. adamanteus, and B. atrox, C. durissus terrificus antigen matrix, one recovers in the indicating that these three venoms contain many similar sum of the two purified antibodies more than 66% of antigens. However, the fourth venom, C. durissus ter the crude antivenom's initial reactivity with all four rificus, is not as strongly reactive with the C. atrox puri VenOS. fied antibody. In fact, the majority of the C. durissus 10 terrificus reactivity by this assay still remains in the - TABLE 6 flow-through fraction from the C. atrox antigen matrix. Retention of the Spectrum of Reactivity of a Horse Antivenom by Sequential Immunoaffinity These examples demonstrate the existence of at least - Chromatography - two antibody populations in the crude Wyeth anti C. C. B. C. durissus venom, one that is polyvalent and reactive with C. atrox 15 and the other three venoms, and one that is C. durissus atrox adamanteus atrox terrificus Crude Wyeth AV 100 100 00 100 terrificus reactive but not C. atrox-reactive (C, durissus C. atrox flow-through 8 10 22 55 terrificus monovalent). before C. durissus EXAMPLE 21 matrix 20 C. atrox/C. durissus 8 8 20 0 terrificus matrices Identification and Purification of Two Monovalent flow-through Antivenom Antibody Subpopulations and One Affinity purified

65