'I'! .5 g . .‘ "oq mum” “mu .0 v 1. (Vi. 4 . n.1,»! ”Mann. la “in". ..- an...” an... in“ _ . . ”qty. Clow' . “EN” 5 an. mm“. «HE. A»? ..:..:.. pk... “93 it. “$.40?“ do} ”I! fit any flu ...:.. 3m 1.-.. ”a! .,u am»? ”‘.|. f.‘ ‘0‘ .008. A r i ll. .3“ I352 INN-.5 LL 4 «1.1!! pH .35 in. 3?. a it A IV "O‘h. .‘ab I” ‘&"O ofli‘ F‘ v. ‘1 v.51 o...l.t.(. If!- C. I“ "". Q“ ”.0--. J .03... y , 7| when“ i‘m‘ u . bl. .I r. O .3. .l. a... “a“ I :‘V 4.. l o Haw-3112 EIIJDIRG CF Vl’i‘AI-lllfi 1312 1 Joan C. Dorris A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of EASTER OF SCIEFCE Department of Food Science 1967 AC RIC Us‘J'LEDG EEK; T 3 To Dr. B. S. Schweigert, Professor and Chairman of the Department of Food Science, the author expresses his most sincere appreciation for guidance, encouragement and stimulation so willingly given.and gratefully accepted. The author also expresses his gratitude to Dr. J. Robert Brunner, Professor of Food Science, for his advice and counsel throughout this research. Graterl acknowledgement is due the Department of Food Science, Michigan State University and the Public Health Service for the facilities and funds which made this research possible. The author also expresses his appreciation to his wife for her help and encouragement. ii ABSTRACT Vitamin B12 must be bound to intrinsic factor before it can be effective in the oral treatment of pernicious anemia. Intrinsic factor is a mucoprotein, and its chemistry has not been completely defined. There are many proteins that bind vitamin B12 but have no intrinsic factor activity. It appeared that the most likely course to take in an effort to elucidate the protein—vitamin B12 binding mechanism.was to study the binding of vitamin 812 by a purified protein. The casein fraction contains about half of the vitamin 312 in milk; however, the vitamin 312 content of casein is much lower than that of the whey proteins on a weight basis, as casein makes up a larger part of the total milk protein. Skim milk was fractionated to give sodium.caseinate, para-kappa complex and fractions that were soluble and insoluble in 10% trichloroacetic acid. No one of these particular fractions contained the predominance of the vitamin B12 in casein. At this point, it was considered more worthwhile to concentrate on one of the whey proteins that contains higher vitamin B12 levels. Electrophoreticaljy homogeneous fi-lactoglobulin was prepared. iii Vitamin B12 content was determined to be about 160,414: per mg of protein. Vitamin L12 binding capacity was found to be approximately 1800].!ng per mg of protein by the use of 0060 labeled vitamin 13:12. When the vitamin 1312 binding capacity of a sample of fi-lactoglo- bulin that had previously been electrodialyzed was determined it was found to have been reduced to about 800% per mg of protein or to about 45% of its original value. Ebccess 0060 —vii‘amin 1332 was added to a sample of fi-lacto- globulin prior to electrodialysis. Then after the sample had been electrodialyzed its vitamin 1312 content was found to be approndmately lBOOflFg per mg of protein. The addition of vitamin 1912 prior to electrodialysis appears to have protected the protein from some of the effects of electro- dialysis. Further evidence of a protective effect by vitamin 812 was seen on U.V. monitoring of the effluent from a P—2 polyacryla— mide bead column when these different preparations were passed over the column. When fi-lactoglobulin was passed through the column, only one peak appeared on the chart paper. When fi—lactoglobulin was electrodialyzed and the material from the electrodialysis sac was passed through the column a second peak appeared. It came off the column much later, indicating a low molecular weight compound. When excess vitamin B12 was added to B—lactoglobulin, iv electrodialysis carried out in the same manner as before and again, the contents from the electrodialysis sac were passed through the column only'one peak was noted. This gives very strong evidence that electrodialysis splits off a peptide or peptides that are necessary for binding vitamin 312, but that if vitamin 312 is bound, to capacity, by the protein prior to electrodialysis, it somehow prevents the peptide from being cleaved. Further work is needed to establish the mechanism.of binding. It will be helpful to determine if a single "specific" peptide or a mixture of peptides is being cleaved, and to determine the nature and chemical characteristics of the peptide or peptides involved. I" ' T ’7‘ C“?- !1‘ 'f‘l"_‘~‘> ["10 .L‘LJLX- J i- ‘v‘; A-JA‘.‘.U .Liorcv‘i-C—JU( TJ- (31-, o o o o o o o o o o o o o o o I o I7 LVIL 7-1! C1 LIJTJ‘L'PUIL . . . . . . . . . . . . . HiStOry. o o o o o o o o o I o o o o o 0 Source 0 o o a o o o o o o o o o o o o o Stpd‘the O O 0 O O 0 I 0 O O I O O O O 0 Forms 0 o I o o o a o o o o o o o o o o o :1zction o o o o o o o o o o o a o o o o tability . . . o o o n o o o o o o o o o BOUHd Form 0 o o o o 0 o o c o o o o o o 1' I ilk Iiilk protein separation methods . . . ElOOd o o o o o o o o o o o o o o o o o o vary -fl -‘~ --~_- 'T“ __'\YY’""1" Shut. :.1-».$::J.Jl H411 Ilir‘ \ILkJ .JJLilm—J . . . . O Q ’Ihe l-icrobiological Assay for Vitamin Assay organism . . . . . . . . . . . Ledia 1.............. Inoculum . . . . . . . . . . . . . . . llethod of determination of vitamin 13-19 Standard. solution . . . . . . . . . Sav.ple solution . . . . . . . . . . Standard curve . . . . . . . . . . . Total vitamin 15112 . . . . . . . . . Preparation of lsoelectric Casein . . . Preparation of Para—Lappa Complex . . . Preparation of TCA Fractions . . . . . Preparation of fl3-Lactoglobulin . . . . Criteria of Purity of fi—Iactoglobulin Vitamin 1312 Eindin g Characteristics of 3- -L.'1<:to ul 0 O Freevitaminlijg........... O O Fitrogen content. . . . . . . . . Polyacmrlamide gel electrophoresis . . 1n 0 O O I O O O O O O I O O O I I I ‘ Content . . . . . . . . . . . . . . . Binding; Capac“ . . . . . Binding Capacitv after Electrodialysis Flectrodialye is of fi-lactoglobulin Content after Electrodialysis of a Sample Contain- ing 1-1xcess COCO Labeled Jitamin 112 , vi 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5;] ob . . . . . . . . . ’3! 9 Cl }J N ‘c '1 (‘1 kan «’2' V7 I 17 17 l7 l7 17 "I cl ”V Q k1 18 19 19 2C 20 G L 23 23 2‘; I 26 / 26 29 29 XO (3 m Luf1nlnu11a1 RESULTS . . . . . . Sodium Caseinate . . . . . Para—Rappa Complex . . . . TCA Soluble Fraction . . . TCA Insoluble Fraction . . Recovery . . . Vitamin B12 Binding Characteristics blllin o a o I o o o o o 0 Content . . . . . . E12 Capacity . . . )3lLactoglo— binding Capacity After Electrodialysis L12 Content After Electrodialysis of a Sample Con- taining Excess C060 Vitamin B12 . . . DISCUSSION AED CONCLUSICNS , , . ‘7‘!"’I")' 7"1' .|'T1'7“1 "‘ AMA—4‘ .1 1““: C .JL)‘ \JJ. LL...) . . . . . . . . Table .TJST 0:? cranes m momma; Vitamin 312 distribution in casein proteins . . . . ;9 Vitamin 1312 content of fi-lectoglobulin following various treatments . . . . . , . . . . . . , , , . 3? Pi gure \7 Vitamin 2312 (c;.ranocobalamin) . . . . . . . . . . . A schematic diagram show-ring; the fractionation of cow's on milk into the casein fractions studied . . . . , . m, A schematic diagrml show-ring the procedure used to obtain electrophomtically homogeneous fi-Jectoglo’o— Ulin used in this study,r . . . . . . . . . . , . . . 2"; motographs of the polyacrylamide gel electrophoreto- grams of various cuts of fi—lactoglo‘oulin . . . . Ellution chrormtograms of fi-lactoglobuljm when treated as indicated . . . . . . . , , , , , , , . b2 Il‘ETELODUCTlCE-T Vitamin 312 is an essential nutrient for man and many animals. The oral administration of vitamin B12 along with a source of gastric intrinsic factor relieves the symptoms of pernicious anemia in man. Pernicious anemia is a macroqytic anemia in which the red blood cells are abnormally large but few in number. In order for vitamin £12 to be effective in the oral treat- ment or prevention of pernicious anemia it must be bound to intrinsic factor. Intrinsic factor is a mucoprotein which con- tains approximately 7% reducing sugars and is secreted by the glands of the stomach walls. The best evidence indicates that intrinsic factor has a molecular weight of around 53,003. In— trinsic factor is by no means the only proteinaceous material which binds vitamin B12. Vitamin 312 binding substances occur in blood, tears, milk, colostrum, cerebrospinal fluid and urine, which have no detectable intrinsic factor activity. A variety of methods has been applied to the resolution of materials that bind vitamin D12. Since these materials are pro— teins they are amenable to the techniques commonly applied for protein resolution. The method of molecular sieving by gel filtra- tion on Sephadex or on polyacrylamide bead columns has been of great value in the determination of the vitamin E12 binding capacity of proteins or peptides, in that it can be used to separate the proteins or peptides with their bound vitamin L12 from free or unbound vitamin B12. Vitamin B12 is involved in a variety of metabolic trans- formations. Among these are the interconversion of succinyl coenzyme A and methylmalonyl coenzyme A, the interconversion of glutamic acid and fi-methylaspartic acid, the dismutation of vicinal diols to the correspondinv aldehydes, the methylation of homocysteine to form methionine and the methylation of deoxye uridylic acid to fol- thymidylic acid. lie detailed chemical mechanisms of many of these transformations remain obscure. Vitamin 212 can always be traced to a microbial origin. The best evidence indicates that there is no vitamin B12 in higher plants and that it is not synthesized by animal tissues. Liver and other organ.meats are exceptionally rich sources of the vitamin. The first crystalline vitamin 312 was isolated from commercial liver extracts. Liver extracts were long used as a source of the vitamin. The purpose of the first part of this study was to isolate the casein fraction of milk and resolve it into components in an effort to determine if any one of these various protein fractions showed a significant variance from the others in vitamin £12 con- tent. if a significant difference had been found an effort would have been made to find the reason for such a difference. Considerable work has been done in an effort to elucidate the mechanism of binding of vitamin 312 by proteins. Earlier work in this laboratory showed that electrodialysis of skim milk removes peptides which have a high vitamin 1‘12 binding capacity. It was thought worthx-rhile, therefore, in the second part of this study to apply the same procedure to a purified milk protein ( fi—lactoglobuljn) and see if peptides are removed by electro- dialysis, and if so, what effect this removal of peptides has on the vitamin {‘12 binding property of the fi—lactoglobulin. These studies were designed to provide further information on the nature and extent of binding of vitamin £12 by milk proteins. Information from these studies could be applicable to other protein- vitamin 812 binding systems including the intrinsic factor. REVIEX OF IITBRATCIE Historz The clinical recognition of pernicious anemia over one hundred years ago marked the beginning of the study of vitamin B12 (1). Pernicious anemia is a macrocytic anemia in which there are too few red blood cells and those present are abnormally large. The bone marrow is megaloblastic. The blood forming cells become enlarged'while still immature. In 1920 Whipple (93) found that feeding liver accelerated the regeneration of red blood cells in dogs made anemic by bleeding. Hinot and Murphy (67) in 1926 tested liver for the treatment of pernicious anemia. They found that a remarkable improvement re— sulted. Similar results were obtained when liver extracts were administered (91). Injection of liver extracts was widely practiced prior to the isolation of vitamin 312. Cohn (l7) began his attempts to extract the anti-pernicious anemia factor from liver in 1928. In the same year Castle (14) postulated that the atrophied stomach glands, associated with ach- lorhydria in pernicious anemia patients might be failing to secrete same essential substance in digestive juice which he called "in- trinsic factor." This "intrinsic factor,“ he concluded, could be necessary for the absorption of an hextrinsic factor" present in food (now known to be vitamin £12), and could yield the liver factor as the product of the reaction (8). Stokstad §t_al. (84) suggested that the animal protein growth factor required by chicks and other animals and the antipernicious anemia factor are identical. Vitamin 512, or extrinsic factor, was finally isolated in crystalline form from commercial liver extracts in l9h8, almost simultaneously by two independent teams (71, 82), Lerck and Co., Inc. in the U.S.A. and Glaxo laboratory in England. Source Vitamin 512 is synthesized by certain microorganisms (27, 62). The best evidence indicates that there is no 312 in higher plants. Ruminants require no dietary source of vi '- 312, as microorganisms in the rumen produce an excess of this vitamin, provided the diet contains sufficient cobalt. Animal tissue is a good source of the vitamin as it is retained by the organism. Liver and other organ meats are exceptionally rich sources of vitamin B12. Three proactinomycetes of the genus hocardia produce a large amount of vitamin 812 (27) and organisms from this group are used for the commercial production of the vitamin. Recommended for the commercial production of vitamin 312 are Streptomyces griseus, g, olivaceous, Bacillus megatherium and propionic bacteria. Yields as high as 3 ’Lg of vitamin B12 per ml of fermentation liquor have been reported (62). Structure Cyanocobalamin is a red, crystalline complex coordination compound containing trivalent cobalt and a cyano group. Degrada- tion studies revealed extensive characteristics of the vitamin L12 molecule, but its exact molecular structure was not established until 1956. hodgkin and associates (50) used x—ray crystallography to definitely establish the structure of the molecule as we know it today (figure 1). The vitamin L12 molecule is the largest vitamin molecule known. It has a molecular weight of 1355. "he bulk of the molecule consists of a porphyrin—likc corrin ring system. Two of the pyrroles are linked direct rather than through a methane bridge as are the other pyrroles in the molecule and in porphyrins in general. Cobalt is in the position held by iron in the heme series. The cyanide group is joined to the cobalt. The cobalt is joined by a coordinate bond to the nitrogen atom.of a molecule of 5,6-dimethylbenzimidazole which in turn is in glycosidic linkage to ribose-B-phosphate. The ribose-B—phosphate is esterified to aminoisoproganol. The nitrogen atom of the aminoisopropanol molecule is in amide linkage with he carboxyl group of a propionic acid molecule substituted in the corrin ring. CHZOH o O\\ H m CH __ 52 {:61 cot-12 co / t... CH2 CH3 ”22 13 CH CC- ~12 H2 / 3 / C 2 ' CaNHg Ii CHZCHZCCZZHZ Fig. 1. Vitamin 312 (cyanocobalamin) here are several naturallyr occurring; vitamin $12 derivatives in which the cyanide group is replaced by other groups (2, 81). L *anide treatment used in the microbiological assay converts all the derivatives to cyanocobalatrjn. All of these derivatives show biological. acti'n'tj,r in humans. azfmmls and microorganisms. L-gany vitamin 1212 analogues in which the 5,6-dianethylbenzimid— azole is replaced by a purine or pyrimidine or other related group have been isolated and tested (Bl). These analogues are generall;,r less active than cyanocobalamin or completely inactive, depending on the test organism. The cobamide coenzyme differs structurally from cyanocobalamin in that the cyanide group of cyano— cobalanin is replaced by a molecrle of deoygl'adenosine. Function rEhe cobamide coenzgane is active in a series of reactions of biological importance (20, 913/). Among those known are the inter- conversion oi‘ succian-rl coenzyr-e A and methyjmalonyl coenzyme A, the fonwtion of fi—methylaspartic acid. from glutamic acid, the disz‘nutation of vicirrial diols, the methylation of I'loztwocysteine to fem methionine and the 111ethylation of deoxgdlridylic acid to form hpfldylic acid. The exact role which the cobamide coenzgme plays in each of these varied aspects of metabolism is not clear. \f) * « Aqueous solutions of cyanocobalamin are most stable between pH a and 6. Within this range autoclaving at 121 C causes a loss of only a small percentage of activitv. Aquocobalamin is less stable than cyanocobalamin, but both are about 905 inactivated by heating for one hour at 100 C at ph 8 (30). Ch exposure to light cy nide is split off and hydroxocobalamin is formed. This change is reversed if the solution is ke t in the dark (39). Irreversible P destruction results from prolonged exposure to sunlight. bound Form Vitamin L12 occurs predominantly in a conjurated form in which it is bound to protein (32). The binding of 312 by intrinsic factor has been intensely investigated (85). hicrobiological growth (12, 22, 43, 52, 53, 80), microbiological absorption (23, 24), charcoal absorption (66), dialysis (8, 72), ultrafiltration (36), paper chromatography (Q9), zone electrophoresis (33, 60), column chromatography (26), gel filtration (21, 5%), and competitive binding; of vitamin 1:12 and €060 labeled vitamin :12 (77), have been used to determine B12 binding capacity of intrinsic factor and other binding materials. Lexy substances and body fluids have been found.which have a significant :12 binding capacity and yet exhibit no intrinsic factor activity. 10 The binding of vitamin 312 (referred to hereafter as 212) depends on many factors including pH and temperature of the medium, ratio of the concentration of 312 to 312 binding material, co- existence of inhibitory materials in the system which compete for ElZ binding, nature of the 312 binding material itself and the technique used to measure the £12 binding. Storage of vitamin 312 in liver has been demonstrated (79). Hedbom (#6, 47, 48) isolated and characterized a native cobalamin—polypeptide complex from liver. Other organ meats have been shown to have a high level of £12 (78). Rosenthal (73, 75) has resolved multiple cyano— cobalamin—binding components from the sera of various animal species. The L12 binding capacity of human serum has been re- ported (51, 74). Gizis gt. El- (31) have demonstrated the binding of vitamin B12 by electrodialyzable peptides in cow's milk and have measured the binding capacity of these peptides. Milk The vitamin B12 activity of milk appears to be almost en- tirely due to cobalamin. In early studies, Gregory (3Q) assayed cow's milk, and also the milk of several other Species, by three different assay organisms - 1. Ochromonas malhamensis, which is rather Specific for cobalamin or other forms of the vitamin ef- fective in animals and 2. .§~ coli and 3. lactobacillus leichmannii both of which respond to naturally occurring £12 analogues which are inactive in animals, as well as to vitamin L12. With each milk studied, all three organisms yielded approximately the same value for the 512 content, indicating that most of the vitamin was present as cobalamin or other fonms of the vitamin active in animals. Ion (58) has pointed out that even though E12 analogues predominate over cobalamin in the alimentary tract of ruminants and are presen in certain fermented fodders and silage, they occur in.milk and in animal tissues only in traces; and, since cow's urine was found to contain a larger proportion of such £12 analogues than of cobalamin, it seems that any of the B12 analogues absorbed are preferentially excreted. The milk of certain species contains peptides and proteins that bind cyanocobalanin and make it unavailable for the test organisms (39, 41). Lilk has been shown to have considerable :12 binding capacity (35, 37, 39, MO), and practically all the E12 contained in it is in the bound form; however, the protein binding zmaterials contained in milk have no intrinsic factor activity (37). On ultrafiltration followed by L. leichmannii assay sow's milk ex— hibited the highest 312 binding, followed by rat milk and human milk (38 . The lowest 312 binding capacity of the milks tested was that of the goat and cow (38). Gregory and holdsworth isolated a purified cyanocobalamin- protein complex from sow's milk by fractional precipitation with ammonium.sulfate followed'by continuous electrophoresis to separate the bulk of the proteins and fractional reprecipitation with iso- propanol (35, 37, 39, #0). This ElZ—protein complex was pink in color, homogeneous on electrophoresis at pH 6.5 and 8.6 and con- tained.23.6»,Lg £12 per mg of protein (35, 37, 39, 40). It had a mobility identical to that of the hog intrinsic factor concentrate these authors had purified, but it had no intrisic factor activity. Host of the values reported for the vitamin £12 content of cow's milk have been determined by microbiological assay, mainly with a strain of L. leichmannii or E. coli. Some bioassays are based on the growth of'weanling rats, rendered vitamin BlZ—deficient by feeding their mothers a vitamin Blz-deficient ration during lactation (25, 45). Values by the rat growth method of assay are comparable to the highest values obtained by microbial assay. him at al. (55) reported values of 5.25 ’L; of 312 per liter of whole raw milk and 5.37 ,Lg per liter of whole pasteurized milk. They also reported that the casein and whey fractions of milk contain about equal percentages of the 512 in milk, but'when 312 content was related to protein content they reported approximately three times as much £12 per milligram of protein in the whey fraction as in the casein fraction. Investigators have published contradicting data on the levels of vitamin £12 in the milk of cows on pasture compared to the milk of cows on dry feed (#2, 83), 'fiorkers similarly disagree on 312 levels in milk during summer (55), as compared to levels in spring (8e). Cther reports (45) state that the 312 content of the milk of cows on pasture feeding is about the same as on barn feeding and that no seasonal variation is noted. Hartman (45) found no significant difference between the vitamin £12 content of Holstein and Jersey milk. Collins and co— workers (18, 19) observed no difference in vitamin 512 content of the milk from the Holstein, Jersey, Guernsey, Ayrshire and Brown Swiss breeds. Similar findings were reported by Gregory 21:. a. <42). The vitamin B12 content of cow’s colostrum is greater than that of the milk later in lactation. From averages of values obtained br several workers, the concentration of B12 in samples of early colostrum.appears to be about three to six times that in milk obtained later in lactation. During the first few days post-partum, the level falls rapidly (3, 19, 42), reaching that of normal milk about the second week of lactation (42) or earlier (19). Pasteurization causes only slight destruction of the vitamin £12 in milk. The content, as determined by microbiological assay, is diminished by about 10% by the holding method (61 to 65 C for thirty minutes) (57, 58) and about the same or somewhat less (15, 57, 58) by the high-temperature—short—time process (71 to 7b C for fifteen to sixteen seconds) and by the flash method (85 C for ap— proximately three seconds) (57). More drastic heat treatment, however, may lead to considerable or even nearly complete destruction 111 of the vitamin (28, 83). Tests made by workers in England (29) of samples of milk sub- jected to ultraviolet irradiation showed that no significant loss of vitamin 13 occurred. 12 111k protein separation methods. Many techniques are available for the fractionation of the various milk protein components. The major protein in milk is casein, a phosphoprotein. It is pre— cipitated by acidification of milk to p11 11.6. Casein has been resolved into several fractions by fractional precipitation with acid. These fractions are designated 0-, fi— and Y—casein (92). a -Casein is the principal component of casein comprising; about 75,”? of the total (33). Another method for the precipitation of casein from milk, at neutral pH, is by the action of the enzyme rennin which is produced in the gastric mucosa. The coagulation of casein by he action of rennin has been described as a conversion of casein into paracascin (16). This coagulation depends on the presence of calcium ions and seems to be caused by a proteolytic action of rennin (90). After casein is removed from milk by acidification and filtra- tion, 1).-Io additional proteins, lactalbumin and fi—lactoglobulin, can be isolated from the filtrate (acid whey), by precipitation with ammonium sulfate. fi-Zactoglobullin is insoluble at low ionic strength and can be obtained in crystalline form when the redissolved 15 protein precipitate is dialyzed against dilute acid (69). Iactal— bumin is obtained.by'salting out from the supernatant solution (4). Acetone precipitation (9) has been reported to be an aid in the preparation of a-lactalbumin. In addition to lactalbumin and lactoglobulin, whey contains a red protein (44). Brunner and Thompson (11) have reviewed and studied some characteristics of the minor proteins in milk. Blood hany investigators have studied the problems of binding of vitamin 312 by human serum.protein fractions and how vitamin 312 is distributed in biological material (6, 7, 6', 66, 68, 70). E. hoff—Jérgensen at al. (51) showed a binding capacity of 1350 ’Lng of vitamin 312 per ml of human serum. In a different type binding experiment (74) an average recovery of 1005 of added 312 was reported when 500, 250 or 100 ’L’Lg of vitamin L12 were bound by one m1 of serum. Rosenthal §t_al. (75) have done considerable 'work on the resolution of cyanocobalamin-binding components in various animal sera. All the sera tested showed two major cyano— cobalaminrbinding components. It had been previously suggested that human serum may contain more than one binding site for cyano- cobalamin (7, 76). The difficulties of isolation of the B12 binding substances in human serum become apparent when the low concentration of these substances in serum is considered. This has been estimated H OX to be in the order of 0.1 ng of binding substance in one ml of serum (6b). In comparison, the $12 binding capacity of human serum ranges from 0.1 to l m P. g per ml (32), of human milk about 80 m ,1, g per ml (33) of cow's milk approximately 0. 5 m P, g per ml (33) and of cow's colostrum: 5.8 mpg per ml (38). In another study (73) it was found that cyanocobalamin binding by a protein makes that protein less susceptible to urea denaturation, suggesting that bound cyanocobalamin may bridge two or more protein chains, thus preventing unfolding of the protein molecule under the influ- ence of urea . *“r —- ‘E’.“"‘ j-fi'nj 1‘ -r T :3 m-f.“ 3 yrs L'leL‘thli'; 51:15.1.) fl ‘.OV;.AUL LLL‘J The Ificrobiological Assay for Vitamin 319 A modification of Gregory's method (34) along: with the method described in U.S.P. (86) and A.C.A.C. (5) was used, Lissa},r organism. Iactobacillus leichmannii ATCC 7830 was used. liedia. Standard official media (Lacto—E’LZ culture agar, Bacto-ifiz inoculum broth and Lacto-L‘lz assay medium) were used. Inoculum. The assay organism was carried in I‘acto-Elz agar. Transfer was made every two weeks and an incubation period of 1+8 hours at 37 C plus or minus 0.5 C was used. On at least three successive days prior to assay the culture was transferred from broth to broth and incubated for 16 to 21+ hours after each transfer. At the time of assembly of the microbiological assay, the culture was washed at least three times with 10—15 m1 portions of 0.9;} saline. rE‘he washing was accomplished by centrifugation and sub- sequent decantation. The washed cells were suspended in 50 ml of saline solution. One drop of cell suspension was used to inoculate each assay tube . I-lethod of detennfmation of vitamin F12. All procedures were carried out under minimum exposure to light. Flasks and beakers were shrouded or actinic red lab-rare was used. Standard solution. One mg of vitamin 312 (cyanocobalamin) from a standard ampule was diluted to contain either 25 or 50 micromicrograms ([Lngfl of vitamin B12 per ml of'working standard solution. Sample solution. Sample material was dissolved in water, the pH adjusted to 4.5, a few drops of cyanide solution were added, samples were autoclaved at 15 pounds per square inch pressure (psi) for 6 minutes, the pH readjusted to about 6.8 and final dilutions were made. Standard curve. Standard solution was added to the standard size (16 mm 1.3. by 1&5 mm long) test tubes at 0.5 ml intervals, in triplicate, ranging from 0-5 ml of standard solution. Total volume was brought to 10 ml'with 5 ml of medium and a sufficient amount of water. Tubes were autoclaved at 15 psi for 4-5 minutes. After the tubes were cooled one drop of cell suspension was added to each tube. The tubes were incubated for 72 hours at 37 C plus or minus 0.5 C. The lactic acid produced was titrated with a Fisher Titrimeter. A standard curve was prepared by plotting titration values (in m1 of C.lh alkali solution) at each level of standard vitamin 312 solution used versus the amount of standard used. Total vitamin 312. Total vitamin 312 was determined in samples by the same procedure used for the standard curve. Appropriate dilu— w "J tions were made so as to bring the r12 content of the diluted sample within the range of the standard curve. Sample tubes consisted of duplicates at levels of l, 2 and 3 ml of diluted sample at each dilution level. Cnly those dilutions with titration values within the range of the standard curve were used. By reference of the titration value of sample solution to the standard curve, the vitamin 1312 content of the dilution assayed was read from the curve. The 112 content of the sample was obtained by multiplying the value obtained by the appropriate dilution factor. Free Vitamin 312. Samples for free vitamin B12 detenrination were obtained by passing the samples through’a Seitz filter and analyzing; the filtrate for vitamin £12. I‘rgparation of Isoelcctric Casein Fresh raw whole milk was obtained from the University Holstein Dairy Earn. While still ram the milk was separated with a cream separator. The cream was discarded. lwo volumes of skim milk and one volume of water were thoroughly mixed. The diluted skim milk was 1i>rought to p11 14-. 5 with 0.11{ HCI solution. The casein precipitate was obtained by filtering through cheese—cloth. r"he whey fraction was discarded. rShe casein was washed three times and recovered after each washing by filtering through cheese—cloth. The washed casein was brought to one third of the original volume of milk used, (‘0 1.) 'by the addition of water, and tLe pH was adjusted to 7 with 0.1T IaCH solution. This yielded a solution of sodium caseinate. Preparation of Fara-Zappa Complex The isoelectric casein solution was dialyzed to remove the excess ions which had been added for precipitation and sulsequent neutralization. One ml of rennin solution (O.h5 mg crystallized rennin per ml) was added to each 100 ml of neutralized, dialyzed casein solution. The casein-rennin solution was held in a 37 C water bath for one hour. Calcium chloride was added to give a concentration of 0.25 H. Centrifuging at 1000 xg for 30 minutes in the Servall yielded the para-kappa complex as a white, waxy solid material. The para-kappa complex was dissolved in 5 molar urea and dialyzed. {reparation of TEA Fraction . The supernatant solution, obtained when tie para—kappa com— plcx was removed by centrifuga;ion, was used for this preparation. J. ,3 Crystalline trichloroacetic acid (TCA) was added slowly and with stirring until a final concentration of 10; TCA was reached. The protein fraction which was insoluble in 165 TCi solution was re— moved by centrifuging at ICOC xg for 30 minutes in the Servall centrifuge. The insoluble material was put in water and slowly neutralized with I faCH solution with stirring. The neutralized solution was dialgrzed for 1+8 hours with C changes of water to remove excess salts. This is called the TCA insoluble fraction. The supernatant, obtained when the 10,3 ICE; 'nsolulzle fraction was removed by centrifugation, t-Ias neutralized with n IEaCPEI and dialg‘zed for MC- hours to remove excess salts. his is called TCA soluble fraction. A summry of the techniques employed is presented schematicalL' in Figure 2. 1‘ Preparation 0.1. fi-Iaetoglo‘euliaq fi—lactoglo‘culin was prepared by the method. of larson and Jenness (59). Freshl;r drazam milk was obtained from the University Jersey Herd. Iafhile still warm, the milk was separated using a DeLaval Cream Separator and the cream I-ras discarded. :hruuonium sulfate was added to the skim milk to bring the concentration to 50;; saturation. The proteins precipitated at this concentration were removed by filtration and discarded. The filtrate I-ras brought to e0}? saturation by further addition of ammonium sulfate. The precipitate was removed by filtration and dissolved in an equal volume of deionized water. This solution was dialyzed in Visking cellulose dialysis tubing: against frequent changes of deionized water. The p11 was maintained at 5.17 by the addition of l.‘ H3 . After one week no crystallization had occurred, even though the preparation was seeded with a commercial preparation of ’7’) LL. ‘igure 2. A schematic diagram showing the fractionation of cow's milk into the casein fractions studied. '21 UK ICLE RAH SILK Separate I Cream (Discard) Skim Kilk Add % Volume of £20 Add 0.1K HCl to pH 4.5 f whey (Discard) Isoelectric Casein Hash 3X Dilute to 1/3 Original Volume Add 0.1E haCh to pH 7 Sodium Caseinate DialyZe Add Rennin Add CaClZ to 0.25 h Centrifuge Supernatant Para-Kappa Complex Add TCA to 10$ Dissolve in 6H Centrifuge Urea Dialyze .- L . N o rara dappa bolution .1 In J - Tb} soluble TCA soluble Neutralize éfld REC. Dialyze neutralize Dialyze ”CE Soluble Solution TCA Insoluble Solution I") K A) fl-lactoglohrlin crystals. however, a significant amount of an amorphous "oil" separated and settled to the bottom of the dialysis sac. This "oil" was separated from the supernatant and dried by lyophilization. The 12 grams of dried protein obtained were dis- solved in 1200 ml of deionized x-rater containing a trace of aztmlonimu sulfate to facilitate solution. Salting-out with 84711301111111“; sulfate was employed to obtain a fraction soluble in 50]; but insoluble in 80}: saturated ammonium sulfate solution. This precipitate was dis- solved in an equal volume of deionized water an dialvzed against fr-quent changes of deionized water at C-l'r C for one week. Again, p11 was maintained at 5.17 until an "oiled-off" bottom layer appeared. This fraction was collected and lyophilized. The supernatant fraction was also lyophilized. The procedures used are surrarized in figure 3. Eectrophoresis in polvacmrlamide gel showed that the supernatant contained an electrophoretieallq,r homo-- geneous protein, identified tentatively,r as fi—lactoglobulin (Figure 19), I-rhich was used for further studies. Criteria of Purity of fi-lactoglotiflin ITitrog;en content. lijeldahl nitrogen determinations gave values of 15.3 and 15.5.3 nitrogen for our fl—lactoglobulin pre- paration which compared favorably,r with a reported value of 15.6,: (59). 2L; 5213511 ~3 l}; Lgfll; 03 Sat'd with ammonium sulfate .’ Filter FlLTJATE PREClPlTATs . (Discard) €05 Sat'd'with ammonium sulfate Filter FILTRATE PRECIPITATE (Discard) Dissolve Dialyze exhaustively Filter 1011 5.17 Dialyze 01L SUPERIAIAIE Dissolve pH 5. 17 Dialyze f 7 OIL 5141’s;{$11133 (Discard) (larophilize) Figure 3. A schematic diagram showing the procedure used to obtain electrophoretically homogeneous AB—lacto- globulin used in this study. 25 Polgacr‘ larlide :el electr phoresis. The gels use J for elect- rop lore31s consisted of ’pol;,*ac1*-‘1a:.1ic.e 1n hora te buffer (pH 3.7). The rafer was prepare ed b; adding 880 grams of boric acid and 19C) Frans of soaiuml . vdioxide to sufficient water to give a final vol tune of 19 litci s of stock solution The final buffer was pre- pared by mixing one part of the stock solution with two and one half parts of water. The gel solution was prepared bv dissolving CC [grams of} yanogum-L'l in sufficient buffer to yield one liter of solution. Individual gels were prpae red L7; adding 0. 1% :11 of ..1T,l;1 Nl- tetra— meth; lethylenediamine (T.?I.Z‘3.;J.) to 250 ml of Cyanogun—L'rl solution and stirringr until- well mixed, followed by the addition of 0.1+ {gram of anionimn per-sulfate. The solution was promptly poured into a gel bed measuring 5 mm deep, 12 cm wide and 26.5 cm long, the slot former was introduced and the gel bed was placed lulder an aUHOSphCI'e of nitrogen until the solution polymerized. The slot former was removed and the sample solution (0.2 ml of 2,: protein) was introduced into the slots. Borate buffer was added to the buffer tanks tor lake a continuous buffer system, platinum electrodes were introduced into the buffer tanks and 100 milliamperes (a-pro:-:— imately lC- milliamperes per 0313, cross section) of current was applied. After 12 to 1'6 hours the gel was removed from the bed and stained with the anionic we, [mid lo lack. .‘xcess dye, not bound by protein was extracted from the gel in an electrophoretic destaining; cell- in which 7“,: acetic acid served as the electrolyte. A mild current of approxfmatel;r 1'; amperes was employed. Proteins were easily located on the destained gel as stained zones. ’lne electrophoretograms were photographically recorded. figure 1+ shows photographs of gels of fi-lactoglo‘etflin at different stages of purification. Vitarin 319 Binding; Characteristics of Q—lactoglobulin Content The vitamin 312 content of the fi-lactoglobulin prepared for these studies was detemlilled by the microbiological assay previous];r described. fl-lactogloblflin (100 milligrams) was dissolved in 10 ml of pH 9 borate buffer (5.67 g borax and 3.09 g boric acid per liter). One half milliliter of (2060 labeled vitamin i312 solution obtained from l-ierel: and Co. was added. The label on the bottle of 3060-112 solution states that it contains 1.29 micrograms of qranocobalarr‘in and 1.52 microcuries per ml. Cy microbiological assay the vitaltlin I212 content was found to be 1.332 micrograms per ml. rEhis micro- biologicallgr determined value was used for all calculations / 0 CC 0 o *3 o a r-~ o involv1nr: 00’ labeled v1ta1711n 1],, 1n these studies. 1he solut10n was mixed and allowed to stand for two hours at room telwlperature to 1. 4:1,)erl'atant after first ”oiling off" . inst '011' s. mined Second “oi?” ottained .. .Zoz'zmercia‘. 'j—times crj',s‘eal'.12el ' . -ievuaercial 'l—times CITKSL 11‘ e ; V '. second ”oil" Obtained £13236 4. :?.l_‘.t.ogra;_:r. of the po‘.‘ zzg'la".l.:e £6; 6.1907 {‘0 {rat-ls L; nous - is 1 fl I z;ernatant after second “oiling ofi‘ allow binding to occur (31). .41 P—Z acrylamide head column had been equilibrated with pH 9 borate buffer. The void volume of the column was 70 ml. ~five mifliliters of the fl-lactoglobliin— (A U'! a o 0 ~ 0 w \ Co ‘ labeled Vitamin L19 solution were passed through the column. The effluent was monitored with an 17.1]. monitor. L‘he ,8--lacto;jlo- bulin peak appeared. on the Ufif. monitor chart immediately following the void volume. Collection of the fraction was star + 'I see as soon as the monitor sensed the presence of protein and.was coatinued until the recorder indicator returned to its baseline. This fraction co: sisted of 3'. -J. n Fractions were collected iollot-ring the 30 :71 sample and W 1 0 0 o ;‘ Kifi' ‘~ '1 n ChGCkGC; on tne seintillation counter for \IOU‘j-1232, 1t was 101an. that 15 1211 of buffer, free from both fi—lactoglol;>ulin and Cove-1'12, passed. from the 001112‘13’) between the fi~lactoglohulin peak with its and the peak attributable to excess free 3000-? 12. '1 o o l . A J _ benarate determinations had also sholm that when Co‘C—Tx:L2 was ssed through the P—Z column with the pl; 9 borate buffer system "8 its elution from he column began when approximately 2120 m1. of buffer had been eluted. fl~lacto;.:lo’:.luli1‘l solution without added (":0 »,. o a . CoL ~22:sz was gassed through the column prior to the binding study to determine the region of its elution (Figure 5). C11]; one protein peak appeared on the recorder chart of the LEV. monitor, and ir;edia— tel;r followed he void volume. ‘Ihe 30 all of effluent solution collected were well mixed. Folin protein determination was o a n s .1, — ‘ / \ carried out according to the mothoa descrieed by Jay-ale (clh A. standard curve had been prepared using our fl—lactoglo‘rmlin pre- paration for the standard, Vitarlin :12 content was determ'ned by O h i ’\ the microbiological assay previous v discussed and coU Vitamin 12 was determined using a Tri-Car’e scintillation counter. Electrodialvsis of B-lactog10bulin. One hundred :llilligrazls of fi-lactoglcbulin were put into solution in 10 ml of p31 9 borate buffer. Ellis solution was placed in a Visking cellulose dialysis sac which had been boiled with ethylenediarrinetetraacetic acid (E.D.T.A.) to remove ions which might interfere with free diffusion of materials through the sae. The sac was placed between plastic grids to flatten it a1 :1 hold it in place inside a plastic, one liter, graduated cylinder which was used for a dialysis vessel. Plastic spacers were used to hold the grids in place. A fez-I drops of toluene were put inside the sac and toluene was added periodically to the dialysis vessel to minimize microbial grmrth. Deionized water (500 ml) was added to the cylinder. P tinmn wires were in— serted for electrodes. The entire assembly was placed in an ice bath and connected to a constant power supply of approximately 100 milliamperes. Dialysis was continued for 14-3 hours. Binding Capacity After Electrodialysis The vitamin £12 binding capacity of electrodialyzed fi—lacto— \J) C) globulin was determined precisely as the vitamin 2'12 binding capacity of our fi—lactoglobulin was determined. The material remaining inside the dialysis sac after electrodialysis was dried by lyophilization and used for this determination. Then the . ff‘ ._ . ‘ .. A protein and excess Goad-1s 9 were separated on the {-4 column, la the U.V. monitor indicated a second protein peak coming off the column much later than the fi-lactoglobulin (Figure 5 ), indicating the presence of a low molecular weight protein or a peptide. 13:01: electrodialysis of a sample of fi-lactoglobulin containing excess (30 ~1 o a *1 . c o o o 9. a C0‘ -.:';‘-l? was carried out and bile material remain lg in the dials/‘51s sac was passed through the f-Z column, no second protein real: was present (Tigure 5). Vitamin B12 Content After Llectrodialysis of a Sample Containing Esteess C360 Labeled Vitamin 1.12. Cne hundred milligrams of fi-lactoglobulin were dissolved in A .. . A r. -, {3U -. . . 10 ml 01 111- 9 borate Llufier. One half 1d 01 to la;;eled v1tarur. .l. 1'12 I-TaS added and the solution was allowed to Sealid at room temper- ature _or two hours. Bleetrodialysis l-ras carried out precise y 93 m U‘ efore. After electrodialysis, vitamin 3112 content was deter- mined. ‘l’he material remaining in the sac after electrodialysis l-ras used for the determination. It was dried by lyophilization L and dissolved in 5 ml of p11 9 borate suffer. The protein was , 1 separated from any unbound vitamin 112 on the P—Z column and protein content and vitamin 3.12 content (microbiological and adiological) were deterrrlined as in previous binding studies. kl) Ix) Y—‘-v.-1»1‘.‘ ..-'.‘,-,‘I .7 1'.'-.‘ ha. . . A." t 1‘ » huh‘ A.) kal.A — ‘1... _—‘LJ~J LulJJ‘i J ILL) Sodium Caseinate The values obtained for the vitamin 312 content of sodium caseinate ranged fram 91 to 128 [JiLg per mg of protein E I 6.2“). An average value of lljuytg per mg of protein is given in I ble 1, under experiment 1. This average is taken from three separate experiments. Para-dapper k301131.03: iaree separate experiments gave an average vitamin E q conten r of 52 [Lng per mg of protein (S X 6.25) with a range of 45 to 55 [I’Lg per mg of protein (Table 1, experiment 1). 23A Solnble Fraction A range of 45 to 107 [L’Lg of vitamin £12 per mg of "protein“ (E A 6.25) was obtained in three separate experiments. An average value of 78 l‘fiLg per mg of “protein" is given in Table 1, ex- periment 1. Pair‘ II. J. bx; Insoluble Traction The average vitamin :12 content, of three separate studies, of this fraction is 153 [ifiLg per mg of protein (1 I 6.25). Ihe range of values found was 122 to 1:38- ,.LFg per m5; of protein (Table 1, experiment 1) . Recover? Table 1, under :perfment 2, contains vitamin 1312 recovery data based on one liter of skim milk. “The sum of the values for para-kappa corr‘lex, To}; soluble fraction and TC": insoluble fraction should be equal to the value obtained for isoelectric casein if no loss occurred. The value for isoelectric casein is 2.6 I: 1043 ftp; per liter. A loss of about 347: is indicated by these values. Chile 1, under experiment 3, contains data obtained in a re— covery study conducted on the initial 5 age of the fractionation of skim milk. Commercial skim milk from th Daily Store was used. Its B12 content was found to be 3.3 K O6 [Lfig per liter of skim :nillc. ri‘he casein was acid precipitated as previousl;r :‘Lescribed. When the values for the vitamin 7:12 content of the three fractions obtained were added, including casein, casein washings and whey, a value of 3.18 II 106 [(41.3 per liter was obtained. The sum of the (J\ 1312 values for the fractions should have equalled the 3.3 Ii 10 P‘P’g per liter obtained for skim milk, if no loss had occurred. These values indicate a recover"r of 965. J / 32+ Tilda? 1 Vitamin :12 distribution in casein proteins L12 Content (,‘*L5} Per mg Protein for liter of Skim Lilk hilk Fraction preriment 1 Experiment 2 Experiment 3 115 5 4 Sodium Caseinate (91—128)a 2.60 2:; 10,, age 3.; 10“ - -. , 52 / A Para nappa Complex (45—55) 0.5e A 10“ —————— I Q . U ,4 TCA (10p; Soluble (45-107) 0.059 I 10‘ —————— , 153 ,4 ,, . \ .,. 1 k TeA (10p) insoluble (128.168) 1,1 A 10 ______ 1. . . - ,, A commer01al okim hilk -—- --------- 3.30 x lOV . . . . 4 Casein‘washings ——————————— - 0,18 A 10L / Whey (acid) —————— ----—- 2.10 I 10“ a values in parentheses denote the range of three separate deteminations . Vitamin 11;; Binding Characteristics of B-Lactog loalin I ‘12 Content :3; the microbiological assag,r described earlier the vitamin T312 content of our fi-lactoglobvlin preparation in two separate assays, was found to be 15:10 and 130 flpg per mg of protein,re- spect' velg,’ (Cable 2) . 1.12 .‘Lindin {5 Capacity The vitamin 312 binding capacity of fi—lactoglobulin was found to be l7o0 and 15320 #Fg per mg of protein, respectively, in U'TO separate experiments by the microbiological assay. Lhe Soc3 labeled vitamin 1112 sound to the fi- lactoglobulin rras found to be 1" '31: and 1430 HLg per mg of protein, respectively, in two separate binding determinations (’i12‘a‘ le 2). i'Lwr Binding apacn. 11,, After Electrodialgfsis After electrOCial’sis, U1.der the co:.ditions pre riouslg, des— cri ed, the vitamin 1312 binding capacitg,r of fi-lactoglobulin, detemL'ze ed microbiolmica l", was reduced to 810 and C30 Mg per mg 01" prote eziz, respectively. and as determined by the bound radio- active isotope to 710 and 750 Mg per mg of protein, respectively, in twodi fferent electrodialgfs is an binding studies (Table 2) . \O ,x ( ample Containing Z‘xces s C) 1112 Content After dlectrodialgrsis of a C050 Vitamin 132 l n q o a o “ (in 5 .‘-1 sample 01 fi-lactogloznflin cox-ltainang excess be“ labeled vitznr'n :12 was electrodialyzed in the same manner as a previously described sa;:1plc of fi-lactoglobulin without added vitamin 1.112. After electrodialgsis the vitamin 1312 bound c; the fi—lactoglob'fligz was measured both microbiological}; and by; determining the a .1ount of C050 labeled vitamin I312. The microliological assay gave values of 11:70 and 1350 #413 per 111;; of protein, and the vita? n can. 1312 as detenzined by the radioactive isotope was fornd to be 31.3. A, A O o and 151313,;ng per 1115; 01 protein. {'1‘ ‘f'r '\ $21.14;” Vitamin 1: content of fi-lactoglobulin fella-ring various treatments [L’Lg per mg rotcin hicrobiological COCO—L12 Treatment Assay hone 140 ---- lCO —--- Excess C060—312 3760 1500 1320 1&80 Electrodialysis, excess 0000—312 810 710 830 750 ,1 A0 dxcess b0” -n12, electro- dialysis 1190 1230 1350 1360 . .- *-\"‘(_‘/17’.flr‘_“rfxl" '. Vr‘h ‘{~’ NT 17"1-“,‘*’r“ .‘J_.QU\--..'u_1_kz;-. 4.1..) LKI-.VJJ~L—),..VI.-J Casein contains between one third and one half of the vitamin 312 in milk. when vitamin 212 content is related to protein content, however, the filZ content per unit of protein is much lower in casein than in the whey proteins. In view of the fact that casein does contain a significant proportion of the vitamin 312 content of milk, it was thought that a fractionation of casein might yield one or more fractions 0 ntaining the br < of the 312 in the total casein fraction; in'which case a closer examination of the :12 rich fraction.would possibly help to elucidate the natrre of the protein—vitamin B12 complex. ho such B12 rich fraction was found. A three-fold difference between the vitamin 312 content of the para-kappa complex and the TCA insoluble fraction was not con- sidered great enough to warrant further investigation at this time. The vitamin 212 content of our [3-lactoglobulin was low compared to values reported.Ly other'workers (55). However, if the prolonged purification procedures employed are taken into consideration the low 512 content is not surprising. Our primary ain.was to study the effect of electrodialysis on the vitamin £12 binding capacity of fi-lactoglobulin, using Co60 labeled cyano- cobalamin. Therefore, the low "native" vitamin 312 content did not affect the purpose of our study. It could be advantageous to use a protein containing no native vitamin 312, if such a protein \P) \ . could be prepared without altering its 14112 binding: capacity. 'I‘T'xen, all the vitamin 3:732 would be of the added isotope, so that values obtained by the microbiological assaj should agree with values ob- tained when the scintillation counter is used and gamma emission of the COCO is measured to determine {'12 content. The 3312 content obtained rfable 2) in these studies by the microbiological assay and those obtained by measuring the garma emission of the radioisotope were in quite close agreement. ' e values obtained. by the microbiological assay were larger for most samples, and they should be higher, as this assay method determines both the "native" 13:12 present in the fi-lactoglobulin preparation and the added Co60 labeled. 73-12, If both methods of 312 assay were absolutely accurate the microbiological assa;r should have given values approximately 160 W5; per mg protein greater for each sample than the gamma emission assay, as this was the average ”native” 312 content found for this fi-lactoglobulin preparation. 13:; a comparison of the values in table 2 we can see that electrodialysis decreases the 312 binding capacity of B—lacto- globulin to approximately 50,3 of the value for undialyzed fl -lactoglobulin. 0n the other hand, if excess Co60 labeled vitamin 13:12 is added to B-lactoglobulin prior to electrodialysis, a remarkable change occurs. The amount of vitamin 1312 remaining bound to the fi-lactoglobulin after electrodialysis is much greater than the D12 binding capacity,r of electrodialyzed fi-lactoglobulin. 1,3,0 This definitely shows that electrodialysis brings about some change in fi-lactoglobulin which decreases its ability to bind vitamin L12 and that addition of vitamin L12 prior to electro— dialysis in some way prevents or lessens the effect of electrodialysis on the 1312 binding capacity of the protein. It would be helpful to know the nature of the change resulting from electrodialysis. Figure 5 is a photograph of the U.V. monitor charts showing the results obtained when fi—lactoglobulin was passed through the column under the noted conditions. When our fl -lactoglobulin preparation was passed through the column only one peak was recorded by the U.V. monitor, the fi—lactoglobulin peak. This peak occurs in the same region as the peak obtained when a solution of three times crystallized commercial ,B—lacto- globulin preparation is passed through the column. ighen fi-lactoglobulin is electrodialyzed and the material remaining inside the dialysis sac is passed through the column, a second peak is recorded. This peak occurs much later in the elution of the column than the fi-lactoglobulin peak, indicating a protein, or very likely a peptide, of much lower molecular weight. Large molecules such as fl-lactoglobulin do not permeate the pores of the acrylaflde beads in the column, are therefore not impeded as they move through the column and are eluted from the column immediately following the void volume. Small molecules permeate the acrylaride beads, are impeded in their passage through the colunm and 141 therefore require more time, or more elution volume to pass through the column. rfhe appearance of a "peptide” peak, due to electrodialysis coincides convincingly with the decreased ability of electrodialgfzed B —lactoglobulin to hind vitamin L12. rI'here certainly appears to be a relationship between the separation of low molecular I-Ieigh protein material and a decrease in 1312 binding; capacity on electro- dialysis. In further support of this relationship is the fact that no “peptide" peak was noted (Figure 5) when excess vitamin 2312 was added to fielactogflebulin prior to electrodialysis and the material from the sac following electrodialysis was passed throught the column. This gives greater evidence of an intimate relationship between what may be a "specific peptide," and the ability of the protein to bind vitamin L112. Ifot only;r does the removal of the pep- tide by electrodialysis decrease the 1312 binding capacity of the protein, but the presence of vitamin L12 prior to electrodialysis protects the peptide from cleavage by electrodialysis. Gizis _e_t_ i1.- (31) electrodialyzed skim milk and lyophilized the electrodialyzate. They found that the dried electrodialyzate contained 70,3 of the vitamin 312 and 1;; of the nitrogen present in the original milk sample. The electrodialyzate was found to consist of five peptides. Vitamin $12 binding determinations showed that two of these peptides were the principal li-lZ-binders 42 J A. Le treatment I 1 J _L B. Electrodialysis A L C. ElectrodiaLysis of sample with excess vitamin L12 Figure 5. EELution chromatograms of fi-lactoglobtflin when treated as indicated. with a binding capacity in the order of l I 105 foLg/mg of protein equivalent. It will be interesting to find the order of ‘he 112 binding capacity of our "peptide" or ”peptides“ as compared to the values reported for milk peptides (31) and as compared to the liver peptides (47). Further'work is needed to establish the exact mechanism of binding. It appears that we should first of all determine whether electrodialysis is cleaving a single "specific” peptide or a mixture of peptides. Peptide mapping by one—dimensional or two-dimensional electrophoresis in combination‘with paper chromatography, end group analyses and amino acid analyses could be helpful in establishing this point. *fe may find that the same peptide or peptides responsible for 312 binding are also involved with dimerization of B-lacto- globulin. ‘Je'will examine the electrodialyzate for the presence cf peptide or peptides there. The bulk of the peptide may be in the electrodialyzate. If a single peptide is involved, the fi-lactoglobulin would be an ideal model system for studying 312 binding. After the vitamin 1‘12 binding mechanism of fi-lactoglobulin is more fully understood, the same techniques used in its study might be applied to intrinsic factor preparations to help us better understand its exact role in the absorption of vitamin 312. (l) (2) (3) (4) 1:1; [‘1 Addison, 1. Anemia: Disease of the supra—renal capsules. lend. Led. Gaz. dz5l7, 13M9. Albanese, A. A. Iewer hethods of hutritional Liochemistrv. Academic Press, Jew York and London, 1963. Anthony; H. E., J. R, Couch, l.'fi. hupel, L. L. henderson and C. Brown. Vitamin T in blood of newborn and colostrumrfed calves and in 00 ostrum and normal milk of Holstein and Jersey cows. J. Dairy Sci. 3#:749, 1951. Aschaffenburg, R. and J. Drez-rry. Improved method for the preparation of crystalline 13-1aeto- globulin and a—lactalbmnin from cow's milk. Biochem. J, 65:273. 1957. Association of Official Agricultural Chemists. Cfficial hethods of Analysis of the Association of Cfficial Agricultural Chemists. Assoc. of Cfficial Agricultural Chemists,‘washington, D. C., 1960. . r -o’ Banerjee. D. 1., 8.1. Chess and J. D. Chatterjee. Free serum.vitamin B level in certain hematologic disorders. Llood 15:030, 1960. Berteher, E. W} and L. E. Meyer. Co0U vitamin L12 binding capacity of normal hum n serum. Proc. Soc. Exptl. Biol. Led. 94:169, 1957. Bird, 0. D. and I. Hoevet. The vitamin 312 binding power of proteins. J. Biol. Chem. 190:131, 1951. Bleumink, B, Preparation of a-lactalbuz'sfin from milk whey by precipitation with acetone. Neth. Eilk and Dairy J. 20:13, 1966. (16) (11) (12+) (15) (16) (17) (12.; 1+5 brunner, J. 3., C. A. Ernstrom, R. A. Rollie, L, L. larson, R. Lcl. Whitney and C. A. Zittle. homenclature of the proteins of bovine milk-first revision. J. Dairy Sci. 43:901, 1960. Lrunrer, J. R. and H. P. Thompson. Characteristics of several minor-protein fractions isolated from bovine milk. J. Dairy Sci. b4:122@, 19cl. mmMman ELK Licrobiological studies on materials which potentiate oral vitamin 312 therapy in Addisonian anemia. Arch, biochem. Liophys. 39:322, 1952. Castle, W. B. Observations on the etiologic relationship of achylia gastrica to pernicious anemia. I. The effect of the administration in patients with pernicious anemia of the contents of the normal human stomach recovered after the ingestion of beef muscle. Am. J. Led. Sci. 178:748, 1929. Castle,'U. B. and h. A. locks. Obsermations on etiological relationship of achylia gastrica to pernicious anemia. J. Clin. Invest. 6:2, 1928. Chapman, H. R., J. E. Ford. 8. 1. Ken, 3. Y. Thompson, S. J. Rowland, E l. Crossley and J. Rothwell. Ptuiher studies of the effect of processing on some vitamins of the L complex in milk. J. Dairy Res. 2h:191, 1957. Cherbuliez, E. and P. Baudet. Recherches sur la caseine. Sur la transformation de La caseine en paraeaseine. Helv. Chim. iota 33:1673, 1950. Cohn, 13. J., G. R. Minot, G. .‘L. Alles and U. T. Salter. The nature of the material in liver effective in pernicious anemia. J. Biol. Chem. 77:325. 1923. Collins, R. A., A. E. Harper, h. Schreiber and C. A. hlvehjem. 116 folic acid and vitamin L12 content of the milk of various species. J. Nutr. #3:313, 1951. Collins, R. A., R. E. Boldt, C. A. Elvehjem and E. I. Hart. Further studies on the folic acid and vitamin L12 content of cow's milk. J. Dairy Sci. 36:2Q, 1953. (20) (21.) (22) (23) (21.3-3- (25) (26) (27) (28) (29) (30) h- 1‘ V7 \ Conn, E. E. and P. 1. Stumpf. ‘ Outlines of Biochemistry. John Wiley and Sons, inc flew York, 1963. x ' Y I' Da1slev,1. Cel filtration of sea water: separation of free and hound forms of vitamin £12. Lature 19 :SCS, 1961. Davis, L. D. and E. S. Tingioli. Mutants of Escherichia coli requiri n me ethionine or vitamin 312. J. Lacteriol. 60:17.1 g 950. Davis, R. L. nd I. F. Chow. Some applications of the rapid uptake of vitamin E.2 bv resting lactobacillus leichnannii organism. Sc1ence 115 351. 952. Davis, R. L., l. l. Lavton and L. F. Chow. rptake of radioactive vitamin L12 bv various microorganisms. Proc. Soc. Exptl. Liol. Led. 7E .a73, 1952. DrVder, L. P., G. I. Riedel and.A. I. Hartman. Comparative assa y for vitamin L] in certain milk products by various rat growth methods. J. Er r. 59zT9, 1955. _*avlkner, J., D. Carroll, T. Driscol l and T. C. Johnson. Co“ vitamin T12 bin nding 0y chromatographic fractions of human gas vric contents. Am. J. Clin. Lutr. 8:512, l9/C. Finogenova, T. V. Vitamin 312 production b3 proactino:13ce.tes of the genus iocardia. Biol. hauki. lOCB: 177, 1964. Ford, J. E. Factors influencing the dcstruction-by heat of vitamin :12 in milk. J. Dairy Res. 24:360, 1957. I\ -J Y'\ n“ Ford, J. 3., T. s. u regor3r and 0. Y. Thompson. Ann. Rept., Zatl. Inst. Ros. Dairying, Reading, 112, 1950. Frost, D. V., K; lapidus, L. A. Plant, 2. scher fliJLg and n. H. Ericka. Differential stabilitv of variOts analogres of cobalanin to vitamin C. Science 11/:119, 1952. (30 (1:1) p. Gizis, Q. J., J. R. Lrunner and b. S. Cchweigert. Vitamin E o-binding peptides in cow's milk. J. Dairy Sci. 98:1704, 1965. Class, C. I. J. Gastric intrinsic factor and its function in tie Meta 0115“ of vitamin 312. 1H13s1ol.Roviews 43: 529, 1963 C-rasbock, R. fractionation of human gastric juice ar d saliva emp103 starch electrophoresis. Gastroenterologia ha99, 1955}. Gregory, M. E. The microbiological assay of vitamin B12 in the milk of different animal Species. Trit. J. Tutr. 8:340, 195+. Gregory, T. E. and E. S. Roldsworth. A cyanocohalamin-protein complex from sow's milk and desiccated pig stomach. Iature 173:830, 1959. “re ery,1€. E. and E. S. holdsworth. Some ooservations on the measurement of “binding" of cyanocobalamin by “intrinsic factor" preparations. Liochem. J. éézb56, 1957. Crecory, L. E. and I. S. holdsworth. Some properties of the cyanocobalamin—protein complex from sow's milk, and the node of linkage of cyanocobalamin with protein. Biochen. J. 59:335. 1955. Crer;or3, h. E. and E. S. Holdsworth. lhe binding of cyanocobalamin and its naturally occurring analogues by certain bodV. fluids and tissue extracts. Ziochem. Ziophys, Acta L2M62,196C. Gregory, E. 3. and E. S. Holdsworth. The combination of some vitamin D Z-like compounds with sow' 5 milk and "intrinsic fac cor” concentrates. hicchem. J. 55:830, 1953. Gregory, L. E. and 2. S. Holdsworth. the occurrence of a cyanocobalamin-binding protein in milk and the isolation of a C3anocobalamin-protein complex from son's milk. Ziochem. J. 59:329. 1955. GregorV, I. E., J. 3. Ford and S. I. fon. fl vitamin Inn binm ng factor in sow's1xilk. Iiochon. J. 51:12:03, 1952. Gregory, :1. L5,, J. I. Ford and 5.1. iron. The I vitamin content of mill: in relation to Breed of cow and stage of lactation. J. Dairy Res. 25:447, 1953. C-rossowicz, 13., J . Aronovitch and Liaclmilewitz. beterrrzination of vitamin 2’12 in hunan serum by a mutant strain of :_j. coli. I‘roc. uOC. -‘brzptl. iiiol. Led. {37:513, 1951+. Groves, 1;. L. The isolation of a red protein from milk. J . Am. Clzezn. Soc. 32:331'r5, 1960. liartman, A. 11., L. 13'. Dryden and G. H. 3316981. Vitamin 11-12 content of milk and milk products as detenained b3 rat assar. J. llutr. 59:77, 1956. Iledl; om , A. A native colnlanin-polypeptide complex 1‘ r011 liver; amino acid composition and terminal amino acid analyses of the peptide part. Liochem. J. 79zb69, 1961. Hedbon, A. A native cobala‘nin-polypeptide complex from liver: isolation and characterization. Biochem. J. 7":307, 1960. Ilcdbon, A. A native vitamin 312 polypeptide complex. L-iochen. 213.101-1355. Acta 1'? :13'37, 1955. Zleinrich, If. C., C. Badel and R. Skib‘oe. itadiopapiercl'lromatog'raphische Untersuchungen uber die {31:62:— ii‘itat der Iiindung von (CoC‘O)-vitamin 1,]2—xmalogen an Intrinsic factor-Lonzentrate. In: l-iadioaktive Isotope in filinik 11nd Jorschung. liunich, Urban und Schuarzenlserg, 9 0 car crfi “lg-4.).) 9 l/JUO Elodgkin, D. C., J. Lamper, IIacI-iay, J. I’icIG-Iorth, ll. Trueblood and J. G. White. Structure of vitamin 212.1?ature 173:6}, 1956. ch. Ioff—Jérgensen, E. and J. Jorm-Tetersen. gfeasurenent of serum binding: capacity for vitardn 3.1g. l‘roc. Soc. Exptl. Jiiol. lied. 110:57l'r, 1952. (52) \C) Eloi‘fmn, C. 33., 33.. L. ‘-.‘.. Stokstad, 3. L. ITutchings A. C. Dorabvsh and 1.11. Juices. The microbiological assay of vitamin L 1.11111 lactobacillus leic‘m 1.2.11111i. J. Biol. Chem. 131:635, 151%). Hutner, 3.11., 1.. lTovasoli, IL. L. 1'1. Stokstad, C. .-=I. 1lof1111a11, 1;. Lelia, 1'1. L. Iranklin and 1‘. 1:. Juices. 1155-3; of anti-pernicious anemia factor with 2111510118.. 1‘1 00 Soc. 13330131. Fiol. Le d. 7C:ll(:=, 19149. Lakei, 11. and” c. 13. J. Glass. Separation of bound and free vita-min 1'17 on Sophaclox ‘1 6-25 colmm. E1100. Soc. Exptl. Liol. Fed. 111:270, 1952. ”rs'wa c‘n'mo 17' "-‘ T“ 9 ‘ ' t" 17178.19"l ‘i .- CII‘v-vh)-nlth, $5. , ‘J. vesal‘a (inc: -—. ::i h) —A 1‘ . T Studies on the vitamin £1170 content in cow milk.1‘roc. Silver Jubilee, 1,113.1“ '1111 11.1.53 3., College of agriculture, - or L; oto Toniversic“ Japan. 115, 1,1. Lin, ‘1’. 17., 23. Cizis, J. 7.. Enumer and 11. £3 Schreigert. Vitamin 1‘32 distribution in cow's milk. J. 111111311. 13:3, 1965 1011,11,... b'erMe ation Proceedings, 20(1). Pt. 111:209, l96l. 1.011, 3. I. 11‘ooc‘.11;_-jricultm‘e Crganization, United I'Zations, 1:10 intrition Studies. 17, l959 Larson, 1:. L. and It. Jenness. 3 ‘I-aCtOElObUJ—‘ln. :Jioche... Fregarations [7° 1055. 05—0.], / 7.1.1121“, 2‘1. 1..., C. C. L'ngley, 73'. V. Cox, 1'7. I-".c’i.Vo;3r-Iloz-re and L. Raine. :31 ectrophoresis of 3111113111 gastric juice in relation to Ca stle' s intrinsic factor. Brit. Zed. J. 1: 'M7, 3953. Iayne,13. Spectrop‘notometric and turbidimetric anetl‘ods for measrring proteins. LI101‘-1od i__1_1 Singinologr 3:247, li‘ 55. .-.:1ci1.1st1t 11a] fe1me11tations.!mn. liev. f’icrobiol. 7:1-933, l" J kn CT) rn (€33) liclleekin, i. L., 1?, J. Hipp and IL. I. Groves. The separation of the components of a-casein. I. The preparation of a—casein. Arch. Biochem. Liophys. 83:35, 1959. ,1 n ‘ ' 1-. —. (614-) Iiendelsolm, R. 3,, c. L-, Uatkin, A. I’. Iioroett and J. Janey. Identification of he vitamin E 9 binding protein in the serum of nonuals and of patients I-Iith chronic :‘rfelocytic let‘liemia. Blood 13:740. 1958. (65) killer, A. and J. 3‘. Sullivan , 'Ihe “in vitro" binding of cobalt00 labeled vitamin 1&2 by normal and leukemic sera. J. Clin. Invest. 37:556, 95?, (66) Killer, 0. Determination of bound. vitamin 2312. Arch Biochem. Ziophys. 68:225, 1957, (~37) Zi’inot, G. 3‘.. and II. 1‘. Lurphy. Ireatment of pernicious anemia by a special diet. J . in. lied. Assoc. 577:1;70, 1926. ’f‘ ~- ' . ‘ V a I f‘ ' (oo .-;;_1.xlgaonx