THE USE OF ION-EXCHANGE RESTNS FOR THE PURIHCATTON OF TYPE E DOMNJS TOXIN Thesis for ”N chru 0‘ pk. D. MICHIGAN STATE UNIVERSITY Alfred Mills Wallbank 1956 “mun Au STATE U SITY LIBRAR K mu lltllflllllllflifliflllrl mum J (a. A 3 1293 01404 2273 This is to certify that the thesis entitled The Use of Ion—Exchange Resins for the Purification of Type E Botulinus Toxin presented by Alfred Mills Wallbank has been accepted towards fulfillment of the requirements for Ph. D. degree in Microbiology Date November 30, 1956 O~169 flinging”) State [‘5 LIBRARY U n i \' cm ty- éva.. ‘ —— n A A 5 3' .40» TV? U53 OF IOY-EICVATGB ET I”? FCR THE 2an hICnr‘Wr'WT (“urn MK"? ‘7‘ T‘ "-mwllwr'r‘ Thai-'1’? I La--I.‘: ‘Li ‘VA‘ bk .L 4.; 3...} .LIJ .' «L K.) ‘k’ J K'n‘. ,L- - uy Alfred Kills Uallbank {7. 'T“!HY‘:\‘{‘| All .1; .J lulu/T Submitted to thr School of Graduate Studies of Michivon 0" State University of Agriculture and Applied Science in partial fu fillment of the requirerents for the degree of -CCTCR OF PHILOSOPHY Department of fiicrobiology and Public Health 1956 Appr O‘Ved “W’N‘TBKXNWVVW ' \ Alfred Hills Nallbank ‘V J- Pure type A ootulinus toxin is the most active toxic M- substance rznown at tne presetit time. One milligram will kill 1,200 tons of living matter. The possible use of the toxin in. biological warfare, and interezt in whv it is more toxic than other conjounflr, led to th5s study of type B Clostridium totulinni toxin. Both tyges A and B toxin have been purified. Type E ootllinus t xin, which is the onlv ot or t .e toxic for humans, had net been gurif iel at the tine this study was undertaken, but while the work was in progress, it was purified ty ethanol pre- cipito.tion at? ort Detrick, lwrgllla by Gordon et al. (1). This stu‘y we a temrt to nurify t"ne E togin illg 10n— exchange.resins. At the beginning of this wort there was no metlod aV31Ml ole to produce ni. 3h concen era tinols of t We E botulinus J. toxin, The SoUtj was started with type A botulinus toxin and was changed to tyn E toxin when a nethod was found A strrng basic anion exchanger was used to absorb crude type A t xin ‘nut no method was found to elute the toxin from the ex- changer. Then, type A toxin was adsorbed ty a strong cation ex chanre resin and a toxic fraction was elizt: with a pnosonate buffer at pH 7.0. Since the stronc cation exchanger mas success fully used in ‘AD adsorntion of the t;;re A t012in, it was tried with the crude type L E toxin with favorable results. Arain, as with the type A toxin, a toxic fraction was eluted from the tyre E toxin with a phosphate buffer at pH 7.0. The eluate moved as o:1e component in paper .3. 1 '1 7‘ 1, q. 1!- ‘\ 1 VA ~ '1‘ - f1 ‘ ‘v‘ r‘, J‘ r1 T“ '7‘ electionnoretlc siloIi.enu. but o 1 not rave eh s re motilit FJ. L ‘H - ~- x J- p L‘s r1 ‘ 1"».1‘. as any of tnC conyonenes 01 one crude to 1n. Using the Ouchterlony technigue, the eluate lid not give evidence of containing an antigenic c3 “onent idsl: oical to any of those agpearing in crude type E toxin, which WOllld lefld to the assunption tha t tre eluate toxicity was caused by a molecule 0 (1 4L. ant1g cnicall" unrelatcl to the orig no iraCeion. I Tyne A toxin loses 30 to 75 per cent of its toxicity upon filtra ivn throuyb filters for bacterioleical filtration. Type V" o s tox n w: )u s not as la 110, for filtration through a Soles 015 vs filter had no significant effect on tie titer of crude type n 9 toxin. Als stort3e in a freezer at —?30 M1 not reduce the titer of the toxin. his study has nresorted no evidence t.at ty;e E botulinus toz-tin can be puri 1ed by ion-excl anrje res i155 Emile the use of ion-exchangers has not given one desired results in these studies, ‘ ner sho uld 1e given considw ation in I.‘ ’J. the wr1eer tel' o H (D 3 4 ‘1 'J C!) (-5- 1:31 __1 c.1- c+ ourification of metabolic products of r1croorrrn1ezs. Alfred Kills Hallhank «ID-r7?“ 1.4"”? 71975 ;-.4.‘. qu..‘J ,. DUff, Jo To . 1-. - .1 -‘IC ixvtr, A.; (,L‘J‘I. 1" r1 - .9 . .. ,1, -,.,.. ' E 1 iifl a1: Toaold ‘ 1.; Fiock, J. A.; Yuri Puriiied Clostriflium totulinum ty3e 1. Gordon, Eact. Proc. THE USE OF ION-EXCHANGE RESINS FOR THE PURIFICATION OF TYPE E BOTULINUS TOXIN By -5 Alfred Mgiwallbank A THESIS Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1956 Acknowledgement The author wishes to express sincere appreciation to Dr. W. L. Mallmann for his interest. guidance and under- standing in these studies. Appreciation is also extended to Robert Christian, Irving Dahljelm, Lisa Neu, and Dr. J. w. Fairley. To Richard Hetherington of the Rohm and Haas Com- pany and Jack D. Kerr of the Dow Chemical Company for the generous supply of ion-exchange resins to carry out the following experiments. TO BARB VITA Alfred Mills Wallbank candidate for the degree of Doctor of Philosophy Final examination: November 30. 1956, 3:00 P.M., Room 300, Giltner Hall Dissertation: The Use of Ion-Exchange Resins for the Purification of Type E Botulinus Toxin Outline of Studies: Major Subject: Microbiology Minor Subject: Biochemistry Biographical Items: Born: May 13, 1925, Farmington, Michigan Undergraduate Studies: Michigan State College, 1944- 1948, 5.3., 1948 Graduate Studies: Michigan State University, 1951-1956, “030' 1953 Major Professor - Dr. H. L. Mallmann Thesis Title: Removal of Botulinus Toxin from Water. Ph. D., 1956 Major Professor - Dr. W. L. Mallmann Experience: Bacteriologist, Barry Laboratories, Detroit, Michigan, 1948-1950. Bacteriologist, Henry Ford Hospital, Detroit. Michigan. Blood Sterilization Project. Graduate Teaching Assistant, 1953—1956. Member: Society of American Bacteriologists, The New York Academy of Sciences, Society of Sigma Xi. TABLE OF CONTENTS Page No. I. Introduction.............................. ........ 1 II. Review of the Literature..........................3 A. Botulism..................................3 B. Types of Clostridiugfbotulinum............6 O. glostrigium botuliggm Type E..............7 D. Use of Ion-Exchange Resins to Purify ViruseBOOOOOOOOOOOOOO0.0.00000.00.12 III. Materials and Methods............................l4 A. Media....................................14 B. Organism.................................14 C. Toxin Production.........................15 D. Experimental Animals.....................l7 E. Dosage and Toxicity......................18 F. Dilution of Toxin for Titration..........l9 G. Ion-Exchange Columns.....................19 H. Conditioning the Ion—Exchange Resin......l9 I. Paper ElectrOphoresis....................2l J. Ouchterlony Technique....................22 K. Production of the Antisera...............23 L. Different Methods Employed for TypeATOXinOOOOOOOOCOOOOOOOOOOOOOOOO.0.O24 M. Column Method for Adsorption and E1Ut10n0f Tox1n.OOOOOOOOOOOOOOOOOOOOOOOO25 N. Ion-Exchange Resins......................25 Page No. IV. Results...........................................31 A. Type A Botulinus Toxin....................31 E. Type E Botulinus Toxin....................37 V. Discussion........................................46 VI. Summary...........................................50 VII. Appendix..........................................51 VIII. References........................................52 LIST OF TABLES Page No. Table I Reported Isolations of Clostridium botulinum Type E From Food..................9 Table II Unsuccessful Methods of Eluting Type A Toxin from Amberlite XE-98 (0H‘)...........33 Table III Resins that Did Not Adsorb Type A TOXin from SelutloDOOOOO00.00.000.000.00.0.35 Table IV Elution of Typfi A Botulinus Toxin from Dowex SO-XB (H ) with Phosphate Buffer, pH 7.000COOOOOOCOOOOOOCCOOO0.0.00.00.00.00036 Table V The Effect of Filtration and Storage in a Freezer on Type E Botulinus Toxin.....38 Table VI Elution of Typ; E Botulinus Toxin from Dowox 50-X8 (H ) with pH 7.0 Phosphate Bafferoooooooooooooo00000000000000.0000...039 Table VII Elution of Typ: E Botulinus Toxin from Dowex 50-X2 (H ) with pH 7.0 Phosphate BUfferoo00000000000000.0000000000000000000.40 LIST OF FIGURES Page No. Figure I ”Intussuscepted" Cellophane Bag unit....................0..............16 Figure II Cation Exchanger...........................28 Figure III Anion Exchanger............................29 Figure IV The Flow of Inorganic Salts Through Ion-Exchange Columns.......................30 Figure V Ouchterlony Test - Plate l.................42 Figure VI Ouchterlony Test - Plate 2.................43 Figure VII Ouchterlony Test - Plate 3.................44 I. INTRODUCTION Botulinus toxin has been suggested as a lethal agent in biological warfare (1). Since 0.25 pg of the pure toxin would kill a 70 kilogram man (2) or seven ounces dis- tributed properly would kill the entire population of the world (3) it would appear that botulinus toxin is highly suitable for warfare. Because of its possible use. an impetus to its study as to production. means of control, and the nature of the toxin itself, has occurred. In a Masters Thesis (4) the writer was concerned with destruction or removal of botulinus toxin in municipal water supplies. He found that chlorine would inactivate type A botulinus toxin but ondy in high concentrations that could not be applied practically in water purification systems. An ion-exchange resin was tried and was found effective in removing the toxin. Sober, Kegeles. and Gutter (5) have proposed that homogenity in ion-exchange chromatographic analysis be con- sidered one of the required criteria of protein purity. Since type A botulinus toxin, which is considered to be a protein (6), was adsorbed by ion-exchange resins. elution should give a “pure" protein. Both type A and B toxin have been purified. The only type of botulinus toxin left which is toxic to man was type E. Therefore a study using ion-exchange resins for purification of type E toxin was undertaken. There are many other methods available today for purification of proteins, but these methods necessitate either elaborate equipment or long involved chemical pro- cedures. A method which can be used for purification of proteins, requiring only a glass column and a few grams of ion-exchange resin, could be used in practically any labor- atory. II. REVIEW OF THE LITERATURE A. Botulism Botulism is an intoxication which was first des- cribed in 1820 by the German poet and medical writer Justinus Kernerm(1)4 The study of botulism, particularly of the toxin responsible for its manifestation, has led to a wider basic understanding of bacterial toxins and their role in mag... ' w The term botulism (from Latin botulus, a sausage) was first applied in south Germany to the paralytic toxic I syndrome resulting from the ingestion of spoiled sausage. In 1896 Van Ermengem (8) isolated from a ham a gram positive large rod which he called Bacillus botulinum, which was responsible for an outbreak of botulism in Belgium. Today, these organisms are placed in the genus Clostridium, com- prising spore-bearing anaerobic bacilli with a species designation of Clostridium botulinum or Clostridium para- botulinum. Van Ermengem demonstrated, as Roux and Yersin had similarly shown for diphtheria, that botulism was caused by a toxin produced by the botulinum organism. Kempner (9) observed that antitoxin could be produced in experimental animals by immunization with the toxin produced in cultures of the botulinum bacillus. These two reports form the basis for the subsequent widespread research into the problems of botulism and the nature of the toxin. Two groups of workers both working independently at Fort Detrick during World War II succeeded in purifying and crystalizing type A botulinus toxin. Lamanna, Eklund, and McElroy (10) based their method upon the observations of Snipe and Sommer (11), and Sommer (i2), that the toxin is acid precipitable from culture medium and can be eluted from the acid precipitate by buffer solutions at appropriate pH values. Abrams, Kegeles, and Hottle (13) used the usual methods of alcohol and salt fractionation of proteins. By both these methods of purification the protein met the usual criteria of protein purity. This purified type A toxin is one of the most highly toxic compounds known to man. The purified toxin is 15,000 times as active on a weight basis as the most toxic drug known, aconitin, and a molecule of toxin is 200 million times as toxic as a molecule of the drug (3). The MLD for a 20 g mouse is 3 X 10'11 g of crystalline toxin. The interesting question now arises; how does this toxin injure the tissues of higher animals? Chemical analysis has failed to answer this question. The toxin is made up of proteins composed of the same amino acids found in the normal tissue protein of the host itself. In the case of type A botulinus toxin, a complete amino acid analy- sis has revealed no unusual chemical groupings that might prove a clue as to why it is toxic. The calculated elementary formula of the toxin is: 040, 298 H62,679 Nio,472 012,634 P15-17 S123. -5- Its amino acid composition is represented by the expression: Glycine 166, Alanine 394, Valine 406, Leucine 708, Isoleucine 820, Proline 203, Pheny- lalanine 64, Cystine SHgo, (cystine 5-) 40, Methionine 64, TryptOphane 82, .Arginine 239, Histidine 60, Lysine 477, (Asparagine -NH2) 1370, Glutamic acid 953, Serine 374, Threonine 642, Tyrosine 672 (6). Comparison of the effects of botulinus toxin and curare has appeared in the literature for many years but Guyton and MacDonald (2) have recently presented data which indicate that its action is different from that of curare. Acetyloholine injected intra-arterially still caused con- traction of the muscle after botulism poisoning. With curare poisoning, intravenous injection of acetycholine does not cause contraction of the muscle. This indicates a fundamental difference between curare and botulinus toxin. Evidence is presented which indicates that the principal action of botulinus toxin is probably at the myoneural junction, though possibly in the terminal nerve fibrils. About the treatment of botulism poisoning Guyton and MacDonald stated, "Treatment of botulinus poisoning consists of massive doses of antitoxin, the use of artificial respiration and in the cases of severe poisoning, the administration of vasoconstrictor drugs. The fact that poisoning lasts for many months makes the results of such treatment discouraging. The use of artificial respiration for several months or longer is not practical, and if a patient is poisoned sufficiently to require vasoconstrictor drugs he will probably die anyway. The only real sal- vation seems to be the early use of anti- toxin in doses greater than 100,000 units of multivalent serum. Though antitoxin has been shown to be of value for guinea pigs as long as two days after poisoning, it is still true that its effect decreases exponentially with time. One must remem- ber that once the toxin has reached the nerve ending and produced its damage this action is irreversible for many months.“ B. Types of Clostridium Botulinum Leuchs (14) found that certain cultures of the botulinum bacillus, isolated from various sources, produced a specific toxin neutralized only by its homolOgous anti- toxin. In 1924 Bengston (15) distinguished the species of C. botulinum on the basis of their ovalytic (digestion of egg white) activity. Van Ermengem's isolate was called C. botulinum and was non-ovalytic. The American species was called C. parabotulinum and was ovalytic. Today emphasis in classifying organisms belonging to the botulinus group rests on the toxin-antitoxin neutralization specificity. Burke (l6)classified 12 American strains on this basis as A or B. On the basis of their specific toxins five types of C. botulinum have been recognized. They are designated as C. botulinum or arabotulinum, type A, B, Ox; Q3, D, and E (17). Neutralization of toxin by the homologous anti- toxin is fairly constant and few discrepancies have been noted. Pfeenninger (18) reported in 1924 that antitoxin produced against the toxin of an Australian strain, re- sembling type C and whose toxin was neutralized by type C antitoxin, would neutralize the homologous toxin but not the toxins produced by other type C cultures. This inter- esting observation indicated that some variation in type specificity can take place. C. Clostridium Botulinum Type E .4 Incidence Type E was not recognized until 1936 when Gunni- son, Cummings, and Meyer (19) reported on two cultures of C. botuligum sent to them by Dr. L. Bier of the Bacteriologic Institute at DnieprOpetrowsk, Ukrania, U.S.S.R. Toxin- antitoxin neutralization tests in mice and guinea pigs re- vealed that antitoxins of types A, B, C, and D in doses adequate to protect against 250 to 170,000 MLD of homologous toxin failed to protect against two to five MLD of toxin produced by these cultures. Conversely, they found that antitoxin produced by this toxin failed to protect against two to three doses of types A, B, C, and D toxins. 0n the basis of these results they prOposed the designation of C.botulinum type E for these cultures. They further pointed out that the designation of the g;_parabotulinum gap; of Theiler and Robinson as type E in T0pley and Wilson, Second Edition, 1936, p. 688, was incorrect since the organism of equine botulism belongs to type C. (It would seem that its neutralization by type C or E antitoxin is of more impor- tance than its effect on the horse). In 1937 Hazen (20) reported on a strain of C. bot- ulinum isolated in December, 1934, from German canned sprats, which had been established as the cause of three cases of botulism. In vivo and in vitro attempts to neutralize the toxic filtrate obtained from this culture by monovalent botulinum antitoxic sera types A, B, and C resulted in fail- ure. In 1938, Hazen (21) reported the isolation of C. bot- ulinum from salmon, smoked and canned in Nova Scotia (later corrected by Dolman and Kerr (22) and established as Lab- rador), and implicated in another outbreak of botulism in New York State. With the cOOperation of Dr. K. F. Meyer of the HOOper Foundation, San Francisco, she was able to iden- tify her two cultures as belonging to type E. Since then many other incidents involving type E have been reported: Geiger (23), 1941; Dolman and Kerr (22), 1947; Dolman, gt 3;; (24), 1950; Prevot and Huet (25), 1951; Meyer and Eddie (26), 1951; Sakaguchi, gt 2;; (27), 1954. In Table I are summarized the reports appearing in the literature on the various isolations of C. botglinum type E that have been identified. J .tu E . ”I‘ll-I'll... i [Pr-Ill fishing .4. . .. .1..A.1 it "81 - 9 - TABLE I Reported Isolations of Clostridium botulinum Type E from Food: Year Reported Source of Designation Reported By Isolation Country of Strain 1936 Gunnison, Russian Soviet Russian 151 Cummings, Sturgeon Ukraine Russian 30-17 & Meyer(l9) 1937 Hazen (20) German Westchest- Sprat No. canned er County, 35396 sprats N.Y. (A.T.C.C.9565) 1938 Hazen (21) Labrador COOpers- Salmon No. smoked town, N.Y. 36208 salmon (A.T.C.C.9564) 1941 Geiger(23) Yugoslav San Fran- Mushroom mushrooms cisco, canned in Calif. California 1947 Dolman & Home Nanaimo, Nanaimo Kerr (22) canned B.C., salmon Canada 1950 Dolman, Home vancouver, VH Chang, pickled B.C., Kerr & herring Canada Shearer(24) 1951 Prevot & Fresh France French Huet (25) perch 1951 Meyer & Uncooked Point No desig- Eddie (26) whale HOpe, nation to flappers Alaska date 1951 Sakaguchi Herring- Japan No desig- gt a1. (27) Izushi nation to date 1952 “ Flatfish- “ " Izushi 1952 " Flatfish & Dace- Izushi " " 1953 " Flatfish-Izushi " " 1953 " Gilthead-Izushi " Tenno - 10 - Toxigenic Pronertieg Type E toxin affects laboratory animals and man in much the same way as do the toxins of the other human bot- ulinum types (A and B) as far as can be judged by the symp- toms that develop. The symptoms are disturbance of vision, muscular paralysis and respiratory failure. A Most investigators have found that the toxin pro- duction in culture media with type E strains is quite vari- able. The Canadian workers have obtained high toxin yields with the VH (Vancouver-Herring) strain of Dr. Dolman (22). By conventional culture methods type E toxin yields have been lower than the other two types (A and B) "h1°fi,9éé§9 human botulism. Toxigenic type A cultures (par- ticularily the Hall strain) regularly produce titers of 500,000 to 1,000,000 MLD per m1 of culture for the white mouse. Rice, Smith, Pallister, and Reed (28) report titers of 600 to 1,400 MLD per ml with type B for the mouse, but point out that they found the mouse much more resistant to type B toxin than the guinea pig. Nigg, gt 5;; (29) obtain- ed titers of 8,000 mouse MLD per ml with their type B culture. Early work with the type E culture by Dolman, 2; a1; (22) yielded 4,000 MLD per ml with the VH strain. Barron (30, 31) has adapted the method of Sterne and Wentzel (32) who described a double surface dialyzing membrane arranged by intussuscspting ("invaginating") cellophane bags for the -11- 1arge scale production of types C and D botulinum toxin and toxoid. The "invaginated" celIOphane bag is filled with water or saline (0.85%) and suspended in the culture medium and the entire unit sterilized. The saline is inoculated with the organism and the culture allowed to incubate. Dialysis of the medium, according to Sterne and Wentzel, is an essential part of the method. As the culture grows waste products presumably dialyze into the medium surround- ing the bag, while fresh nutrients continue to dialyze into the bag. Since botulinus toxin is non—dialyzable the slab- orated toxin accumulates inside the bag. The maximum titer is reached, when incubated at 30 C, in 10 days. The toxin is harvested by withdrawing the contents of the bag. Sterne and dentzel referred to the toxins produced in the cellophane bag as "dialysate" toxins. After this work had been in progress for some time Gordon, gt gtt (33) at Fort Detrick, Maryland, purified type E toxin by a procedure involving precipitation of toxin by ethanol in the cold, extraction of toxin from the precipi- tate with calcium chloride solution and reprecipitation twice with ethanol in the cold. The product obtained con- tained 9 X 104 LD50/mg N. Duff and co-workers (34) also reported a study on the activation of type E toxin by trypsin. Their investigations showed that the toxicity of cultures of _g; botulinum were increased from 12 to 47 fold by treatment - 12 - for one hour with one per cent trypsin (Difco 1:250) at pH 6.0 and 37 C. A group of Japanese workers have isolated an an- aerobic organism which appears to belong to the genus gigg-l tridium which increases type E toxin production in a mixed culture. They have called this organism strain No. 13. This strain itself was proved to be non-toxigenic for mice. Fur- ther work has shown that a sterile culture filtrate of strain No. 13 will give higher titered toxin from type E organisms and is likely to be an enzyme in regard to its action and protein like prOperties (27, 35, 36, 37). 9. Use of tggggxchangers to Purtfy Vtguggg Muller (38) was the first to employ ion~exchange resins to aid in the purification of viruses. He used a cation exchange resin (XE-64) to remove inert nitrogenous material from suspensions of chick embryos and mouse brains infected with neurotrOpic virusesa Lo Grippo (39) used XE-67, an anionic exchanger to adsorb the Lansing strain of poliomyelitis and Theiler virus to the resin along with nitrogenous material. The virus was then selectively eluted from the resin. In his second paper Muller (40) adsorbed the PR8 strain of influenza virus with a cation exchanger (XE-64) and then eluted the virus from the resin with sodium chloride solutions. -13- Takemoto (41) indicates that a cation exchange resin, Nalcite HCR-X12 (Dowex 50), can be used to adsorb type A influenza virus from human throat washings. The virus can be isolated more frequently with the use of resin eluates than without the exchanger. -14- III. MATERIALS ANQ METHODS A. Media Barron's broth (31) was used for toxin production in this study. The medium is composed of: Heart-infusion broth (Difco) 25 g Tryptone (Difco) 10 g Glucose (Reagent) 10 g Calcium carbonate (Reagent) 5 g Distilled water 1 L pH 7.4 - 7.6 The pH was adjusted with N/lO sodium hydroxide. The medium was made up without glucose and autoclaved at 121 C for 75 minutes because of the large volume of medium. The glucose was added aseptically after being autoclaved at 116 C for 20 minutes. Stock cultures were kept in cooked meat medium (Difco) and stored in the refrigerator. As cultures were needed the organism was transferred to fluid thioglycollate medium (Difco) and incubated for 24 hours to obtain an actively growing culture. B. Organism The VH strain of C. botulinum, type E, isolated from herring in vancouver, Canada, by Dolman was used during this entire study. The VH strain was obtained from - 15 - Dr. Jack Konwalchuk of the Defense Research Board, Kingston, Ontario, Canada. All cultures were incubated at 30 C to obtain maximum toxin production. C. Toxin Production A nine liter Pyrex carboy was used for a medium container. (See Figure I). The organisms were added to a cellOphane bag sus- pended in six liters of Barron's broth. Using this method the high molecular weight portion of the medium was kept out of the toxin. The cellophane used was 8.5 cm wide seamless cel- lulose tubing obtained from Visking Corporation, Chicago, Illinois. In order to "intussuscept" the tubing it was found that a thorough soaking of the cellophane in water greatly facilitated this procedure. The tubing, once attached to the unit, was kept moist since drying of the cellophane presented the possiblity of cracking and also made it difficult to handle. Care was taken to avoid damaging the cellophane in arranging the set up. Five m1 of an actively growing culture taken from fluid thioglycollate medium were added to 200 m1 of Barron's broth in an eight ounce prescription bottle. After 48 hours incubation the supernatant was decanted to avoid the cal- cium carbonate on the bottom of the bottle. The bacteria -15- ______ Withdrawal Tube Gas Glass Outlet Rod Glucose Inlet Hater \ and Inoculum __—_——_—_*' Inlet ”fi- l I I 'I ' l Annular Space ”‘ Medium ‘4razz’0arboy Culture Cellophane -\ Bag FIGURE I "Intussuscepted" Cellophane Bag Unit -17.. were then spun down in an International Refrigerated Centri- fuge PR-l (Angle Head) at 3,000 RPM for 30 minutes. The organisms collected from one prescription bottle were resuspended in 350 ml of water and added to the "intussuscspted' cellOphane bag in a carboy. Then, these large containers of broth were incubated at 30 C for 10 days. The contents of the bag were then aspirated into a flask. The toxin plus organisms were centrifuged in an Internation- al Refrigerated Centrifuge PR-l (Angle Head) at 3,000 RPM for 30 minutes. The supernatant was poured off and filter- ed through an 015 Selas filter. The toxin was immediately put in screw capped test tubes and either frozen or refrig- erated. A sterility test was made by adding two m1 of toxin to both fluid thioglycollate medium and cooked meat medium. These tubes were incubated at 37 C for seven days. D. EXQerimental Animatg The mouse was used for assay of the toxin because of its sensitivity to type A and type E toxin. Carworth Farms strain CF #1 was used for this entire study because of their uniform response in toxicity studies. The mice were fed Rockland Mouse Diet in pellet form. The mice were housed in cages made of either stain- less steel or Monel metal measuring 21 x 33 x 16 cm or 32 x 40 x 16 cm. The cages had suspended wire-mesh (0.8 cm mesh) food receptacles built into the center of the cover. --18- The water bottles were made available to the animals by in- serting the glass tubes through the covers. Shavings were used for bedding and were changed once a week or more fre- quently if necessary. The room temperature was approximately 21 0, except during the summer months, when it was difficult to maintain this temperature without air-conditioning. After discarding rough-furred or unhealthy looking mice, the remaining animals were separated into weight groups. Animals were selected at random from the 16 to 24 g group for preliminary work, and from the 18 to 22 g group for more critical experiments. E. Dosage and Toxicity The dosage of diluted toxin used was 0.5 ml ino- culated intraperitoneally in the following manner: the mouse was picked up by the tail in the left hand, held by the fur on the back of the neck with the abdomen facing up- ward, and the needle was introduced at approximately a 45 degree angle into the abdominal cavity. Preliminary toxicities were determined in terms of the minimum lethal dose (MLD). Using three mice per dilution, the highest dilution killing two or more mice was accepted as the MLD. The lethal dose for 50 per cent of the mice (LDSO) was obtained by using eight mice per dilution and calculations made by the method of Reed and Muench (42). *2“? The animals were checked daily for five days at the beginning of the work. After some time it Was noted that only two deaths had occurred after three days; so to facil- itate the work animals were checked for three days. Obser- vation revealed that death was usually preceded by labored breathing and what appeared to be muscular paralysis. F. Dilution of Toxin for Titration Serial, tenfold dilutions of the toxin being tit- rated were made in a gelatin-phosphate buffer (43) consist- ing of 0.2 per cent gelatin in a one per cent phosphate buffer, pH 6.9, autoclaved at 121 C for 15 minutes. A diff- erent sterile pipette was used for each successive dilution. G. Ton-Exchange Column; The Pyrex glass columns used in these studies were 44 cm long x 2.5 cm in diameter. The bottom portion of the column had a removable sintered glass filter that would re- tain the resin, yet allow flow of a liquid. The columns were always cleaned in dichromate cleaning solution and were well rinsed before using. H. Conditioning the Ion-Exchange Resin he ion-exchange resin was added to the column in a slurry and allowed to settle out. The new resins, as received from the manufacturers, -20- contained a small amount of fines, (extremely small particles of the resins) which had to be removed to prevent excessive resistance to downflow operations. Removal of the fines was accomplished by backwashing, which is the process of allow- ing enough water to flow up through the column to expand the resin bed 100 per cent. Backwash was continued until the fines had been washed from the bed. To obtain the hydrogen cycle, the cation exchange res in was treated alternately with 500 m1 of 1}; sodium hydrox- ide and 500 ml of lg hydrochloric acid. This procedure was repeated three times. The exchanger was washed to neutral- ity with distilled water after each reagent. The sodium cycle with cation exchangers was ob- tained by treating the resin alternately with 500 ml of l_I\_l_ hydrochloric acid and 500 ml 131 sodium hydroxide. The resin "‘8 then rinsed with distilled water until the effluent was at, neutrality. The Amberlite IRC-SO (Rohm and Haas) was treated wi th 500 m1 of 23 sodium hydroxide followed by water. 2.1! hYdrochloric acid, and washed with distilled water. Since the THC-50 was to be used in the (MINI) cycle, the resin "is treated with ammonium hydroxide and finally washed with ‘11 Stilled water until the pH was approximately 7.0. The anion exchange resins were treated as follows: L3 hydrochloric acid and then l_l§ sodium hydroxide to estab- 11 ah the exchanger in the hydroxide cycle; and with 113 -21- sodium hydroxide and 13 hydrochloric acid' to maintain the resin in the chloride cycle. The exchangers were washed to neutrality with distilled water after each reagent. I . Paper Electr0phoresi s The power source was made according to the speci— fications of Kunkel (44). Uhatmann filter paper No. 1, cut into 2 x 18 inch strips or 4 x 18 inch strips, was used for all experiments. The paper was rinsed with water several times and dried before use. Both the crude type E toxin and eluate were dia- lyzed against the buffer used in the experiments. The buffer was changed five times over a period of 24 hours. The dia- 1y sis was done at 4 C. Barbital buffers (pH 8.6) at an ionic strength of 0.05 for the five hour runs, and an ionic strength of 0.075 fol" the 18 hour runs, were used in this study (45). The paper strips were immersed in buffer solution and blotted until almost dry. Strips were then placed on a ' 31-:L1conized glass plate (10 x 16 inch). The eluate (0.005 ml) was placed, by means of a pipette, in the exact center °f one strip, and this procedure repeated with the toxin on “10 trier strip. A second siliconized glass plate of the same 512-e was placed on top of the plate with the paper strips, and the two plates were secured by clamps. The electrOphoretic runs were made at room tem- -22.. perature (23 to 25 C) at seven milliamperes for five hours or eight milliamperes for 18 hours. The paper strips, after each electrOphoretic experiment, were dried at 37 C and then sprayed with a 0.5 per cent ninhydrin solution. Rubber gloves were used at all times for handling the paper. J. Ouchterlony Technique This technique is used for the identification of molecules (especially protein molecules) which are generally characterized better by immunologic specificity than are molecules having the same specificity by other tests. The test consists of adding antigen (AG) and anti- body (AB) to suitably Spaced reservoirs in an agar plate. The AG and AB diffuse through the agar and when the concen- trations of the two reactants are in equivalence a line of precipitation appears. These specific lines appear as straight lines or with only a slight arc. The non-specific lines appear as an arc around the wells (at least that has been the eXperience in these studies). The agar used in this work consisted of the follow- ing: agar (Difco), 20.0 g; sodium chloride (Reagent), 8.5 g; methyl orange, 0.02 g; merthiolate, 0.1 g; and dis- tilled water, 1 L; adjusted to pH 7.3 - 7.5. Forty ml of the melted agar was added to each of three plates. After the agar had solidified, a No. 5 sterile cork borer was used to out three cups out of the agar in each plate, thus forming wells. Two-tenths ml of melted agar was pipetted into the bottom of the wells to form a seal. The outside edges of the three wells were 18 mm apart and formed an equilateral triangle. In these tests, 0.2 ml of the following reagents were added once a day for seven days to their respective wells: in the first plate - toxin, normal serum, and anti- serum; the second plate - eluate, normal serum, and anti- serum; the third plate - toxin, eluate, and antiserum. After each of these additions the plates were returned to the 37 0 incubator. After the last addition on the seventh day, the plates were incubated for 24 hours, and then stored at # C. K. Production of the Antisera Three to four kilogram rabbits were used for pro- duction of antisera. The rabbits were given repeated intravenous in- Jections of type E botulinus toxoid and then toxin. The toxoid was made by adding 0.5 per cent formalin to 2,000 mouse LD50/ml of type 3 toxin and incubating the mixture at 37 C for 10 days. During the incubation period the mix- ture was gently rotated daily to insure prOper mixing. Type E toxin was diluted to 200 and 2,000 mouse LDSO/ml- The rabbit was first given 0.4, 0.8, 1.6, 3.5, and 5.0 ml of the toxoid and then 0.1, 0.2. 0.5, and 2.0 m1 of the -24.. 200 LD5o/m1 toxin at three day intervals. Fifty m1 of blood were bled from the heart 10 days after the last inoculation. The serum from this blood was used for preliminary studies. Three weeks after the last inoculation the rabbit was given 0.5 and 1.0 ml of the 2,000 LDSO/ml toxin at three day in- tervals. After 10 days 50 ml of blood were again bled from the heart. The serum from this blood was used for the more critical studies. The vessel containing the blood was allowed to stand undisturbed at room temperature until the blood was clotted and then placed in a 37 C incubator for two to four hours. The blood was then separated from the walls of the vessel, and placed at 4 0 for 24 hours. The serum was re- moved and added to sterile screw capped test tubes. One ml of the serum was added to fluid thioglycollate medium for a sterility test. L. Different Methods Employed forglype A Toxin The work by Lewis and Hill (46) has shown that clarified corn steep liquor, 4 g total solids; powdered milk, 20.0 g; commercial grade glucose (cerelose), 6.0 g; and tap water, 1 L; adjusted to pH 7.4 - 7.6, dispensed into eight ounce prescription bottles, and sterilized in the autoclave for 20 minutes at 121 C, gives high yields of toxin. In this study this medium was inoculated with two per cent of an actively growing culture of the "Hall strain" of Clos- -25.. tridium botulinum, Type A, from fluid thioglycollate medium CDifCO) and incubated at 35 C for 48 to 72 hours. After incubation the culture was centrifuged in an Intmernational Refrigerated Centrifuge PR-l (Angle Head) for 45 :ninutes at 3,000 RPM. The supernatant was the crude toxin and it was not filtered. M. Column Method for Adsorption and Elution of Toxin The water level was brought to one inch above the conditioned resin in the column and then the toxin was al- lowed to flow through the column at the rate of 0.5 ml per minute. After the solution of toxin had passed through the resin, the solution used for elution was then passed through at, the same rate. Both the toxin and eluate were collected in test tubes and immediately stored at 4 C until they were as flayed. A portion of the original toxin solution was also Stored at 4 C to obtain the titer of the toxin added to the column, Wharge Resins A solid exchanger must have three characteristics. First. it must contain ions of its own. Second, it must be insoluble in water under all conditions. Third, there must be enough space between its molecules so that other ions can move freely in and out of the solid. Low crosslinked resins have less selectivity for Specific ions and are capable 0f - 25 - exchanging ions of high molecular weights. An exchanger molecule is a long chain polymer. It magr have either a negative or a positive charge. To neutra- lizes this charge, smaller ions of Opposite charge are pre- serit. It is these small ions, that are not held by bonds to thee rest of the molecule, that exchange with the ions in soZLution, The chemical structure of two typical ion-exchan- ge1?s is shown in Figures II and III. Figure II shows a crosslinked styrene polymer that hass been sulfonated which leads to a polystyrene sulfonic acixd in which every benzene ring in the polymer contains one su1;fonic acid group. The resulting resin is a strong acid anti many commercial cation exchange resins have this general composition. Figure III shows crosslinked polystyrene that has beset) chloromethylated and then reacted with tertiary amines '00 give a "strongly" basic anion exchanger. The flow of inorganic salt solutions through a r953irl bed containing cationic, anionic, and both a cationic and anionic resin is shown in Figure IV. This will give anus) idea how ion-exchangers work in a simple system. Cohn has described in simple terms the principles lnvc“'leed in the ion-exchange system: "The most useful method for the separation of components in biochemical mixtures seems to be that of "elution analysis“. This is a two step process, wherein the - 27 - mixture is first adsorbed at the t0p of a column, previously prepared in a given form, and then eluted in such a manner as to bring each substance to the bottom of the column as a separate and distinct “peak" without significant change in the form of the exchanger. In general, the adsorption step is carried out under conditions where the affinities of the ions in question for the exchanger are maximal, or at least greater than during the elution sequence. Elution utilizes conditions which decrease the affinities, releasing the ions from the exchanger. In so far as the adsorption is due to ionic forces, elution conditions fall into two main groups, (1) simple in- crease in the concentration of the com- peting ion to the point where it dis- places the adsorbed ions by mass action (ionic-strength adjustment) and, (2) change (usually decrease) in the charge of the adsorbed ion by pH change, com- plex formation, the use of a nonpolar sol- vent, etc., (charge adjustment). The method of charge adjustment is ob- viously applicable only to ions with variable charge, a not unusual situation in biochemistry. Amino and carboxyl groups are readily ionized or deionized, phosphate esters become singly or doubly ionized, sugars form borate complexes, etc., at pH values that are not difficult to attain in the laboratory and usually are not destructive to the material under investigation." (47) -28- FIGURE II TION EXG GER 503' Na/ 303‘ Na/ / \\ va-——C 2 ———-CH CH2 CH CH2 ————-GH-——-— CH2 ANMNN‘ SO}- Na/ /’ /’ \ \ “N“'—-CH2 ---CH CH2 CH -—‘—'CH2 CH ’—-—'CH2--"- WWNNN‘ /’ \ _ 29 - FIGURE III ANION EXCHANGER NR3/ 01" NR3/ 01" / \ “AAA-CH2 — CH --—— CH2 — CH CH2 ca — CH2 NR}?! 01" / / H ll ' \ wvw — CH2 CH CH2 CH CH2 CH CH2 / l \ - 3o - FIGURE IV TNfiE FLOW OF INORGANIC SALTS_$HROUGHglON-EXCHANGE COLUMNS Taaoucs A CATION EXCHANGER THROUGH AN ANION EXCHANGE CaSO4 Na2504 Na/ Resin OH‘ Resin] Na2804 NaOH '_I'_iiROUGH onion? ANQ ANION EXCHANGERS CtClz i IH/ Resin 1 KCl 1 OH- Resin HOH - 31 - lg. assutzs At the beginning of this work there was no method available to produce high concentrations of type E botulinus toxin. This study was started with type A botulinus toxin but was changed to type S toxin when Barron and Reed (31) (‘5 published their study on the production of type a toxin. A;_lype A Botulinus Toxin It has been established that type A botulinus toxin can be adsorbed in a column containing Amberlite XE-98 (OH‘), a strong anion exchange resin (4). In this study many buffers with different pH values and several chemicals of various concentrations were employed and found to be ineffective in eluting type A toxin from Amberlite XE-98 (OH'). These different buffers and chemicals are listed in Table II. It should be noted that l.O§_sodium hydroxide will inactivate type A toxin in 30 minutes, but 0.1! sodium hydrox- ide has very little effect in the same length of time (4). Therefore, the eluate was collected in beakers set up with a magnetic stirring apparatus, and a Beckman H-2 pH Meter. The pH of the eluate was adjusted to keep it in the range of pH 5.5 to 7.0. As indicated in Table III many ion-exchange resins would not, at least in the cycle in these eXperiments, ad- - 32 - sorb type A toxin from solution. The anionic exchangers XE-67 (01'), X2-98 (Cl‘) and IR-4B (OH‘) did not adsorb any toxin at all. It should be noted that XE-98 operating in the hydroxide cycle will adsorb toxin from solution. The cationic exchangers XE-66 (Na/), 50-X2 (Na/), 50-X8 (H/), and IRS-50 (NH4/) also do not remove toxin from solution. Later it will be shown that both 50-X2 (H/) and 50-X8 (H/) will adsorb toxin under the prOper conditions. After unsuccessful attempts in using a cationic exchanger for the adsorption of type A toxin, the pH of the toxin solution was lowered below the iso-electric point (pH 5.6). This gave the protein a positive charge favoring the replacement of the hydrogen ion of the cationic resin with protein. This eXperiment was tried with type A toxin buffered to a pH of 4.2. A toxin solution (50 ml) contain- ing 6,200 LD5O/m1 was adsorbed by Dowex SO-XB (Hf). The type A toxin was eluted from the column with a phOSphate buffer at pH 7.0. (See Table IV). - 33 - TABLE II Unsuccessful Methods of Eluting Type A Toxin From Amberlite Xvi-93 (OH-I (Resin - 13.5 x 2.5 cm in a column) pH of Toxin Experiment Before Chemicals for Elution and No. Adsorption Fractions Collected 1 6.7 50 ml - 0.13 Sodium Chloride 2 7.2 100 ml - 10% NaQHP04 3 5.5 10 ml - 0.1! Sodium Acetate plus 0.1M Acetic acid " - pH 71.1 Acetate buffer" 90 m1 - pH 4.1 ” 4 6.7 10 ml - pH 4. 6 Acetate buffer“ ” - pH 4.1 " “ - 1% Acetic acid fl _ I? “ - Sodium Chloride N _ 10% N 5 6.9 50 ml - 0.001N Sodium Sulfate " - 0.0m " " n _ OolN " fl " - 1.0K “ " 6 6.7 50 ml - 0.13 Sodium Hydroxide 7 7.4 100 ml - 0.13 Sodium Hydroxide " - Experiment No. 8 _ 34 _ TABLE II CONT. pH of Toxin Before Chemicals for Elution and Fractions Collected Adsorption 6.6 100 n N N H N N m1 - 0.53 Sodium Hydroxide I! N N ml m1 ml N N N N N N 1.03 " N N Distilled water a a 0.53 Hydrochloric Acid 00 u n n Distilled water (A flow rate of 10 ml per minute was used for this experiment) “Buffers: See Appendix (#3 and #5) -35.. TABLE III Resins (In The Indicated Cycle)_That Did Not.Agsorb Type A (Resin - 13.5 x 2.5 cm in a column) Basia Amberlite XE-67 Amberlite XE-66 Amberlite XE-67 Dowex 50-X2 Dowex 50-X2 Amberlite XE-98 Dowex 50-x8 Dowex 50-X8 Amberlite THC-50 Amberlite IRC-SO Amberlite IR-4B Toxin From Solution Type Anionic Cationic Anionic Cationic Cationic Anionic Cationic Cationic Cationic Cationic Anionic pH of Toxin 6.8 7.4 6.7 6.7 7.1 6.7 7.4 7.2 7.5 4.1 6.7 -35- EABLE I! EXPERIMENT £2 Elution of T pe A Botulinus Toxin from Dowex 50-X8 (Hi) (13.54x 2.5 cm) with Phosphate Buffer, 23 1.0 (50 ml of Type A Toxin buffered to pH 4.2 Con- taining 6,200 LDSO/ml were added to the column) Eluate diluted l-lO 10 ‘ml Fractions (No. of mice killed) 1 0/3 2 0/3 3 2/3 4 3/3 5 3/3 6 2/2 7 3/3 8 3/3 9 3/3 10 1/3 11 1/3 12 1/3 13 0/3 14 0/3 15 013 50 0/3 Buffers: See Appendix (#1 and #4) - 37 - E. Type E Botulinus Toxin In view of the fact that Rice 23 3;; (28) noted that bacteriological filtration removes from 50 to 75 per cent of the toxicity of type A botulinus toxin, a study was made of the effect of both filtration and storage in a free- zer on type E toxin. Neither filtration through a Selas 015 filter nor storage in a freezer for nine days at -23 0 had any signi- ficant effect on type S botulinus toxin. (See Table V). The next experiment was a successful attempt to adsorb type E toxin (50 ml), buffered to a pH of 4.2, con- taining 200 MLD/m1, with Dowex 5o-x8 (Hf). An experiment was arranged to ascertain if the type 3 toxin could be eluted with a phosphate buffer at pH 7.0. The type E toxin (50 ml), buffered at pH 4.2, con— taining 5,400 LDSO/ml' was added to the column. The toxin (u? some toxic component was eluted. (See Table VI). The toxic portions of the eluate, which were pool- emi after preliminary titration, contained 2,000 LD50/m1' imne eluate gave a positive Biuret test and a negative Mo- lisch test. The results which are sumnarized in Table VII de- amnistrate again that a toxic fraction was eluted from Dowex 50-)(8 (ml). Type E toxin (50 ml), buffered at pH 4.2, con- tadiiing 44,000 LDSO/ml' was added to the column, and phos- Ifiuate buffer (pH 7.0) was used for elution. After prelimin- ary titration the eluate was pooled. The eluate titer was 4. 400 LI350/1111 - - 38 - TABLE V The Effect of Filtration and Storage_;n a Freezer on Type E Botulinus Toxin Control Filtered Stored in a Titer Toxin* Freezer“i 26000 Lose/m1 23000 LDso/mi 32000 LDSO/ml * 015 Selas Filter ** Nine days at -23 C -39.. EABLE VI EXPERIMENT £10 I Elution of T me E Botuiinus Toxip from Dowex 50-X8 (Hr) (13.5 x 2.5 cm) with pH 7.0 PhosphateAQuffeg (50 m1 of Type E Toxin buffered to pH 4.2 Con- taining 5,400 LDSO/ml were added to the column) Eluate Diluted Titer of 10 m1 1-10 (No. of Titer Pooled Toxin Fractions Mice Killed) in MLD in LDEOle 1 0/3 2 0/3 3 0/3 4 0/3 5 0/3 6 2/3 0 7 3/3 200 g 3/3 2000 3/3 20 2000 10 3/3 air 11 3/3 0 12 2/3 0 13 2/3 0 14 0/3 15 0/3 16 0/3 17 0/3 18 0/3 19 0/3 20 0/3 Buffers: See Appendix (#2 and #4) Flow rate - 10 m1/30 minutes - 40 - TABLE VII EXPERIMENT 1 Elution of T e E Botulinus Toxin from Dowex 50-X2 (Hf) (15.5 x 2.5 cm) with pH 1.0 Pho§phate Buffer (50 ml of Type E Toxin Buffered to pH 4.2 cone taining 44,000 LDSO/ml were added to the column) Eluate Diluted ' Titer of 20 ml 1-10 (No. of Titer Pooled Toxin Fractiong Mice Kgiled) in MLD in LDSO/ml 1 0/3 0 2 3/3 20 3 3 3 20000 4 3 3 2000 4400 5 3 3 onQJ 6 2 3 200 7 1/3 0 8 2/3 20 9 1/3 20 10 0/3 0 11 0/3 0 12 0 3 0 13 2 3 20 14 0 3 0 15 0/3 0 16 0/3 17 0/3 18 0/3 19 0/3 20 0 3 Buffers: See Appendix (#2 and #4) Flow Rate - 10 ml/30 minutes -41- The Ouchterlony test was chosen as the most crit- ical method to ascertain if the eluate contained toxin or if the eluate contained any fraction that was identical to one of the components of the toxin. For this test, an antiserum was prepared against the toxin and the antigens (eluate and toxin) were placed in the wells of an agar plate to react under standard conditions. In figures V, VI, and VII are photographs of agar plates showing the results of the Ouchterlony test. In plate 1 (Figure V), which contained type E toxin (T), normal serum (NS), and anti-type E botulinus serum (AS), six lines 0f precipitation appeared, at 37 C, between the wells con- taining T and AS, revealing at least six antigenic compon- ents in the toxin. A line of precipitation appeared between T and NS when the plates were placed at 4 0 (this was noticed after 10 day storage). The continuation of this line of precipitation between T and AS into T and NS indicates that toxin reacts with a component common to both AS and NS. In plate 2 (Figure VI) containing eluate (E), NS and AS, there were no specific lines of precipitation at either 37 or 4 C. Plate 3 (Figure VII), containing T, E, and AS had three lines between the T and AS wells, which signifies at least three antigenic components in the T. There were no lines of precipitation between E and AS, indicating the absence of any antigenic components common to the toxin. Plate*3, when placed at 4 C did not give any more lines of precipi- -42- FIGURE V Ouchterlogy Test Plate 1 NS - Normal Serum T - Type E Toxin AS - Anti-Type E Botulinus Serum - 43 - EIQHB§.!I Ouchterlony Test Plate 2 NS - Normal Serum E - Eluate AS - Anti-Type E Botulinus Serum -44.. FIGURE VII Ouchterlopy Test 1m; T - Type E Botulinus Toxin E - Eluate AS - Anti-Type E Botulinus Serum -45.. tation. The toxin was the same material used in experi- ment No. 11 and the eluate was the material eluted in experiment No. 11. The antiserum was produced in rabbits by a series of intravenous injections of the toxin used in experiment No. 11. In attempting to make the antisera, four rabbits were used, but in two of the rabbits, what was considered one-tenth of a rabbit LD50, was injected and killed the rabbits. Therefore, a toxoid was used for the first in- jections and then toxin was used. One of the remaining rabbits was killed while bleeding it the second time to ob- tain serum. The antiserum of the other rabbit produced up to six lines of precipitation with type E toxin in the Ouch- terlony test system. Patterns obtained with closed strip paper electro- phoresis with the eluate from experiment No. 11 and the type E toxin used in eXperiment No. 11 which had been dialyzed against barbital buffer, ionic strength 0.075 and pH 8.6, indicate the eluate moves toward the cathode slightly faster than one of the toxin fractions. The toxin formed four spots, two moving toward the cathode and two moving toward the anode. Other electrOphoretic experiments gave the same number of fractions but the mobilities differed due to the changes in unric strength of the buffer and the length of time the ex- Periments were run. -45.. V. DISCUSSION This work has established that both type A and type E botulinus toxins can be adsorbed and a toxic frac- tion eluted from ion-exchange resins. Elution from an ion-exchange resin has been pro- posed as a criteron of purity for proteins. The type E eluate fractions should be pure whether they are antigenti- cally the same as the toxin molecule or have been changed by action of the ion-exchange resin. Paper electrOphoresis also indicates that the eluate is pure since only one component was found. The mobility of the eluate fraction is different from that of any component of the crude toxin. This shows that there has been some change in the toxin molecule that was eluted. According to the results of the Ouchterlony test the eluate is not immunologically identical to any fraction of the crude type E toxin. This is possible, when the work with both type E and A toxins is examined critically. The molecular weight of “pure" type A botulinus t<1xin is considered to be 900,000, therefore it is hard to believe that a molecule of that size could be readily ab- sormed from the intestinal tract. The difference between txrtulinus toxins and other toxins is that botulinus toxin is toxic by mouth and that toxicity is not destroyed by pPCDteolytic enzymes of the intestinal tract. It is diffi- CUI-t to believe that these particular protein molecules - 47 - should be resistant to proteolytic enzymes. The assumption that toxin is not broken down is based on the fact that toxicity is not destroyed by pepsin or trypsin. Therefore it may be possible for the eluate to contain a breakdown product of the type E toxin molecule and still possess toxicity. The work at Fort Detrick by Duff 23 2;; (34) shows that treatment of type 3 botulinus toxin with trypsin in- creases the titer of the toxin when it is inoculated intra- peritoneally. The same toxin administered orally to mice shows no significant difference in toxicity. This certainly indicates that the proteolytic enzymes of the intestinal tract activate the type E toxin. It would be very interesting to compare the mole- cular weight of purified type E toxin with the molecular weight of the purified toxin after treatment with pepsin, trypsin or papain. In fact, a similar study of type A and type B would be helpful in understanding this whole problem of toxicity of botulinus toxins. Since one of the pressing problems concerning the Study of botulinus toxins is their use in biological warfare, it: would be highly significant to identify the smallest por- ticbn of the toxin molecule that still causes toxicity. It seemns logical to assume that the large molecule of type A doses not pass through the intestinal wall and that the size or the toxic fragment is quite small in comparison to the - 48 - original molecule. Of course the small fragments might com- bine after passage through the intestinal wall to again be- come toxic, for it seems that the toxin formed from bacteria must pass through the cell wall as small molecules and com- bine on the outside of the cell wall. The alteration of the toxic portion of the toxin or a breakdown product of the toxin may be caused by secon- dary reactions taking place between the ion-exchanger and the toxin molecule. The acid or basic prOperties of the resin may cause some change in the toxin. This may be es- pecially true with the strongly basic exchanger, Amberlite XE-98, since botulinus toxins are very labile in alkaline solutions. Also, the eXposure to strong electrical fields of a strong-acid changer, Dowex 50, may be deleterious (47). The different mobility of the eluate fraction as compared with the crude toxin indicates that there is a Change in the net cnarge of the eluate molecule. It is usually considered that only the "surfaces” of the particles are involved in a change in mobility (48). It should be remembered that a protein molecule dOes not have to be altered very much to change its anti- SGIUAHIJ. Landsteiner has shown that Just the addition of EHI acetyl, methyl or ethyl group to proteins shows a marked effect on the specificity of antigens. Type E toxin must not be as labile as type A toxin to filtration, for filtration through a Selas 015 filter did -49.. not significantly effect the toxicity. This study has presented no evidence that type 3 botulinus toxin can be purified by ion-exchange resins. The writer is still of the Opinion that with the number of new ion-exchangers that the chemists are making, one will event- ually be found that can be used to purify type E botulinus toxin. The chemist and biologist are studying proteins more every day and therefore more ion-exchangers will be designed for work with proteins. In fact, some laboratories have started "tailor-making" polymers for specific uses (49). Ion-exchangers should be of great interest to the microbiolOgist for they may eventually be used for purifi- cation of bacterial, rickettsial and virus toxins, and also many other metabolic products of microorganisms. - 50 - VI. SUMMARY Crude type B botulinus toxin was adsorbed by a strong cationic exchanger, Dowex 50-X2 (H/), and a toxic frac- tion was eluted from the exchanger with phosphate buffer, pH 7.0. The eluate did not contain an antigenic component iden- —\ tical to any of those appearing in crude type a toxin. The eluate moved as one component in electrOphoretic eXperiments but did not have the same mobility as any of the components of crude type E toxin. One of the components of the crude type E toxin reacted with normal serum to form a precipitate at 4 C but not at 37 C. Filtration through a Selas 015 filter and storage in a freezer had no significant effect on the titer of crude type B toxin. Crude type A botulinus toxin was adsorbed by a strong cationic exchanger, Dowex SO-XB (H/), and a toxic frac- tion was eluted from the exchanger with phosphate buffer at pH 7.0. - 51 - APPENDIX Buffers Used in These_§xperigent§ l. Phosphate buffer - pH 7.0 39 m1 of 0.43 monobasic sodium phOSphate (NaH2P04-H20) 61 ml of 0.4M dibasic sodium phosphate (NaQHPO4) 2. Phosphate buffer - pH 7.0 39 ml of 0.2M monobasic sodium phosphate (NaH2P04-H20) 61 ml of 0.2g dibasic sodium phosphate (NagHP04) 3. Acetate buffer - pH 4.1 35 ml of 0.2! acetic acid 10 ml of 0.2fl sodium acetate 4. Acetate buffer - pH 4.2 30 ml of 0.23 acetic acid 10 ml of 0.2fl sodium acetate 5. Acetate buffer - pH 4.6 10 ml of 0.25 acetic acid 10 m1 of 0.2fi sodium acetate - 52 _ REFERENCES Rosebury, Theodor and Kabat, Elvin A. Bacterial Warfare J. Immunol. 56; 7-96, 1947 Guyton, Arthur C. and MacDonald, Marshall A. Physiology of Botulinus Toxin Archives of Neurology and Psychiatry 57; 578-592, 1947 van Heyningen, W. E. "Bacterial Toxins" Blackwell Scientific Publications, Oxford, 1950, p. 15 Hallbank, Alfred M. The Removal of Botulinus Toxin from later Master Thesis, Michigan State University, 1953 Sober, Herbert A.; Kegeles, Carson and Gutter, Frederick J. Chromatography of Proteins. Frontal Analysis on a Cation Exchange Resin J. Am. Chem. Soc. 74; 2734-2740, 1952 Buehler, Henry J.; Schantz, E. J.; Lamanna, Carl The Elemental and Amino Acid Composition of Crystalline Clostridium botulinum Type A Toxin J. Biol. Chem. 169; 295-302, 1947 Jordan, E. 0. “Food Poisoning" 1917, Chica o Cited by - T0pley and Wilson Principles of Bacteriology and Immunity' Wilson, 0. S. and Miles A. A. The Williams and Wilkins 00., Baltimore, 1955, Vol. 2, p. 1816 8. 10. ll. 12. 13. 14. Van Ermengem, E. Untersuchungen uber Falle von Fleishvergiftung mit Symptomen von Botulismus Centrabl. f. Bakteriol 19; 442-444, 1896 Kempner, W. Welter Beitrag zur Lehre von der Fleischvergiftung. Das Antitoxin des Botulismus Z. Hyg. Infect. Kr. 26; 481, 1897 Cited by - "T0pley and Wilson Principles of Bacteriology and Immunity" Wilson, 3. S. and Miles A. A. The Williams and Wilkins 00., Baltimore, 1955, Vol. 2, p. 1824 Lamanna, Carl; Eklund, Henning 3.; and McElroy, Olive E. Botulinum Toxin (Type A); Including a Study of Shaking with Chloroform as a Step in the Isolation Procedure Snipe, P. T. and Sommer, H. Studies on Botulinus Toxin 3. Acid Precipitation of Botulinus Toxin J. Infect. Dis. 43; 152-160, 1928 Sommer, H. Purification of Botulinus Toxin Proc. Soc. EXptl. Biol. Med. 35; 520-521, 1937 Abrams, Adolph; Kegeles, Gerson; and Hottle, George A. The Purification of Toxin From Clostridium Botulinum Type A Leuchs, J. Beitrage sur Kenntnis des Toxins und Antitoxins des Bacillus botulinus Z. Hyg. Infect. Kr. 65; 55, 1910 Cited by - Barron, A. L. (30) a.) IE. .r.‘ Ad 15. l6. 17. 18. 19. 20. 21. 22, Bengston, I. A. Studies on Organisms Concerned as Causative Factors in Botulism U.S.P.H.S. Hyg. Lab. Bul. No. 136; 1-96, 1924 Burke, G. S. The Occurrence of Bacillus botulinus in Nature J. Bact. 4; 541-553. 1919 Jordan, Edwin 0. and Burrows, William "Textbook of Bacteriology" W. B. Saunders Co., Philadelphia, 1945, p. 538 Pfenninger, W. J. Toxico-ImmunolOgic and Serologic Relationships of Bacillus botulinus, Type C, and Bacillus parabotulinus, "Seddon'. ‘ :‘ J. Infect. Dis. 35; 347-352, 1924 Gunnison, J. B.; Cummings, J. R.; and Meyer, K. F. Clostridium botulinum Type E Proc. Soc. EXptl. Biol. Med. 35; 278-280, 1936 Hazen, E. L. A Strain of Bgcillus botulinus Not Classified as Type A, B, or C Hazen, E. L. Incitants of Human Botulism Science 87; 413-414. 1938 Dolman, C. E. and Kerr, D. E. Botulism in Canada with a Report of a Type E Outbreak at Nanaimo, B. 0. Can. J. Pub. Health 38; 48-57, 1947 23. 24. 27. 28. 29. -55- Geiger, J. C. An Outbreak of Botulism JAMA 117; 22. 1941 Dolman, C. E.; Chang, H.; Kerr, D. E.; and Shearer, A. R. Fish-borne and Type E Botulism; Two Cases Due to Home- Pickled Herring Can. J. Pub. Health 41; 215-229, 1950 Prevot, A. R. and Huet, M. Existence en France du Botulisme Human d'Origine Piscaire et de Clostridium botulinum E . Bul. Acad. Natl. Med. (Paris) 135; 432-435, 1951 Meyer, K. F. and Eddie, B. Perspectives Concerning Botulism Z. Hyg. Infektionskrankh 133; 255-263, 1951 Sakaguchi, GenJi; Tohyama, Yuzo; Saito, Seuchiro; Fujisawa, Soichi; and Wada, Akira An Outbreak of Type E Botulism in Akita Prefecture Due to Gilthead-Izushi Jap. J. Med. Sci. Biol. 7; 539-546, 1954 Rice, C. E.; Pallister, E. F.; Smith, L. C.; and Reed, G. B. Clostridium botulinum Type A Toxoids Can. J. Res. 25E; 175-180, 1947 N156: 0.; Hottle, G. A.; Coriell, I. L.; Rosenwald, A. 8,; and Beveridge, G. W. Studies on Botulinum Toxoid Types A and B I. Production of Alum Precipitated Toxoid Types A and B J. Immunol. 55; 245-254, 1947 Barron, Almen L. Clostridium botulinum Type E Toxin and Toxoid Doctoral Thesis, Queen’s University, 1953 31. 32. 33. 34. 35. :36. 37. - 56 - Barron, A. L. and Reed, G. B. Clostridium botulinum Type E Toxin and Toxoid Can. J. Microbiol. 1; 108-117, 1954 Sterne, M. and Wentzel, L. M. A New Method for the Large-Scale Production of High- Titre Botulinum Formol-Toxoid Types C and D J. Immunol. 65; 175-183, 1950 Gordon, M.; Flock, J. A.; Yarinsky, A.; and Duff, J. T. Purified Clostridium botulinum Type E Toxin and Toxoid Bact. Proc., p. 90, 195 Duff, J. T.; Wright, G. G.; and Yarinsky, A. Activation of Clostridium botulinum Type E Toxin by Trypsin Bact. Proc., p. 89, 1956 Sakaguchi, Genji and Tohyama, Yuzo Studies on the Toxin Production of Clostridium botulinum Type E I. A Strain of Genus Clostridium Having the Action to Promote Type E Botulinual Production in a Mixed Culture Jap. J. Med. Sci. Biol. 8; 247-253, 1955 Sakaguchi, Genji and Tohyama, Yuzo Studies on the Toxin Production of Clostridium botulinum Type E II. The Mode of Action of the Contaminant Organisms to Promote Toxin Production of Type E Orgamisms Personal Communication Kagenori Matano The National Institute of Health, Japan 38. 39. 41. 42. 43. 44. 45” -57— Muller, Robert H. Application of Ion-Exchange Resins to the Purification of Certain Viruses Proc. Soc. Exptl. Biol. Med. 73; 239-241, 1950 LoGrippo, Gerald A. Partial Purification with an Anion Exchange Resin Proc. Soc. Exptl. Biol. Med. 74; 208-211, 1950 Muller, Robert H. and Rose, Harry M. Concentration of Influenza Virus (PR-8 Strain) by a Cation Exchange Resin Proc. Soc. Exptl. Biol. Med. 80; 27-29, 1952 Takemoto, Kenneth Use of a Cation Exchange Resin for Isolation of Influenza Virus Proc. Soc. EXptl. Biol. Med. 85; 670-672, 1954 Reed, L. J. and Muench, H. A Simple Method of Estimating Fifty Percent Endpoints Am. J. Hyg. 27; 493‘497. 1938 Boor, Alden K.; Tresselt, Hugh B.; and Schantz, Edward J. Effects of Salts and Colloids on Potency of Botulinum Toxin Proc. Soc. EXptl. Biol. Med. 89; 270-272, 1955 Kunkel, H. G. Concerning an Electrophoresis Apparatus to Separate the Components of a Mixture Sci. Amer. 193; 92-98, 1955 Block, Richard J.; Durrum, Emmett L.; and Zweig, Gunter "A Manual of Paper Chromatography and Paper Electro- phoresis" ‘ Academic Press, Inc., New York, 1955. p. 406 46. 47. 48. - 58 - Lewis, Keith H. and Hill, Edwin V. Practical Media and Control Measures for Producing Highly Toxic Cultures of Clostridium botulinum, Type A J. Bact. 53; 213-230, 1947 Cohn, Waldo E The Separation of Biochemically Important Substances by Ion-Exchange Chromatography Annals N. Y. Acad. Sci. 57; Art. 3, 204-224, 1953 Lamanna, Carl and Mallette, M. Frank "Basic Bacteriology" The Williams & Wilkins Co., Baltimore, 1953. p. 188 Heming, nrthur E. and Flanagan, Thomas L. Considerations in the Selection of Cation Excharge Resins for Therapeutic Use Annals K. Y. Acad. Sci. 57; 239-251, 1953 - r i 1;er 't g ' _/" ‘ i, r .-/ "Willis/ill“Mil/1111'“