. "I‘ll, I Junipljl. :95... g. “4.1: . .c . lwufl“. ‘00?” ' ..-.vv. It.“ .vxu: ., .Au..fl‘ux<0v.\llun\ts-. : ~10va hf: w ‘ I 1le :1" : . : : .lo ,1I.::...a.v n I ~ .8. c. . n ,V : 00.! h r. I , o . ,1. 1.25411 . . . . 1| u, I '9 0v 1 | 1 i V l i V. II l I ' O H" . :4 in... u: . .0 . . . u .1. 2: a Q. .u : : : .. . \. o 1.: - I. o . . : \ v.3 . . . a I : .. “\Iaqll‘éln‘ul thWWk-Dv . : : o. . 3 -. vlnhfldflflduflfi ”\iih. . . .. . . - :1 1: . 1.4 1.): ‘”.tt...l.ri$ .:.!.-. .2- - no. .I | .‘v|‘D-|r\. . ‘..|'ol«'|s v t : ¢ . . A :I. N“: ‘DL 7 .II‘ 2...... ni:§ . 7 . . .1 ‘11:? m... 5.1 m -. ::!Lr:xu-;.. - hulk": " ‘ a: .: " : ' “I ”y. ‘ : | ' .. v " I 1‘ | "V. . l' " 'I O. ‘4 O :: :' ‘lvil |~n n I. v u: :. : Il‘l - I”:”:u..:.l|l$:\.l..\.uk.t.l: : . ::m:...\l::l \ 31) s {I I‘llllxt\lltn\:l1\.th - I. I! : “willful: ::(l|| b :1 J: i :1 . “out“ H Id ““41 :ll : o- :l: l\l|(.l:| ,: \ll - (”1-1 . “HA turn“ VI 6 . 'WMY'f: I" I; ,I 4| IZ'J‘Z‘ 6:;‘1‘ .AL‘A“ '. r‘« - ”(it I.":‘: A n‘ r \ This is to certify that the dissertation entitled STUDIES ON XANTHINE OXIDASE IN COW'S MILK presented by Greg. Shu-Guang Cheng has been accepted towards fulfillment of the requirements for Ph.D. degree in Departgent of Food Science and Human Nutrition fl] Major professor MSU is an Affirmative Action/Equal Opportunity Institution Date May 26 1983 0-12771 MSU LIBRARIES .—:,—- RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. w 3 i: (:3 3pm ligE mm Git-.1. In . (“if 83“” 41 D «D /<’//~ 00 STUDIES ON XANTHINE OXIDASE IN CON'S MILK By Greg. Shu-Guang Cheng A DISSERTATION Submitted to Michigan State University in pattial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 1983 ABSTRACT STUDIES ON XANTHINE OXIDASE IN CON'S MILK By Greg. S. G. Cheng Xanthine oxidase (E. C. 1. 2. 3. 2.) was purified from fresh cow's milk by differential centrifugation and hydroxylapatite chromatography. The final product possessed A(280nm)/A(450nm)=4.84; A(1cm, 280nm, 1%)=11.9; activity/A(450nm)=l4l; specfic activity=3.59 IU/mg; and a detectable dehydrogenase activity. Purified enzyme was a reversible oxidase form and could be converted to dehydrogenase with 10mM dithiothreitol (DTT) or 1% mercaptoethanol (ME). Chemical analyses of the enzyme preparation indicated the presence of 14.8% protein nitrogen and the absence of lipid. The enzyme contained 82 sulfhydryl groups per mole (302,000 daltons) with 44 of these occurring in disulfide bonds. Amino acid composition revealed that the enzyme was hydrophobic in nature and contained lysine as its N-terminal residue. Triton x-100 did not affect the enzyme activities while 1% mercaptoethanol enhanced its dehydrogenase activity. Six molar urea reduced the ability of oxidase to convert to dehydrogenase. An active monomer of X0 with an estimated molecular weight of 155,000 was obtained from a Sephacryl 5-200 gel filtration column in a 6M urea environment. Sodium dodecyl sulfate polyacrylamide gel electrophoresis of purified enzyme revealed a single, sharp zone with a molecular weight of 151,000 i 4.000. The enzyme retained its oxidase activity after limited proteolysis by trypsin, chymotrypsin, plasmin, pancreatin, pepsin and papain. The proteolyzed X0 remained unresolved Greg. S. G. Cheng in polyacrylamide gel electrophoresis with and without dissociating agents such as 1% mercaptoethanol or 6M urea. Three major zones with molecular weights of 85,000-100,000, 30,000-35,000 and 18,000-20,000 were commonly observed in sodium dodecyl sulfate gels. Amino acid content of the four principal subunits of trypsinized X0, e.g., 136,000, 85,000, 35,000 and 18,000 dalton species, indicated a hydrophobic nature and lysine as the N-amino terminal for all subunits. Measurements of riboflavin and its derivatives in aging milk as well as the nature of its enzymic activity indicate that X0 could not be the source of milk riboflavin. Surface and interfacial activities as well as its hydrophobicity indicate that the enzyme is lipophilic. Fat globules in an enzyme—butter oil emulsion were more stable than those in emulsions of casein- or whey-butter oil. The observed experimental evidence supports the hypothesis that X0 serves to stabilize milk fat globules in its intracellular transport as well as in secreted milk. ACKNOWLEDGMENTS I would like to express my sincere gratitude to my major professor, Dr. J. R. Brunner, for numerous discussions during the course of my graduate studies and his aid in the preparation of this dissertation. The influence of his research approach, encouragement, patience and inspiration significantly contributed to the success of this study. Appreciation is also extended to Dr. J. N. Cash and Dr. J. I. Gray of the Department of Food Science and Human Nutrition, to Dr. E. S. Beneke of the Department of Botany, and to Dr. H. A. Lillivek and Dr. N. N. Hells of the Department of Biochemistry for reviewing this manuscript and serving on my graduate committee. I wish to acknowledge the fine assistance provided by Ursula Koch, and especially appreciate her help with the amino acid analyses. I would like to thank Dr. C. M. Stine for the use of equipment during this study. The interaction and discussions with several fellow graduate students during the course of this study is also gratefully appreciated. I would also like to thank the Department of Food Science and Human Nutrition for the use of facilities and financial support. And last, but not least, to my beloved wife, Lin, go my thanks for her patience, understanding and encouragement. ii TABLE OF CONTENTS Page LIST OF TABLES O 0 O O O O O O O O O O O O O 0 O 0 O O O O V LIST OF FIGURES. O I O O I O O O O O I O O O 0 O O O O O O Vii INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW. . . . . . . . . . . . . . . . . . . . . 4 Occurrence in Milk . . . . . . . . . . . . . . . . . . 4 Xanthine Oxidase and Milk Membrane . . . . . . . . . . 8 Milk Fat Globule Membrane (MFGM) . . . . . . . . . . 8 Skim Milk Membrane . . . . . . . . . . . . . . . . . 11 Isolation of Xanthine Oxidase. . . . . . . . . . . . . 12 Chemical and Physical Properties . . . . . . . . . . . 18 Enzymic Characteristics. . . . . . . . . . . . . . . . 22 Substrates and Inhibitors. . . . . . . . . . . . . . 22 Assay. . . . . . . . . . . . . . . . . . . . . . . . 23 Catalytic Kinetics . . . . . . . . . . . . . . . . . 25 Xanthine Dehydrogenase Conversion. . . . . . . . . . 28 Biological Role and Applications . . . . . . . . . . . 29 EXPERIMENTAL . . . . . . . . . . . . . . . . . . . . . . . 37 Materials and Chemicals. . . . . . . . . . . . . . . . 37 Preparative Procedures . . . . . . . . . . . . . . . . 37 Fractionation of Whole Milk. . . . . . . . . . . . . 37 Isolation of Xanthine Oxidase. . . . . . . . . . . . 38 Isolation of Polypeptide Fragments of X0 Cleaved by Trypsin . . . . . . . . . . . . . . . . . . . . 39 Physical Methods . . . . . . . . . . . . . . . . . 41 Discontinuous Polyacrylamide Gel Electrophoresis (Disc-PAGE).oooooooooooooooocoo 41 ureaDisc-PAGEo00000000000000.coo 42 SDS'PAGEoooooooooooooooooooooo 42 Surface and Interfacial Tensions . . . . . . . . . . 43 Enzyme Assay . . . . . . . . . . . . . . . . . . . . . 44 Xanthine Oxidase . . . . . . . . . . . . . . . . . . 44 Xanthine Dehydrogenase . . . . . . . . . . . . . . . 45 NADH-Ferricyanide Reductase. . . . . . . . . . . . . 45 Chemical Methods . . . . . . . . . . . . . . . . . . . 45 Nitrogen Content . . . . . . . . . . . . . . . . . . 45 Lowry Determination of Protein . . . . . . . . . . . 45 iii Available and Total Sulfhydryl Groups. Disulfide Groups . . . . . . . . . . . Amino Acid . . . . . . . . . . . . . . Tryptophan . . . . . . . . . . . . . . N-Amino Terminal Free riboflavin, Flavin Mononucleotide Flavin Adenine Dinucleotide (FAD). Limited Proteolysis of X0. . . Flavins Content and X0 in Milk Stability of Milk Fat Globules RESULT AND DISCUSSION. Isolation. . . . . . . . . . . . . Purification of Xanthine Oxidase Criteria of Purity of X0 . . . . Enzyme Nature. . . . . . . . . . . Kinetics . . . . . . . . . . . . Xanthine Dehydrogenase Properties Active Monomer . Inhibition by Chloroquine. . Chemical Compositions. . . . . Effects of Dissociating Agents Effects of Limited Proteolysis Amino Acids Contents of the Fragments xo 0 O O O O O O O 0 O O O O 0 O O O O Quaternary Structure of Milk X0. . . . . . Free Riboflavin, FMN and FAD Contents and X0 in Cow's Milk. of O O O O A. O O O O o—fi'oooooooooo '11 00.03.0000 a O O k 0 O O O O O C O O O 0 Stabilization of Fat Globules by X0. . . . . CONCLUSIONS. . . . . . RECOMMENDATIONS. . . . APPENDIX . . . . . . . BIBLIOGRAPHY . . . . . iv U 00100000000000. oooovooooo o o g oooomooooo ooflooaooooooooooo .0. o11.7 (Nagler and Vartangan, 1973). The activity of the enzyme investigated over a pH ranging from 5 to 12, using 0.1M Tris-HCl, Tris-NaOH, Tris-acetate buffer to cover the pH spectrum, showed an optimun pH of 8.3 (Mangino and Brunner, 1977a). Greenlee and Handler (1964) reported that me of xanthine oxidation is constant from pH 11.0 down to 8.5 and was decreased by a factor of 2 on lowering the pH to 7.0, and by an additional factor of 2 at pH 5.0. The activation energy of the enzyme reaction derived from an Arrhenius plot of initial velocity across a temperature range of 15 to 40 C as 14.1 kcal/mole (Mangino and Brunner, 1977a), or 14.5 kcal/mole for the range of 10 to 38 C (Massey et al., 1969). Temperature stability of the pure X0 has not been studied extensively. Most work on the heat inactivation and cooling effect of X0 has been investigated in the milk system. X0 activity can be detected in whole milk heated at 170 F (76.7 C) for 15 min. The enzyme was inactivated after heating to 195 F (90.1 C) for 15 sec and could be partially reactivated when the same milk was condensed to 50% total solid and homogenized at 4500 psi. Further increasing homogenization pressure did not enhance the amount of activation (Greenbank and Pallansch, 1962). They suggested that an inactive form of X0 which associated closely with fat globules is displaced from the fat globule by homogenization and hence is activated. Excessive homogenization, therefore, not only would free the enzyme but rapidly denature it since free enzyme is apparently sensitive to heat denaturation. Gudnason and Shipe (1962) found that X0 activity in fresh milk 20 increased after storing at 4 C, heating to 70 C for 5 min, or homogenization. X0 activity was also increased when fresh milk was incubated with comnercial lipolytic and proteolytic enzymes such as steapsin, lactivase, and pancreatin. The activity of both milk and cream increased on storage, whereas there was no activity increase in skim milk. However, skim milk obtained from milk stored at 4 C for 24 hr showed an increase in X0 activity while the corresponding cream showed a decrease. These investigators supported that some of X0 is not active in fresh uncooled milk and the observed increase in activity was related to the redistribution of the enzyme. This redistribution was believed to be associated with the dispersal of MFGM in milk after cooling, heating, or mechanical stress. Bhavadasan and Ganguli (1980) studied the temperature effect of X0 activity in milk by investigating the changes of activities of free and membrane-bound X0. They found an increased activity of membrane- bound X0 and a decreased activity of free X0 in buttermilk concomitant with a significant increase in the activity of free enzyme in skim milk after whole milk stored at 5 C for 24 hr. These observations indicated that the increase of X0 activity in whole milk could not be a consequence of desorption of microsomes from fat globules and their release to skim milk as suggested by Gudnason and Shipe (1962). They suggested that the increased activity of membrane-bound X0 in buttermilk was due to structure changes occurring in MFGM as a result of cold storage of milk (Coleman, 1973). The increased activity of the free enzyme in skim milk and its decrease in buttermilk was attributed to a distribution of part of the free enzyme from the fat phase into the milk serum during cold storage of milk. They also found a 21 significant increase of X0 activity in the cream phase and a slight increase of the activity in skim milk after heating of milk at 60 C for 5 min. The observation of the state of X0 activity in skim milk from heated reconstituted milk led them to conclude that the increased X0 activity in milk after heat treatment was not due to the release of the enzyme from the fat phase to the serum phase. They suggested that the increase of X0 activity on heated milk was possible due to the presence of a heat resistant activator in milk which reactivated the enzyme or the inactive form of X0 became activated during heating of milk. Bhavadasan et al. (1982) reported that agitation of milk at low temperature, 10 to 25 C, resulted in high lipolysis of milk fat concided with a great release of membrane-bound X0 from MFGM into skim milk. They suggested that release of membrane-bound X0 from MFGM was due to the disruption of MFGM by agitation. This disruption of MFGM increases the susceptibility of milk fat to the lipase system and hence, enhanced the milk lipolysis. Mechanical force used in homogenization of milk is much stronger than that used for agitation. During homogenization of milk, MFGM is subjected to an extensive rupture which causes the release of free and/or membrane-bound X0 from the fat phase and subsequently increases the enzyme activity in milk. Homogenization studies on milk heated at 48 C for 5 min showed that increase of X0 activity was a linear function of the pressure between 70.3 and 281.2 kg/cm2 . Each additional kg/cm2 of pressure resulted in additional 0.16 milliunits of activity per ml of milk (Demott and Praepanitchai, 1978). 22 Enzymic Characteristics Substrates and Inhibitors Xanthine oxidase is capable of reacting with a wide variety of substrates, requiring two substrates, i.e., both reducing and oxidizing, to complete its reaction. Generally, the reducing substrates are purines, pteridines and their derivatives as well as aliphatic, aromatic, and heteroaromatic aldehydes which can also be the substrate of aldehyde oxidase. Bray (1975) listed a number of reducing substrates and demonstrated the difference of the substrate specificity between X0 and aldehyde oxidase. Reducing substrates bind to the molybdenum active site except for NADH which interacts with the enzyme via the flavin molecule. Bunting et al. (1980a) found that nitrogen heteroaromatic cation substrates reacted with the enzyme only at pH values above 9.6. They attributed this change in specificity to an alteration in the substrate-binding site at high pH. This phenomenon provided them with a mechanism for investigating the reducing substrate binding site of X0. They proposed that an electronegative atom (oxygen or nitrogen) of the substrate interacts with a hydrogen-bond donor in the active site during the reaction. This interaction allows the sequence of oxidation at C6, C2, and C8 of purines and C4, C2, and C7 of pteridine. For example, uric acid is generated from xanthine, xanthine is formed from hypoxanthine, and hypoxanthine is produced from oxidation of purine (Bunting and Gunasekara, 1982; also see the Appendix, Figure A4 and A5). Oxidizing substrates acting as an electron acceptors are oxygen, ferricyanide, ferritin, cytochrome c, and a large number of dyes such 23 as methylene blue, indophenols, etc. Interaction of oxidizing substrates and the enzyme is via the flavin active site. The enzyme specificity varies depends on not only the differences in the substrates encountered but also on the concentration of both reducing and oxidizing substrates and the pH values employed in assay conditions. Most reducing substrates such as purines and aldehydes are most rapidly oxidized at neutral pH while the quaternary heterocyclic compounds only react with X0 at pH>9.6 (Greenlee and Handler, 1964a; Bunting et al., 1980a,b). Also, the X0 reaction is susceptible to inhibition by excess substrates (Bray, 1975). Inhibitors of the enzyme consist of amino group reagents such as 2,4-dinitrofluorobenzene, 2,4,6-trinitrobenzenesulfonate, benzylaldehyde (Greenlee and Handler, 1964b), alloxantnine (1H- pyrazol0(3-4-d)pyrimidine-4,6-diol) (William and Bray, 1981), 9-(p- aminoethoxy-phenyl)guanine (Baker and Siebeneick, 1971), and allopurinol (4-hydroxylpyrazolo(3,4,d)pyrimidine) (Bray, 1975). Cyanide can inactivate the enzyme due to cyanolysis of a persulfide group in the molybdenum active site. Cyanide-inactivated enzyme can be largely reactivated by incubation with Naz S (Massey and Edmondson, 1970). Hwang et al. (1967) found that all the major protein fractions in milk can act as an activator or inhibitor of the enzyme depending upon their concentration in the reaction mixture. .As_sax Since X0 catalyzes oxidation of a large group of substrates, a number of enzyme assays have been reported. Assays of X0 or dehydrogenase generally present no unusual problems. Assays quantitating the catalytic activity of X0 through the rate of 24 substrate disappearance or product appearance are normally employed. The most widely used procedure utilizes xanthine as reducing substrate and oxygen as oxidizing substrate; liberating uric acid. An international unit (10) of enzyme activity is defined as the amount of the enzyme needed to convert one micromole of xanthine to uric acid per minute at 23.5 C in a pH 8.3 solution of 0.1 mM xanthine and saturated with air (Bray, 1975). Avis et al. (1955) introduced a spectrOphotometric method to monitor the appearance of uric acid by absorption measurements at 295nm under conditions specified to yield results in 10. An initial rate measurement via A(295nm)/min allows a direct calculation of the enzyme activity, utilizing the molar absorption coefficient difference 3 1M."1 (Kackler, 1947). between xanthine and uric acid, e.g., 9.6x10 cm_' This method has been widely adopted by later researchers. Other spectrophotometric methods reported measure the color reactions of methylene blue (Dixon and Thurlow, 1924), triphenyl tetrazolium chloride (Zittle et al., 1959), cytochrome c (Horeker and Heppel, 1949), conversion of ferricyanide to ferrocyanide (Krenitsky et al., 1972), and conversion of vanillin to vanillic acid which reacts with 2,6-dibromoquinonechloroimide (Kuramoto et al., 1957). The dye reaction of tetrazolium chloride with the enzyme has been utilized qualitatively to visualize the active X0 species in gel electrophoresis (Zikakis, 1979a). Ball (1939) was the first to use a manometric technique to monitor the disappearance of oxygen in the reaction. Zikakis and Treece (1971) developed a polarographic technique to measure the oxygen consumption by using a Clark oxygen electrode. This method is claimed to be more 25 rapid than spectrophotometric methods. Three milliliters of sample and 0.2 ml of 0.1M phosphate buffer, pH 7.2, are equilbrated at 38 C with air purging. After inserting the electrode,~50 ul of 3.13x10-3M hypoxanthine are added to start the reaction. Completion of the typical reaction requires two to ten minutes. The electrode has been incorporated into an integrated automatic system (Yellow Springs Instrument Biological Monitor, Model 53) which is capable of plotting the oxygen uptake. Results in microliter of oxygen per ml sample per minute (polarographic unit) are readily converted into the conventional I.U. by using the gas law of dele and Charles (Zikakis, 1979b). A radiochemical assay employing (14C)-labeled xanthine was reported to be approximate 4,000 times more sensitive than either one of the above assays but required more analysis time (Daugherty, 1976). However, the polarographic method, following oxygen consumption, and the spectrophotometric technique, following the appearance of uric acid, are the most frequently used assays. Both provide accuracy and convenience since the results are expressed directly in or readily converted to I.U. of X0 activity. Assay for xanthine dehydrogenase activity generally employs xanthine and NADI'as substrates. Measuring the formation of NADH at 340 nm is the most comnonly way to obtain dehydrogenase activity. Catalytic Kinetics Analysis of steady state kinetics of X0 showed a ping-pong binary mechanism which involves two half-reactions separated from each other. The half-reactions are reduction of the enzyme by xanthine followed by release of uric acid and reoxidation of the enzyme by oxygen with subsequent dissociation of hydrogen peroxide (Massey et al., 1969; 26 Edmondson et al., 1972) Each half molecule of XO contains two sites for interaction of the various reducing and oxidizing substrates. These are the flavin site whose function depends on the presence or absence of a nearby thiol group and a molybdenum site whose functioning is dependent on having a nearby persulfide group intact. The reducing substrates, except NADH, interact at the molybdenum site, whereas NADH and oxidizing substrates interact at the flavin site. Evidences relating to the roles of these sites are: [1] the demolybdo enzyme is inactive toward xanthine, [2] reducing substrates influence Mo(V) electron paramagnetic resonance signals from the enzyme in a number of specific and direct ways but have no such effects on signals from other chromophores, [3] substrates, except NADH, no longer reduce the enzyme after removal of the persulfide group by cyanide, [4] substrate analogs, particularly allopurinol, inhibit reduction of the enzyme by normal substrates but not by NADH, [5] removal of the flavin abolishes reoxidation of reduced enzyme by oxygen, but does not affect its reducibility by xanthine (Bray, 1975; Komai et al., 1969; Massey and Edmondson, 1970). Olson et al. (1974b) described the reaction of X0 in terms of relative reduction potentials of the various electron acceptor groups. The mechanism of the X0 reaction is best described by the order of the electron affinity constants which is roughly FADZFe/SZMO. Differences between reductive substrates can be interpretated in terms of differential perturbation of these reduction potentials when substrate is bound to the enzyme. The results of these studies supported the conclusion of Edmondson et al. (1972) that the rate-limiting step in both reduction and oxidation is the decay of ES-complex which is 27 accompanied with rapid electron redistribution among M0, the Fe/S center and FAD. The Fe/S centers act as electron reserviors functioning to maintain Mo(VI), (for efficient reduction) and flavin as FADH2 , (for efficient oxidation). Each half molecule of X0 contains one 2-iron labile sulfur center, thus each half molecule is capable of accepting any number of electrons between 0 to 6 from reducing substrate (before reoxidation)-2 on 2 Fe/S, 2 on No, and 2 on FAD, a total of 6 electrons. X0 is capable of reducing oxidizing substrates in one-electron, two-electron, or both reactions. Nith oxygen, superoxide is produced in a one-electron reaction whereas hydrogen peroxide is produced via a two-electron transfer. High pH, high oxygen concentration, and low xanthine concentration all tend to favor the one-electron pathway relative to the two-electron pathway. Both one-electron and two- electron transfer occur at the flavin molecule (Olson et al., 1974a; Bray, 1975; Hille et al., 1981a). Porras et al. (1981) and Hille and Massey (1981b) have proven Olson's hypothesis of one- and two-electron transfer on reoxidation of the enzyme by molecular oxygen. They demonstrated that the electron transfer followed a sequential removal of electrons, e.g., 6--4--2--1--0. The first two steps of the two electron oxidation forms hydrogen peroxide and the last two steps represent a one-electron transfer to form superoxide. The schemes of the reduction and oxidation mechanism of X0 and electron transfer proposed by the above researchers are illustrated in the Appendix Figure A's. The pathway of internal electron transfer within X0 molecule is beyond the scope of this discussion, but has been discussed by Barber's group (Barber and Siegel, 1982a,b; Spence 28 et al., 1982). Oxidase Dehydrogenase Conversion Stirpe and Della Corte (1969, 1970) found that many mammalian xanthine-oxidizing enzymes exist in vivg_and freshly prepared tissue extracts as NAD‘I-linked dehydrogenase. During isolation and storage, the activity with NAD+ is gradually lost while that with oxygen increases. Conversion of the dehydrogenase (type D) to the oxidase (type 0) can be brought about by heating, proteolysis, storage at -20 C, aerobiosis, organic solvents, incubation with subcellular fractions, and sulfhydryl modifying reagents. This conversion, except for proteolysis, is reversed by treatment with dithiothreitol or dithioerythritol. It was suggested that a specific thiol group(s) in the vicinty of the flavin active site is essential to utilization of NAD+ as an acceptor. Modification or oxidation of this thiol group(s) results in the loss of reduction of NAD+ and an increase in the affinity to use oxygen as an acceptor. Proteolysis is assumed to remove the thiol group or displace it from the vicinity of the flavin (Della Corte and Stirpe, 1972; Naud and Rajagopalan, 1976a). These finding were recently confirmed by Haud and Rajagopalan (1976b) and Coughlan and Cleere (1976). Both teams proposed that the conversion involves a formation of disulfide bonds from the vicinal sulfhydryl groups when the dehydrogenase converted into the oxidase. Proteolysis removes a peptide containing some free sulfhydryl groups which may function to stabilize the dehydrogenase formation, and thus results in an irreversible oxidase form (Coughlan, 1980). Most evidence relating to the interconversion between X0 and dehydrogenase are based on the investigation of liver X0. Battelli et 29 al. (1973) was the first to provide evidence that milk X0 possessed NAD‘ -reductive activity. They found both crude and purified milk X0 could be converted almost completely into the type 0 form by treatment with dithioerythritol or dihydrolipoic acid, but Only to a small extent by other thiols. Results from their kinetic studies and proteolytic treatment of the enzyme are similar to those reported previously on liver X0. Clare et al. (1981) treated xanthine dehydrogenase purified from various mammalian tissues with an immobilized preparation of crude bovine sulfhydryl oxidase. A Comparison of the rates of conversion of the 0 form to the 0 form in the presence and absence of the immobilized enzyme indicated that sulfhydryl oxidase catalyzes the conversion. The 0 form (dehydrogenase) of milk xanthine oxidase, obtained from purified milk X0 (type 0) treated with dithiothreitol was also enzymically converted back to the 0 form with the concomitant disappearance of sulfhydryl groups. They suggested that sulfhydryl oxidase may serve an important role in the formation of the 0 form of the enzyme in a given tissue. Biological Role and Applications X0 in mammals exists jn_vivg_as a dehydrogenase which performs a reduction-oxidation reaction in purine metabolism in the cell. Some microorganisms can utilize purines as their major source of carbon and nitrogen through X0/dehydrogenase reactions which may also balance the nucleotide pool (Coughlan, 1980). The real biological significance of X0 in mammals is still not clear since Johnson et al. (1974) found that rats grow and reproduce normally after feeding sufficient 30 tungsten to destroy all X0 activity. Fired et al. (1973) suggested that the main function of X0 is to provide a source of hydrogen peroxide and superoxide radical, both of which would then be available for coupling biological oxidation. Bray (1975) questioned the validity of this proposal since, as mentioned above, the enzyme exists as a dehydrogenase in 1112. However, Fired et al. (1973) pointed out that one advantage to the proposed mechanism is that a large amount of substrate can be oxidized by a proportionally small amount of oxygen. Part of the oxygen is regenerated from the hydrogen peroxide and/or superoxide by catalase and superoxide dismutase and can, therefore, be utilized for further reaction. These two enzymes are widely distributed and could be part of an overall reaction scheme whereby X0 participates as a generator of an oxidizing agent which is deactivated by specific enzymes. Based on this theory, Bjorck and Claesson (1979) suggested that X0 serves as a natural source of hydrogen peroxide for the lactoperoxidase system in milk. The lactoperoxidase system is proposed to function in the preventing of bacterial infections in the gastrointestine tract of the neonate (Reiter, 1978). The system catalyzes the oxidation of thiocyanate by hydrogen peroxide to hypothiocyanate (OSCN—), which is bactericidal for enteric pathogens including multiple antibiotic resistant strains of E. coli. Lactoperoxidase and thiocyanate are natural constituents of milk whereas hydrogen peroxide is not. It was found experimentally that hydrogen peroxide or superoxide which was generated by X0 reacting with low concentrations of free purines (hypoxanthine or xanthine) had little effect on bacteria. But indirectly, in combination with thiocyanate and lactoperoxidase, gave rise to a substantial 31 antibacterial effect. A decisive factor for the validity of this hypothesis is the availability of substrates for X0. The activity of X0 necessary for this function is less than half of that normally found in milk. This may be due to other functions of the enzyme in bovine milk. As previous discussion, X0 serves as an integral part of the intracellular milk fat globule membrane. Stabilization of fat globules by the enzyme may, presumably, be one of its biological roles in milk. Presumably, X0 contributes to spontaneous oxidized flavor in milk. Aurand and Hoods (1959) indicated that the occurrence of a spontaneously oxidized flavor in milk was dependent upon a high level of X0 activity. The use of an enzyme inhibitor prevented flavor development. They suggested that an intermediate product of hydrogen peroxide served as the oxidant for developing flavor. Smith and Dunkley (1960) did not find a high correlation between X0 activity and spontaneous development of oxidized flavor. X0 may be involved as a catalysis of oxidized flavor but it was not a limiting factor. Ascorbic acid and copper are the essential reactants for spontaneous flavor development (King and Dunkley, 1959; Smith and Dunkley, 1962). However, Aurand et al. (1967) found that milk with a level of X0 activity adjusted to greater than 120 ul uptake of oxygen per ml per hour developed oxidized flavor. Flavin-free xanthine oxidase added to raw milk at the same level had no effect on off-flavor development. They suggested that endogenous acetaldehyde was the substrate used by X0 to produce oxidized flavor. Pederson and Aust (1973) found that superoxide produced by X0 can form singlet oxygen which promotes the peroxidation of unsaturated 32 lipids. Kellogg and Fridovich (1975) also found that X0, acting aerobically upon acetaldehyde, caused the peroxidation of linolenate. Since superoxide and hydrogen peroxide can directly give rise to singlet oxygen, Aurand et al. (1977) proposed that superoxide anion produced by X0 in milk may undergo non-enzymic dismutation to form singlet oxygen which could catalyzes lipid oxidation. They reported that the addition of X0 with hypoxanthine in milk resulted in increased lipid peroxidation, whereas heat inactivated X0 added to milk inhibited the oxidation. More recently Allen and Hrieden (1982) investigated the lipid oxidation influenced by milk proteins and concluded that under normal condition X0 may not be a very significant factor in lipid autooxidation. However, a strongly pro-oxidation effect was found in the presence of 10 uM Cu+2 . More interestingly, the pro-oxidation effect of the enzyme was greatly enhanced by heat denaturation. They concluded that the role X0 plays in lipid peroxidation could be very important when cupric ions are added to milk for nutritional reasons or by contamination. The proteins of the MFGM are particularly effective at binding copper. Initiation of lipid peroxidation by singlet oxygen is commonly induced by light, metal and/or enzyme. X0 may involve the enzymic induction of lipid oxidation. However, superoxide dismutase, a native constituent of milk and usually considered together with X0 in the reaction system, would catalyzes the conversion of superoxide anion into triplet oxygen and hydrogen peroxide (Shipe, 1977; Hicks, 1980). Ascorbic acid in the milkwould be oxidized by superoxide anion prior to oxidation of lipid (Nishikim, 1975). Therefore, more research is required to determine the role of X0 in lipid peroxidation and 33 oxidized flavor development. It appears that non-enzymic as well as enzymic roles are played by the enzyme. Oster (1971) postulated that X0 in homogenized bovine milk is involved in the development of atherosclerosis in humans. According to his hypothesis, X0 survives through the gastrointestine tract and passes through the intestine barrier, especially the micronized droplets found in homogenized milk. After entering the circulatory system, X0 is deposited in the arterial wall and heart muscle. At these sites, the enzyme causes a depletion of the phospholipid plasmalogen, an important constituent of arterial wall and the myocardial cell, by oxidizing the aldehyde moiety-plasmal. Irreverisible removal of plasmal by X0 produces alterations in phospholipid balance and changes in the structure integrity of the cell membrane. Defective cell membrane causes failure in the active transport system and eventual cell death (myocardial infarction). In the vessel wall, cholesterol ester and other lipids accumulate at the site of injury as a compensatory repair mechanism and, hence, initiate the formation of atherosclerotic plaque. After Oster's presentation, several articles have been presented which refute this theory; others support it without actual evidences. Oster (1971, 1976a) assumed that human milk does not contain X0 and found a correlation existed between the titer of antibody to X0 in human sera and milk consumption. In fact, in 1977, Zikakis et al. reported the X0 activity in human milk. Both active and inactive forms of milk X0 elicit a similar antigenic response (Ultmann et al., 1962; Zikakis and Rzucidlo, 1976b). Hence, a high antibody titer does not necessarily indicate a high level of biologically active X0. The main 34 issue between these two opposite groups is whether X0 can be absorbed through the intestinal wall in an active state. Volp and Lage (1977) presented evidence that X0 does not retain its activity in the stomach and is not detectably absorbed by the small intestine. McCarthy and Long (1976) also reported that there was neither causal nor statistical significance between X0 activity in blood and average daily milk consumption, age, or sex. Mangino and Brunner (1976) found that X0 was completely inactivated below pH 3.2. Gastric juice has a pH of less than 2 and with a high volume to volume ratio to milk X0 activity is destroyed completely. However, Zikakis et al. (1977) found that the pH was reduced to 5.16 when simulated gastric juice mixed with milk with a ratio of 1:2 and 14.2% of the enzyme activity remained viable. Ho and Clifford (1976) reported that a very small amount of the active enzyme could reach the intestinal wall and was absorbed as intact enzyme. Evidences from animal studies of the intestinal absorption show that the size of proteins no greater that 80,000 molecular weight could be absorbed. Ho and Clifford (1976) found that only a very small amount of X0 with a molecular weight of 300,000 could be absorbed. Zikakis et al. (1977) indicated that the probability for X0 absorption may be greater if it can exist in the low molecular weight ((75,000) form which was isolated by Biasotto and Zikakis (1975). Oster (1971) believed that the small XO-containing fat globules ((1 u) in homogenized milk could pass through the intestinal wall. Clark and Pratt (1976) suggested that administration of half cream and half milk (H/H) supplemented with X0 may cause the absorption of intact X0 across the gastrointestinal tract. But it is inconclusive, since they 35 also found that H/H stimulated endogenous X0 activity. Gandhi and Ahuja (1979) indicated that X0 is not destroyed completely in gastrointestinal digestion and is absorbed when associated with milk fat globules. Ross et al. (1980) demonstrated that absorption of X0- containing particles in homogenized milk was linked to liposomes as a vehicle for the persorption. Hence, they suggested that active X0 could be readily absorbed in this form. It appears that intact, active X0 may be absorbed, at least in small amounts, and enters the circulatory system. However, the initiation of atherosclerosis by this absorbed X0 is still questionable. Ho and Clifford (1977) found that large intravenous doses of x0 over a prolonged period did not deplete arterial or coronary tissue plasmalogens and did not induce the formation of arterial plaques. Though the evidence is issufficient and unsatisfactory to support Oster's hypothesis, it caused some concern among physcians, dairy manufacturers and consumers. Most dairy products with the exception of powdered and evaporated milk products and all butter contain detectable X0 activity (Zikakis and Hooters, 1980). Deeth (1981) published a critical review on homogenized milk and coronary heart disease suggesting that milk X0-induced atherosclerosis deserves further examination. He concluded that until experimental evidence is presented which proves that absorbed milk X0 causes tissue damange directly related to atherogenesis the Oster hypothesis cannot be accepted. The suggested food applications of X0 are limited. Actual food application of the enzyme are unknown due to a wide range of substrates and their low specificity for the enzyme. 36 Alfa-Laval (1977) indicated that native X0 in milk could generate hydrogen peroxide upon the addition of hypoxanthine (0.5% w/w). This enzyme reaction could improve the keeping quality of milk through cold pasteurization and has been suggested as a method for preserving milk on the farm. Xanthine oxidase can be used as a marker enzyme to assay for the churning of milk. The release of X0 from the fat globule into skim milk indicates that churning has occurred (Stannard, 1975). Groman and Groman (1975) suggested that quantitation of X0 in the skim milk due to agitation could be used for assessing the correctness of stirrer choice for dairy storage and fermentation tanks. The level of hypoxanthine in fish muscle is an index of fish quality or freshness. Burt et al.(1968) introduced an automatic colorimetric method to measure the hypoxanthine concentration in fish tissue by the X0 reaction. Uric acid produced by X0 induced a color change in the redox dye 2,6-dichlor0phenolindophenol. Jahns et al. (1976) developed a rapid and simple semi-quantitative test utilizing dry strips containing X0 and dye, a sort of "litmus" test. The freshness of fish products was estimated by a color change of the stripe within a short time. EXPERIMENTAL Materials and Chemicals The milk used in this study was obtained from Holstein cows of the Michigan State University dairy herd. No attempt was made to collect milk from a specific cow except for the riboflavin study. The principal chemicals and their sources used in this study are listed in the Appendix, Table A1. All chemicals are reagent grade unless otherwise specified. Deionized water was used in the preparation of all buffers and solutions. Equipment regularly used in the course of this study is listed in the Appendix, Table A2. Instrumentation specific for a certain experiment will be referred to the appropriate section. Preparative Procedure Fractionation of Whole Milk Milk was collected inmediately after milking and separated as soon as possible at 40-45 C with a Westfalia separator. The cream was collected and washed by adding three volumes of deionized water at 40- 45 C followed by gently mixing and reseparation. This washing step was repeated three more times to ensure adequate removal of casein and whey protein from the cream. Following storage overnight at 4 C, the washed cream was churned at room temperature. After the fat emulsion was broken, the aqueous phase was filtrated through four layers of cheesecloth. Unchurned butter granules were removed by centrifugation at 1,000xg for 20 min. The resulting aqueous suspension served as 37 38 buttermilk for subsequent fractionation and analysis. Butter oil was prepared from churned, washed milk fat globules. The churned fat was melted at 55 C and washed with 55 C deionized water. The mixture was centrifuged at 1,000xg for 15 min to separate butter oil from the aqueous phase. Butter oil was stored at cold room for further use. Whole casein was prepared by adjusting pH of skim milk to 4.6 with 1N HCl. The casein precipitate was collected and washed 2 times with a volume of deionized water equal to that of the original skim milk. Washed casein was resuspended in deionized water by adjusting pH to 7.0 with 1N NaOH and reprecipitated at pH 4.6 with acid. After washing 2 times with deionized water, the casein was solubilized, dialyzed against deionized water at‘4 C for 48 hr and lyophilized. Whey protein was obtained by dialyzing the milk serum after the removal of casein at 4 C for 48 hr and dried by lyophilization. Isolation of Xanthine Oxidase The differential centrifugation method of Mangino and Brunner (1977a) with slight modifications was employed to isolate bovine milk X0. A Beckmen L2-65 preparative ultracentrifuge and a Type 30 rotor were used. Buttermilk obtained as previously described was centrifuged at 27,500 rpm (80,000xg) for 150 min at 10 C. The pellet-material (1005) was purified by additional centrifugations and by hydroxylapatite adsorption chromatography. The amount of wet hydroxylapatite gel added to the extract was 50 mg/ml. The purified X0 was stored at 4 C in 25% (w/v) ammonium sulfate solution pending further analysis. When required, the enzyme was lyophilized after dialysis against deionized 39 water at 4 C for 24 hr. Isolation of Polypeptide Fragments of X0 Cleaved by Trypsin Twenty electrophoresed SDS-gels of trypsin digested-X0 were sliced into segments corresponding to a reference gel from the same run which was visualized with Coomassie Blue R-250. Gel segments representing discrete polypeptide bands were combined and diced into small pieces. Polypeptide was extracted by overnight soaking diced gels with boiled deionized water containing 0.05% 505 at 37 C. The mixture was centrifuged and the gel pellet was washed three times with the same solution. The clear supernatant fractions were combined, dialyzed and dried with vacuum evaporation in preparation for N-amino terminal analysis. For amino acid composition analyses, a homemade plexiglass electrophoretic elution apparatus was used to elute the polypeptide from SDS-gels (see Figure 1). About 100 SDS-gels were cut to correspond with stained zones in a reference gel pattern. Gels representing each band were combined and immersed in a small volume of boiled deionized water to avoid dehydration of the gels before subsequent elution. A layer about 0.3 cm thick of 1.5% agarose in veronal buffer, pH 8.6, ionic strength 0.02, was formed at the bottom of the elution cell. Polyacrylamide gel pieces were dispersed on top of the agarose layer. The free water phase was mixed with agarose and veronal buffer to make a final solution containing 1.5% agarose solution which was poured into the elution cell, covering the PAC fragments to a depth of 1.5 cm. About 10 ml of veronal buffer was layered on top of the gel for collecting eluted polypeptide. The bottom piece of the elution 40 Upper buffer chamber Rubber ring Elation cell Upper butler Collecting buffer Membrane [100001100 2323:? \— Lower butler lower buffer container Bottom plate with holes <)()(> cowumuccpzaa accruagm auw>Puo< evapoea me=_o> mmmuwxo mcvzucmx xppe mo cowumuFLnga .H wpnmh 56 chromatography for obtaining the best enzyme preparation is recommended. The purified X0 did not contain protease activity. This was significant because proteases alter the characteristics of the enzyme. Milk contains an endogeneous protease, plasmin, which affects the enzyme like other proteases during isolation. Data reported in Table 2 show the results of incubating purified X0 with 1% and 5% plasmin at 37 C for up to 24 hr. Within 10 hr, X0 activity did not change significantly after treatment and showed only a slight decrease after 24 hr. Though the dehydrogenase activity was not measured here, the data indicate that the enzyme activity would not be affected by plasmin. In fact, the concentration of plasmin in milk is much less than that used here. Also, the purified X0 still possessed a high specific activity with a molecular weight of 151,000 and a detectable dehydrogenase activity with a capacity of conversion to dehydrogenase (see discussion below). It was concluded that milk endogenous protease did not influence the purification of X0. Della Corte and Stirpe (1972) suggested that xanthine dehydrogenase could be converted to an intermediate form by proteolysis (see Figure A2 of Appendix). It appears that milk plasmin could affect XO only during the time milk was held in the mammary gland. Thus, the addition of protease inhibitors during the isolation procedure is not useful since proteolysis of XO-if indeed there was any-already occurred before isolation. Fortunately, the observed properties of the purified X0 indicated that the time from milk secretion to the end of purification was not long enough to alter the enzyme properties. Possibly, the activity of plasmin was directed to more susceptible substrates, the 57 Table 2. Effect of plasmin 03 the xanthine oxidase activity of the purified enzyme Activity (IU/ml) Time (hr) Control 1% plasmin 5% plamin 0 0.125 0.124 0.121 0.5 0.123 0.124 0.123 1 0.125 0.124 0.121 2 0.124 0.121 0.117 3 0.128 0.121 0.119 4 0.127 0.127 0.116 5 0.123 0.125 0.117 6 0.124 0.122 0.117 7 0.123 0.129 0.119 8 0.122 0.126 0.118 9 0.124 0.126 0.118 10 0.122 0.120 0.117 24 0.112 0.109 0.100 a. Using the simulated milk ultrafiltrate as the incubation buffer (Jenness and Koops, 1962). Incubation temperature was 37 C. 58 milk caseins. Criteria of Purity of X0 The specific activity of X0 prepared for this study was 3.59 IU/mg. Its spectral properties were: A(280nm)/A(450)=4.84, A(lcm, 280nm, 1%)=11.9 and activity/A(450nm)=141. Compared to the published data (see Table 3), these values agree favorable with the results of Hart et al. (1970), Massey et al. (1969) and Mangino and Brunner (1976). A decrease in the concentration of non-X0 protein (at 280nm) and a simultaneous increase in the concentration of X0 (at 450nm) should give an increasingly lower PFR value (Zikakis, 1979a). In other words, the lower the PFR value, the higher the purity of the enzyme preparation. Thus, the enzyme preparation obtained here is purer than preparations reported by other groups. By SOS-PAGE, purified X0 revealed a single band with a molecular weight of 151,000 1 4,000 (see Gel C in Figure 2 and Figure 3). The estimated molecular weight of X0 is 302,000 since X0 contains two identical subunits. This value is in the range of 270,000 to 370,000 reported for the enzyme from various sources and is in excellent agreement with the generally accepted molecular weight of 300,000 (Bray, 1975). The purified enzyme revealed three zones in Disc-PAGE gel patterns (Figure 1, gel 0). All three bands showed positive staining for enzyme activity and a single band with an estimated molecular weight of 151,000 in SOS-gel (Figure 1, gel E). Nathans and Hade (1978), employing a 7.2% (T) gel, obtained a single band by disc-PAGE. However, they suggested that the use of size as the primary criterion of homogeneity would minimize the problems encounterd when charge or cowu=FPu an» an wexncm an» .Anmemfl .mcxmx_~v p_== ucgamemoeopoq eoec caua_=a_mu .u .xammo on» yo Lac Asufi .seom¢v< an uauw>cu .=*E\As=mmmv< .»p_>wpum mm um=Pcmo .n .Ascomev<\fie=owuv< co oppme a an um=_cma .a ¢.HH HQH em.¢ mm.m xusum m_;h murgoaapaxocuag a copummsepcucmu ---- mafi o.m --- Aflmmfiv .Fo pa oc_;mcz agaoemouasoegu auccpcc< ---- --- H.¢ oc.m Aamkafiv mwxox_N amcoguxa-=o. a 832:: .8 ---- --- o.4 o.~ hmsmfiv mum: new mcmguaz cozpceupscoeupa mu ~¢.HH --- mo.m mm.m ANNmHV emecaem ecu o=_m=az a»_p~qa_xxoLuxg a cowummacwgucmu n--- mmH in- o.m Amumfiv .—o no comucoEum annucmouosocgu xuccccc< m.~H oefi ~.m m.m Aoumav .pc no age: cowpmcsumcwu muo_»uppam m.HH msfl ¢.m m.m Ammmfiv ._~ “a sumac: crummeuema m.HH ON ~.m --- Anomafl .mmmHV .Fa um m_>< =o.uo~___mumaeu vco crummgucmm Aafi .euH .scom~V< nma< «mam me\.=._ mueaemcaa =o_umeaaaea mmmuvxo mcvzpcmx x—PE um'cwcaa uo mm'ucoaocq pmguuoqm new aup>yuu< .m open» 60 Figure 2. Electropherograms of purified X0. (A) low MW protein standard in SOS-PAGE, (8) high MN protein standard in SOS-PAGE, (C) purified X0 in SOS-PAGE, (D) purified X0 in disc-PAGE, (E) bands cut from (C) in SOS-PAGE. SDS- gel is 9% T and 2.6% C in running gel and 5% T with 2.6% C for running gel of disc-PAGE. 61 N 059". 62 Figure 3. Standard curve for the estimation of molecular weight in SOS-PAGE. Protein standards used for this plot are: thyroglobulin (330,000), ferritin (half unit, 220,000), phosphorase b (94,000), albumin (67,000), catalase (60,000), ovalbumin (43,000), lactate dehydrogenase (36,000), carbonic anhydrase (30,000), trypsin inhibitor (20,100), and alpha- lactoalbumin (14,400). Arrow indicates the relative mobility value of purified X0. SOS-PAGE (9%T and 2.6%C) is according to the method of Laemmli (1970). 63 activity were employed as criteria. Thus, the enzyme purified for this study was very close to 100% pure as estimated from spectral analyses as well as electrophoretic analyses. The three bands observed in disc- PAGE gel patterns could be explained by either genetic polymorphism or by an association-dissociation phenomenon which occurred during electrophoresis, resulting in species differing in charge but not in size. They may also be isozymes of X0. In fact, Jarasch et al. (1981) found that X0 showed three isoelectric variants with the same molecular weight in two dimensioned gel electrophoresis of MFGM. Enzyme Nature Kinetics Figure 4 represents the Lineweaver-Burk plot of the enzyme reaction with different xanthine concentrations. Treatment of these data by linear regression yielded a Km=1.61:0.08X10"'5 M and V...“ =3.68:0.12 IU/mg which compared favorable with similar values obtained by Mangino and Brunner (1976a) and Massey et al. (1969). Xanthine Dehydrogenase Properties Purified X0 possessed a detectable NAD‘Ldependent reductase activity (see Table 4). After incubated the enzyme with 10 mM DTT or 1% ME at 37 C for 20 min, the dehydrogenase activity increased about 23 folds and about 20% of the X0 activity was retained when oxygen was the electron acceptor. This indicated that approximately 80% of purified X0 could be converted to xanthine dehydrogenase. A comparison of uric acid production by the enzyme treated as above, using oxygen and oxygen-NAD'I as electron acceptors, indicated that NAD+'masked or 64 Figure 4. Lineweaver-Burk plot of xanthine oxidation by X0. Reaction mixture contained 0.1 M perphosphate, pH 8.3, 10.6-212 mM xanthine and 0.16 mg X0. l/S is reciprocal units of xanthine concentration and 1/V is reciprocal units of velocity in micromole/min. 65 Table 4. Conversion of xanthine oxidase-dehydrogenase activities of purified X0 by dithiothreitol (DTT), mercaptoethanol (ME) and chymotrypsin Activity (IU/ml) Treatment Acceptor oxygena oxygen+NAD+a NAD+ b X0 0.112 0.112 0.002 X0+DTT ° 0.023 0.072 0.050 X0+ME° 0.024 0.069 0.046 X0+chymotrypsi a“ 0.114 0.110 0 X0+chymotrypsi n+DTT ° 0.100 0.100 0 a. Measurement of uric acid formed at 295nm, IU/ml=micromole xanthine oxidized per ml. 0. Measurement of NADH formed at 340nm, IU/ml=micromole NADH formed per ml. c. The enzyme was incubated at 37 C with 10mM or 1% ME of final concentration in 0.1M phosphate buffer, pH 6.8, for 20 mim. d. X0 was incubated with chymotrypsin, 10ug/ml, in 0.1M phosphate buffer, pH 6.8, at 37 C for 45 min. e. After incubation with chymotrypsin at 37 C for 45 min, DTT was added to the final concentration of 10 mM in the mixture and proceeded the incubation for another 20 min. 66 influenced the reaction of FAD with oxygen. The reduction in the amount of uric acid produced by treated enzyme in the oxygen-NAD+ assay when compared to untreated enzyme is attributable to the product, NADH, which serves as a competitor of NAD+ for binding the FAD molecule (Bray, 1975; Battelli et al., 1973). The interconvertible properties of the enzyme was lost when it was treated with chymotrypsin. Battelli et al. (1973) isolated a convertible X0 from milk by treating fresh milk with ME. They obtained approximately an 8-fold increase in xanthine-NAD‘ reductase activity after treating the isolated enzyme with 10 mM DTT. It was suggested that X0 occurs as a reversible oxidase in milk and could not be obtained as a dehydrogenase. Though no attempt was made to separate xanthine dehydrogenase from the purified X0, data from this study showed that purified X0 is not only a reversible oxidase but also possesses dehydrogenase activity. Additionally, no proteases and/or reducing agents were added in the isolation procedure. Thus, it is concluded that X0 in milk is capable of interconvertibility between D and 0 forms, as is the case for the enzyme isolated from other mammalian sources, and that the reversible 0 form is the major state of the enzyme in fresh milk. Also, it is postulated that X0 in the mammary gland exists in the dehydrogenase form as it existes in the liver. Somehow, it is converted into a reversible and/or irreversible oxidase form. The postulated conversion may effected by (1) a redox reaction in the presence or absence of glutathione (6886) in the cell, (2) by endogenous protease in the cytoplasm of the secretory cells, (3) by sulfhydryl oxidase when fat globule complexes reach the apical plasma 67 membrane of secretory cells during the secretory process, and/or (4) all or anyone of the above possibilities during storage of milk in the mammary gland. Active Monomer During the investigation on the effects of dissociating agents on X0, it was observed that the enzymic activity remained after the purified X0 was treated with 6 M urea in 0.1 M phosphate buffer, pH 6.8, containing 5 mM salicylate. Since urea is generally used to dissociate the quaternary structure of proteins, it is possible that the active species is the monomer (150,000) of X0 rather than the dimer (300,000). Haud et al. (1975) found that 8 M urea, 4 M urea or 0.5% Triton X- 100 in pH 8.9 polyacrylamide gel electrophoresis did not dissociate the enzyme into its subunits or fragments. The enzyme released its iron moieties and was denatured after exposing to 6.25 M urea at pH 6.25 during electrophoresis. In the presence of 6 M guanidine-HCl, he found that the enzyme dissociated into a fragment possessing a molecular weight of 157,000. Since guanidine-HCl is known to be a strong protein denaturant, it is understandable why this fragment did not exhibit activity. In this study, it was determined that gel patterns of purified X0 did not change when disc-PAGE was performed in the presence of several dissociating agents. However, the bands showing enzymic activity migrated further after the purified X0 treated with 6 M urea. Since disc- and SOS-PAGE did not resolve the urea-dissociated enzyme into its active subunits, gel filtration chromatography was employed to examine whether the enzyme could be dissociated into an active monomer with 6 M urea. 68 Figure 5 represents a chromatogram of urea treated enzyme eluted from a Sephacryl S-200 column which was calibrated with standard proteins of known molecular weight. From the elution pattern and standard curve, it is apparent that the active species was eluted as a single peak close to the IgG peak, possessing an estimated molecular weight of 160,000. SOS-gel patterns of eluted active fraction revealed a single band with molecular weight of 155,000 which is similar to that of the untreated enzyme. Table 5 summarizes the results of analysis of X0/dehydrogenase conversion of the eluted enzyme monomer. No uric acid formation was detected when the eluted enzyme samples were in the assay in the absence of xanthine. It is, therefore, confirmed that urea is not a substrate of X0 and does not influence the results of subsequent treatments of X0. There was no NADH form in the urea-treated enzyme. Nevertheless, a low dehydrogenase activity was obtained after the eluted species was treated with 10 mM DTT or 1% ME. This result may be due to the presence of urea, a strong dissociating agent, which does not allow NAD+ to reach FAD molecule while DTT or ME could mitigate this effect. The significance of the data is not so much that of detectable dehydrogenase activity, but they indicate that the active monomer retained the capacity to convert oxidase to dehydrogenase. Although urea was present throughout the enzyme samples during analysis to maintain a dissociated state, the results provided direct evidence in support of the hypothesis that the monomer of X0 is active. Inhibition by Chloroquine Eigel (1980) found no reduction of X0 activity in MFGM prepared in 0.03 a O a N 0.01 AC'iVi'Y , AA295 llTIil'I Figure 5. 69 Bovine milk lgG (106,000) Bovine serum albumin (57.000) Ovalbuniin(43,000) 0" o a Chymotrypsln A (25.000) Ribonuclease A (13,700) .“ log MW Fraction number Gel filtration chromatogram on Sephacryl S-200 of urea- treated X0. Fractions of 4.6 ml were collected. Activity of each fraction was measured by enzyme assay at 295 nm. Arrow indicates the void volume. Upper straight line indicates calibration of the column by standard proteins. 70 Table 5. Conversion of xanthine oxidase-dehydrogenase activities of active X0 monomer by DTT and ME Activity (IU/ml) Treatment Acceptor oxygena oxygen+NAD+a NAD+b Active monomer without xanthinec 0 0 0 Active monomer with xanthinec 0.136 0.149 0 Active monomer+DTTd 0.103 0.119 0.005 Active monomer+MEd 0.114 0.117 0.006 a. Micromole xanthine oxidized/ml, (IU/ml), by measuring uric acid at 295nm. b. Micromole NADH formed/ml, (IU/ml), by measuring NADH at 340nm. 6. Only enzyme species to check the effect of urea in the assay. d. Enzyme species was incubated at 37 C with 10mM DTT or 1% ME of final concentration in 0.1M phosphate buffer, pH 6.8, for 20 min. 71 .wczuxvs cowuummc on» cw zE om.H ea c soc; mcpmcmc .Apv mcwzcoeo—zu no wucmmmca an» :_ ox an cowumuwxo acmnucmx uc papa xcamicm>emzo=vb .o «camwd o 28%... SEE.— » _ SE2”.— u _ 72 the presence of protease inhibitors such as chloroquine and Kallikrein inhibitor. During the investigations on the increase of X0 activity in milk stored at different temperatures, chloroquine and/or aprotinin were added as protease inhibitors to prevent the proteolysis of X0. It was found that X0 activity decreased after the addition of inhibitor(s). This observation raised the question concerning the possible effect of these inhibitors on the enzyme. The enzyme activity of purified X0 did not change in the assay mixture containing aprotinin, whereas a 20% decrease in activity was found after adding chloroquine to the assay mixture. This suggests that chloroquine may inhibit the reaction to a limited extent or may be a substrate competitor of the enzyme. Figure 6 shows the 1/V vs 1/S plot of the study of enzyme kinetics in the presence of different fixed concentrations of chloroquine. The intersection of the plotted data below the 1/S axis indicates that chloroquine acts as a hyperbolic (partial) mixed-type inhibitor in the enzyme reaction (Segel, 1975). No further analysis of this inhibition was attempted since the evidence is sufficient to indicate the influence of chloroquine on the enzyme. Chemical Compositions Lyophilized, purified X0 contained 14.8% protein nitrogen which is equilvalent to 92.5% protein based on ly0philized weight and a factor of 6.25. No lipid was found after staining the disc-gel of purified X0 with lipid staining. Table 6 compares the amino acid composition of bovine milk X0 73 Table 6. Amino acid composition of four preparations of bovine milk X0 Amino Acid Content (mole %Q Bray and Nelson and Mangino and this Residue Malmstrom Handler Brunner (1964) (1968) (1977a) study Lysine 6.8 6.9 6.6 6.8 Histidine 2.3 2.3 2.3 2.3 Arginine 4.4 4.7 4.9 4.3 Aspartic Acid 8.4 8.6 8.9 8.4 Threonine 7.1 7.0 6.9 6.7 Serine 6.5 6.5 6.3 5.7 Glutamic Acid 10.2 10.0 10.5 9.5 Proline 5.5 5.5 5.3 5.6 Glycine 8.2 8.2 8.2 8.0 Alanine 7.5 7.6 7.5 7.3 Half Cystine 2.6 2.7 2.7 3.5 Valine 6.9 6.8 6.7 7.0 Methionine 2.0 2.2 2.9 3.0 Isoleucine 5.0 4.8 5.2 5.0 Leucine 8.7 8.9 8.8 9.1 Tyrosine 2.6 2.4 2.5 2.1 Phenylalanine 5.0 4.9 4.1 5.0 Tryptophan 0.4 0.4 0.4 0.6 SAQ(this study vs) 3.66 2.87 3.90 74 prepared in this study with values reported by Bray and Malmstrom (1964), Nelson and Handler (1968) and Mangino and Brunner (1977a). The values of SAQ calculated by comparing the data of this study with that of the latter three groups were 3.66, 2.87 and 3.90, repectively. These values indicate that the analyses are in excellent agreement since a value of 4 was considered by Heltman and Dowben (1973) as the limit of analytical precision between different laboratories for the same protein. Comparing the data of the individual amino acid residues showed a general agreement, except a high value of half-cystine for the enzyme preparation in this study. Common to all is the extreme low value for tryptophan which is characteristics of the enzyme (Bray, 1975). Based on tne data of mole percentage of each amino acid residue, the ratio of the acidic residues, Asp and Glu, to the basic residues, Arg, Lys and His, is approximate 1.3. The ratio of the hydrophobic residues, Leu, Ileu, Val, Pro, Phe and Net, to the hydrophilic residues, Tyr, Asp, Glu, Lys and His, is 1.2. This illustrates that bovine milk X0 possesses an acidic but hydr0phobic nature. The average hydrophobicity of purified enzyme calculated by the method of Bigelow (1967) is 1135 cal which is 5% higher than 1074 cal obtained by Mangino and Brunner (1977b). Values of SAQ and average hydrophobicity is commonly used to examine the relatedness of various proteins. Generally, the discrimination of the 530 method is better than that of the average hydrophobicity procedure. Mangino and Brunner (1977b) employed both methods to demonstrate the relatedness among selected proteins including X0 according to the following criteria: (1) related 75 proteins, in general, have similar average hydrophobicities (Bigelow, 1967), (2) value of $30 less than 50 for paired comparison indicates a high degree of primary structural homology of proteins (Marchalonis and Neltman, 1971). They deduced that bovine milk X0 exhibits a high degree of compositional relatedness to membrane ATPase, actin and other contractile proteins. Although many of these proteins can be related to purine metabolism or recognize the purine ring, some of them such as acetylcholinesterase and phosphatase have no connection with purine metabolism. However, all of the proteinS'were either membrane-associated, or form an insoluble complex in water. Thus, they suggested that these proteins may be "actin-like" in composition but not in primary sequence or function. Their relatedness may be due to the evolutionary convergence toward similar amino acid compositions capable of interacting with the environment such as a lipodial membrane. It is more important to consider here the hydrophobic nature of X0 and its compositional similarity to the membrane-associated proteins than to compare the homology of the primary sequence between proteins. This characteristic could offer logical explanation why the enzyme is concentrated in the cream phase and MFGM in milk and can be entrapped in liposomes for intestinal absorption. The specific volume of the enzyme calculated from its amino acid composition was 0.736 which is in excellent agreement with values of 0.737 and 0.74 reported by Avis et al. (1956a) and Andrews et al. (1964), respectively. The data in Table 7 represent the results of analyses for available (exposed), total (unexposed and exposed) and reduction- 76 Table 7. Sulfhydryl content of xanthine oxidase Component No. of -SH per mole of X0. Available -SH 4 Total -SH 38 Total -SH after reduction of S-S 82 Total half-cystine content (-SH) from amino acid composition: This study 88 Bray and Malmstrom (1964) 73 Nelsons and Handler (1967) 76 Mangino and Brunner (1977a) 75 Nalger and Vartanyan (1973) 120 a. Based on molecular weight of 302,000. 77 induced sulfhydryl groups in the purified X0. There were 4 available sulfhydryl groups per mole of the enzyme (300,000) in the native comformation. The enzyme, after exposure to 0.2% SDS, revealed a total 38 detectable sulfhydryl groups which indicated that 34 sulfhydryl groups were buried inside the molecule. After treating X0 with a strong reducing agent, sodium borohydride, a total of approximate 82 sulfhydryl groups per enzyme molecule was determined. Difference between the number of sulfhydryl groups before and after reduction of the enzyme suggested that there were 22 disulfide bridges distributed throughout the interior of the molecule (subunits). From the ratio of moles of half-cystine/moles protein, using a molecular weight of 302,000, it was calculated that the purified X0 contained 88 total half cystine residues (-SH groups) per mole which is close to 82 obtained by chemical analysis. Only the total half-cystine content has been reported which was based on amino acid composition of X0 (see Table 7). The data of the total half-cystine residue or sulfhydryl groups obtained in this study, either 88 or 82, is higher than the 73, 76 and 75 residues calculated from the data of amino acid compositions of Bray and Malmstrom (1964), Nelson and Handler (1968) and Mangino and Brunner (1977a), respectively. Nalger and Vartanyan (1973) obtained a value of 120 semi-cystine residues which is higher than any one of the above values. They also studied the kinetics of the reaction of PCMB with SH groups of X0 in the pH range of 6.9-12.9 at 22 C. Approximately 30% of all SH groups were found to react slowly with PCMB only at pH>11.7. Since at high alkalinity disulfide bond-splitting is accompanied by degradation of protein, the possibility exists that the reactive SH ‘Illl-ll filliiti‘vi 78 groups were generated by the dissociation of disulfide bonds at pH>11.7. Approximately 19 disulfide bonds could be calculated from their data which approximates the 22 disulfide bridges obtained in this study. The significance of these data is not so much the numbers of sulfhydryl and disulfide groups found but the observation that disulfide bonds exist in the enzyme molecule. Haud and Rajagopalan (1976) and Coughlan (1980) suggested that many of the free sulfhydryl groups of the dehydrogenase form of X0 are oxidized to disulfide bonds during the conversion of the dehydrogenase form to oxidase form. Battelli (1980) converted Histar rat liver xanthine dehydrogenase to the oxidase form in the presence of glutathione (0350) and a crude fraction obtained from rat liver homogenates. He reported that bovine milk sulfhydryl oxidase failed to catalyze the conversion and suggested that the conversion is via a thio-disulfide interchange mechanism. In contrast, Clare et al. (1981) found that purified milk X0 form 0 could be converted to the 0 form when catalyzed by purified milk sulfhydryl oxidase. They suggested that the conversion of dehydrogenase to oxidase of milk X0 could be via a gg_ngxg_disulfide bond generation catalyzed by sulfhydryl oxidase. Regardless whether by thio-disulfide interchange or gg_ngvg_disulfide formation, disulfide bonds were formed from sulfhydryl groups during the conversion. Clare et al. (1981) reported the disappearance of the protein's sulfhydryl groups concomitant with the conversion of enzyme activity. Thus, the content of disulfide groups could be a special characteristics of the enzyme and may be used as a tool to estimate the purity, enzymic state and/or degree of the conversion of the enzyme. 79 The techniques applied here could not distinguish between intra- and inter-molecular disulfide bonds. However, results and discussion with regards to effects of dissociating agents on purified X0 indicate that the disulfide bonds could be considered to be the intrachain type because treatment with ME did not change the electrophoretic patterns of X0. Data presented in Figure 7 show the results of chromatography of the N-terminal amino acid of the purified enzyme. Four solvents were used as described in the EXPERIMENTAL section. Solvent 3 and 4 were run in the same direction as solvent 2 and were designated as dimension 3 and 4, respectively. The first two dimensions separated most amino acids but not lysine, histidine and arginine which remained in the same position. Dimension 3 separated histidine from lysine and arginine while dimension 4 distingished arginine from lysine and histidine (Heiner et al., 1972). Thus, the four dimensioned chromatography run can resolve lysine, histidine and arginine (Figure 6A). The acid hydrolysate of dansylated purified X0 showed the same chromatogram as that of the dansylated lysine standard run on the reverse side of the micr0polyamine sheet (Figure 68). Therefore, lysine appears to be the N—terminal amino acid of milk X0. Effects of Dissociating Agents The effects of five dissociation conditions-(1) ME, (2) urea, (3) urea plus ME, (4) SDS and (5) Triton X-100 in combination with each of former four conditions-on the enzyme was studied by PAGE. Gels for SOS-PAGE contained 9% (T) with 2.6% (C) while gels with 5% (T) and 80 Figure 7. Chromatograms of dansylated amino acids on micropolyamide sheet. (A) dansylated standard amino acids, (8) dansylated N-terminal amino acid of X0. Legend: A-arginine, H-histidine, NH -dansyl NH , 0H- dansyl 0H, L-lysine. See text for dgscription 8f solvent system. 8'l N O NH3 E a» at H l .2 . -<> 0 1 ,3 OH A o ) NH3 cl. '2 :4 fl 0 H c: O 2 1 '6 ' OH O a, OI. ‘- of °". ° N O A ‘5 OH g - «S -— Solvent 1 Figure NH3 OI. - <3 01 on 0 Mia cl 0 ' OH 0|. 01 o 1 CD OH O -——-—>- 7 82 Figure 8. Electrophoretic patterns of purified milk X0 in the presence and absence of various dissociating agents: A, no reagents (5%T); 8, sample equilibrated against 10 mM (1%) ME (5%T); C, sample equilibrated against 6 M urea (5%T); 0, sample equilibrated against 6 M urea plus 10 mM ME (5%T); E, with 505 according to Laemmli (1970) (9%T). 83 7* Figure 8 84 2.6% (C) were used for other electrophoreses. The results are shown in Figure 8. Except for SDS, the treated enzyme showed a positive reaction to enzyme activity-staining after electr0phoresis. Gel 1 represents the undissociated X0 with three enzymically active bands- one major and two minor zones. After purified X0 was exposed to 10 mM (1%) ME for 20 min, disc-gel patterns revealed a single active band (Gel 2). ME can convert the reversible 0 form of X0 to 0 form which possessed different electrophoretic mobilities than the irreversible 0 form (Battelli et al., 1973). It appears that the three bands in gel 1 indicate that the enzyme preparation is a mixture of these two forms as was evident from previous studies on the nature of its activities. Gel 3 represents the pattern of purified X0 equilibrated with 6 M urea and subsequently applied to the gel system containing 6 M urea. Three positive activity-staining bands were observed which migrated further than those in gel 1. The addition of ME consolidated these bands into a single enzymic active band (Gel 4) with higher electr0phoretic mobility than observed in the gel 2 pattern. The faster migration of protein bands in urea-gel may be due to the differences of size of the protein since urea could dissociate the enzyme into active monomer, resulting in a high charge to size ratio-electrophoretic mobility. SDS-gel patterns of the enzyme sample revealed a single polypeptide band (Gel 5 in Figure 8). A comparison of the relative mobility of the band with those in the standard curve (see Figure 2) revealed a molecular weight of approximate 151,000. The value falls in the range of 150,000-155,000 reported from SOS-PAGE estimation for milk X0 (Nalger and Vartanyan, 1973; Haud et al., 1975; Mangino and Brunner, 1977a; Frendenstein et al., 1979). It is in excellent 85 agreement with the value of 150,000 commonly recognized as a monomer of X0 from various sources (Bray, 1975). Addition of 1% Triton X-100 to each of the treated specimen had no effect on enzyme activity. Results of PAGE following the addition of Triton X-100 showed similar gel patterns to that of corresponding treatments without Triton X-100, see Figure 8. These data indicate that Triton X-100 does not affect the enzymic activity or physical characteristics. Effects of Limited Proteolysis Results showing effects on enzyme activities by limited proteolysis with trypsin, chymotrypsin, plasmin, pancreatin, pepsin and papain are presented in Figure 9. Oxidase activity was resistant to proteolysis. The six proteases investigated did not change the ability of purified X0 to convert xanthine to uric acid with oxygen or neotetrazolium as an electron acceptor. However, following proteolysis purified X0 lost its xanthine dehydrogenase activity with NAD7'as an electron acceptor. Furthermore no NADH formtion could be detected in the dehydrogenase assay when proteolyzed X0 was treated with 1% ME or 10 mM DTT. Nalger and Vartanyan (1976) treated milk X0 with trypsin, chymotrypsin and subtilisin at pH 8. They found that the enzyme retained its oxidase/dehydrogenase activities and the ability to transfer electrons from xanthine to other acceptors such as tetranitroblue tetrazolium, phenazine methosulfate and indophenol. They also found that there was no effect on the catalytic properties 86 .02 53 LAW/min l 2 3 4 Time, hr Figure 9. Effect of limited proteolysis by various proteases on the oxidase/dehydrogenase activities of X0. Solid lines indicate oxidase activity after treatment of X0 with: a-trypsin, b- papain, c-plasmin, d-chymotrypsin, e-pancreatin, and f- pepsin. Dotted line indicates the dehydrogenase activity of protease-treated X0. 87 of the enzyme at pH 10.7, whereas the activity reduced gradually and remained at 5-6% of initial activity after incubating X0 at pH 11.0 to 11.2 for 24 hr. The ability of the enzyme to transfer electrons to other acceptors mentioned above was reduced three folds. This behavior is attributed to the dissociation of FAD from the enzyme at high pH. The enzyme specimen digested by subtilisin at pH 11 retained the ability to oxidize xanthine and to transfer electron to the above mentioned acceptors. This characteristic was completely lost when X0 was digested by subtilisin at pH values higher than 11.3. Thus, the above evidences led to the conclusion that X0 activity is very resistant to proteolysis. This characteristic may be a ramification of compact molecular configuration rendering enzymic active sites stable in the condition of proteolytic digestion. Figure 10 and 11 represent electrophoretic results of the digested enzyme with and without treatments of ME, urea and ME plus urea. Gel patterns of X0 which were digested individually by six different proteases were similar to that of an undigested enzyme sample, except that the protein band in some gels migrated slightly faster as apparent in Figure 10A. Mobilities of the band of trypsin-, chymotrypsin- and pancreatin-treated samples, designated gel 2, 3, and 5, repectively. were slightly higher than that of the control X0 specimen. Haud et al. (1975) and Nalger and Vartanyan (1976) obtained similar results, noting that treatment of X0 with trypsin, chymotrypsin and pancreatin resulted in an increase in the electrophoretic mobilities of the protein bands in disc-PAGE. Figure 108, 11A and 118 show the gel patterns of proteolyzed X0 further treated with 1% ME, 6 M urea and 1% ME plus 6 M urea, respectively. 88 Figure 10. Disc-PAGE electropherograms of purified X0 subjected to proteolytic digestion by the following enzymes: l-control (no enzyme), 2-trypsin, 3-chymotrypsin, 4-plasmin, S-pancreatin, 6-pepsin, 7-papain. Group A represents data from 4 hr digestion. Group 8 illustrates effect of 1% ME added subsequent to digestion. iflu 1M... 89 ‘1! , i” ;.n .1 1‘ u ’ Illivl . i .4.— : .... . ‘il‘ . _ _ 4 ' l1! 1‘ {I 111111. , 4 i '1 Cc . . T Figure 10 90 Figure 11. Gel electropherograms of proteolyzed X0 treated with 6M urea (A) and 1% ME plus 6 M urea (B) in disc-PAGE (5%T). Legend for gel patterns: 1-no protease, 2-trypsin, 3-chymotrypsin, 4-plasmin, 5-pancreatin, 6-pepsin, 7-papain. 91 ii I! Figure l l 1.l — 44 92 The numbers of bands representing the digested enzyme samples were similar to those of undigested X0 with the same treatment, whereas their electr0phoretic mobilities were different. All the bands visulized by Coomassie blue staining also revealed positive staining for enzyme activity. These studies indicate that, under the above conditions of treatment, proteolyzed X0 could not be electrophoretically resolved. Thus, it is apparent that the enzyme is very resistant to digestion by proteases at pH 6.8 and that integrity of its activity and molecular structure are conserved. The data of Nalger and Vartanyan (1976) represent support for this conclusion. They found that proteolyzed polypeptides of X0 were resistant to further digestion by proteases at pH 8. They, too, failed to separate the proteolytic fragments under conditions at which the enzyme remained enzymically active. They found that separation of these fragments was possible only in an electrophoretic environment containing 1% $08 or 5 M guanidine-HCl. Under these conditions, enzymic activity was lost. Nalger and Vartanyan (1976) also studied the limited proteolysis of X0 under partially denaturing conditions such as high pH. They reported that the oxidase activity of the enzyme remained unchanged before and after the addition of subtilisin A at pH 11.2. SOS-gel patterns of the enzyme after partial denaturation and proteolysis indicated that a selective digestion occurred. They suggested that the protein molecule of X0 may be a bi- or poly-globular type, since a protein made up of only one type of globule would be unfolded under the conditions employed resulting in a non-selective and/or complete digestion of the polypeptide chain. However, the FAD molecule of the 93 enzyme is dissociated from the enzyme at high pH which may result in a partial alteration of the protein structure. Subtilisin A is a serine protease similar to chymotrypsin which has a limited substrate specificity. Investigation of the hypothesis of poly-globular structure of the enzyme is best performed with the intact protein and a protease capable of performing random cleavage. If the X0 molecule is made up of only one type of globule, proteolysis should consist of a non-selective digestion of polypeptide chains. Papain was chosen here for this purpose since papain has broader substrate specificity than serine proteases. The results obtained with papain digestion showed that the oxidase activity of purified X0 did not change (see Figure 9) and its protein molecule remained compact and could not be further dissociated by ME and urea (see gel 7 in Figure 11A and B). The SDS-gel pattern of papain digested-X0 (Gel I in Figure 12) indicated that a restricted digestion occurred. Although papain is not an ideal protease to use for this investigation, the results obtained, when considered with the results obtained from the limited proteolysis of X0 by the other five proteolytic enzymes and the work of Nalger and Vartanyan (1976), provide sufficient evidence to conclude that the molecular structure of X0 is a polyglobular type. SOS-gel patterns shown in Figure 12 represent the purified X0 specimen digested by six proteases at 37 C for 4 hr. Molecular weights of the polypeptide bands revealed in all gels were estimated by referring their relative mobilities to the standard curve shown in Figure 2. Gel C represents the undigested X0, revealing a single band with 94 Figure 12. SOS-gel electropherograms of purified X0 digested by various proteases. Legend: Aplow MN protein standard, B-high MW protein standard, C-undigested X0, and the enzyme treated with D-trypsin, E-chymotrypsin, F—plasmin, G-pancreatin, H-pepsin, and I-papain. SDS- PAGE (9%T) is according to Laemmli's method (1970). 95 .A. — i :1 _——.v.1 - AL... n i 'Iv‘l' w—w a. 2.6a 2.1.: .n—_ .4“ .c.- hum— 96 molecular weight of 149,000. As discussed previously, this polypeptide is the monomer of X0. Gel 0 shows the dissociation of trypsinized X0 after a 4 hr digestion. Four major bands were observed with molecular weights of 136,000, 85,000, 35,000 and 18,000 (from top to bottom of the gel, respectively). Degradation of X0 by trypsin digestion was also examined by SOS-PAGE. Sample were removed from the digestion mixture after 10 min and at subsequent intervals of 30 min for 4 hr. Only the 136,000 dalton species and several very faint bands migrating close to the marker dye were observed after 30 min of digestion. Thirty minutes later, the SOS-gel pattern revealed four major bands of molecular weight 136,000, 95,000, 33,000 and 20,000. Afetr 90 min an additional 85,000 dalton band appeared in the gel pattern accompanied by a minor band of 95,000 daltons. This minor band disappeared after two hours digestion and the remaining four species were resistant to further breakdown. The results indicate that trypsin rapidly converted the subunit (150,000) of X0 into a 136,000 dalton species with a subsequence breakdown into fragments. The 85,000 dalton species was derived from either the 95,000 or the 136,000 dalton species. Haud et al. (1975) obtained one major band of approximately 130,000 molecular weight after treating X0 with trypsin for 30 min. They suggested that trypsinization caused a single cleavage in the subunit at one of the termini resulting in 130,000 and 20,000 molecular weight fragments. Mangino (1976) also observed a limited breakdown of the enzyme by digestion with trypsin. Two major bands with molecular weights of 153,000 and 90,000 were found after 50 min of digestion. Only the 90,000 dalton band and several trace bands were seen in SOS-gel after the digestion proceeded 24 hr. He found the only 13% of the specific 97 activity of the original enzyme was present in the 90,000 dalton species. Nalger and Vartanyan (1976) observed that the subunit (150,000) of X0 was rapidly converted into a 135,000 dalton species during trypsin digestion. Three major bands in SOS-gel, possessing molecular weights of 135,000, 92,000 and 20,000, were observed after trypsinization of X0 for 24 hr at pH 8. They suggested that the 92,000 band was formed from the 150,000 and 135,000 dalton species by tryptic activity. They also found that a high concentration of trypsin split the 92,000 dalton polypeptide into a 84,000 dalton species. Gel E shows the results of incubating purified X0 with chymotrypsin at 37 C for 4 hr. Three major XO-derived bands with molecular weights of 80,000, 29,000 and 18,000 and, just above the 18,000 species, a band representing chymotrypsin. The location of the chymotrypsin band was identified by running SOS-PAGE with a sample containing chymotrypsin only carried through the incubation. Two minor bands of molecular weight of 63,000 and 33,000 were also seen in the gel pattern. After the first 30 min of digestion, a 110,000 band was observed in addition to above bands. The staining of band 33,000 was more intense than that of band 29,000. After 1.5 hr the disappearance of band 110,000 and a reduction in color density of band 33,000 accompanied by an enhanced color density for band 29,000 were recorded. The gel pattern remained unchanged after further digestion. Nalger and Vartanyan (1976) reported that treatment of X0 with chymotrypsin and subtilisin for 24 hr gave rise to three major fragments with molecular weights of 92,000, 42,000 and 20,000. They found that high concentration of the proteases could further digest the 92,000 and 42,000 dalton species to 84,000 and 35,000 dalton 98 species, respectively. Gel F represents the result plasmin treated X0 after 4 hr at 37 C. Molecular weights of the observed three major zones were 136,000, 115,000 and 18,000. Degradation of X0 by plasmin was slower than that by the other five proteases. After the first 2 hr, the SOS-gel pattern revealed a single band of molecular weight 146,000. This species was converted to a 136,000 dalton species after 2.5 hr. After 3 hr of digestion an additional minor zone of 115,000 molecular weight and a trace band of 18,000 daltons were observed. The latter two bands became slightly darker when digestion was terminated. Since plasmin is a endogenous protease in milk and its concentration in milk is much less than that used in this experiment, the results allow the suggestion that milk endogenous plasmin would not influence the purification of X0 if started with fresh milk. Degradation of X0 by pancreatin was observed in the first 10 min. The subunit (150,000) completely disappeared and was replaced by a 134,000 dalton band. This molecular species was further degraded into three major fragments with molecular weights of 85,000, 33,000 and 18,000 after 30 min of digestion. These three species did not show further breakdown, constituting the SOS-gel patterns to the conclusion of digestion (see Gel G in Figure 12). Haud et al. (1975) obtained similar results when X0 purified by a non-proteolytic method was degraded into 89,000 and 38,000 dalton polypeptides by digestion with pancreatin for 30 min. An additional three bands of 80,000, 75,000 and 19,000 molecular weight were observed after 6 hr digestion. Data obtained here indicate that pancreatin affects the molecular properties of X0 and its dehydrogenase characteristics. Addition of 99 pancreatin in the isolation procedure would promote the degradation of X0 molecule although its catalytic properties would be retained. This conclusion was supported by Haud et al. (1975), Nalger and Vartanyan (1976) and Battelli et al. (1973) who showed that X0 purified in the presence of pancreatin was dissociated into 92,000, 35,000 and 15,000 dalton species when assayed by SOS-PAGE. Pepsin and papain effected a fast digestion of X0. The degraded products after 10 min were resistant to further cleavage. X0 was degraded into 136,000 dalton species by pepsin, whereas fragments of 130,000, 105,000, 35,000, 29,000 and 18,000 molecular weight were obtained after 10 min of digestion with papain. The gel patterns remained unchanged for up to 4 hr of digestion (see Gel H and I in Figure 12). Since the optimum pH for pepsin activity is 2, the digestion pH of 6.8 used here-close to the tail of its pH-activity profile, was too far removed from its optimum to achieve a full catalytic reaction. Whether or not X0 is resistant to proteolysis by fully active pepsin remained unanswered. However, it may be suggested that X0 in milk does not degrade extensively when pepsin is added in fresh milk since the pH of milk is about 6.7. The importance of degradation of X0 by papain is not merely the characteristics of the degradation products but also the resistance these products exhibit to further digestion. From the above data, it was apparent that three polypeptides of molecular weights ranging 85,000-100,000, 30,000-35,000 and 18,000- 20,000 were conmonly found after cleavage of X0 by proteolysis. Nalger and Vartanyan (1976) obtained the same results noting that X0 was converted to three fragments with molecular weight of 92,000, 42,000 100 and 20,000 by proteolytic enzymes and that the 92,000 and 42,000 dalton fragments could be further converted to 84,000 and 35,000 dalton species, respectively. The results obtained in the present study were essentially similar and support their postulation of a subunit molecule (150,000) of X0 composed of three globular fragments. Furthermore, the results further suggest that a strong hydrophobic binding force is required to hold the three globules in a tight configuration. Hence, the intact molecule could not be separated under the dissociating conditions afforded by ME and urea, retaining its catalytic properties after proteolysis. Amino Acids Contents of the Fragments of Trypsinized X0 All major fragments of trypsinized X0, e.g., 136,000, 85,000, 35,000, and 18,000 contain lysine as their N-terminal amino acid. Figure 13A shows typical results of four dimensional chromatography of dansylated N-terminal amino acid of the 85,000 fragment on micropolyamide sheet. Two spots were observed in the sample plate which were absent in the chromatogram developed from the blank gel after third dimension development (Column 8 in Figure 13). After running the dansylated lysine standard on the reversed side of the sample sheet, the two spots coincided with the spots of lysine standard in all 4 dimension runs. The chromatogram of a lysine standard is shown in Figure 7A. Chromatograms of dansylated N-amino terminal of the remaining three fragments were similar to that shown in Figure 13A. It is unusual that all fragments from trypsinized X0 and the intact enzyme showed lysine to be the N-terminal amino acid. 101 Figure 13. Chromatograms of dansylated N-amino terminal residues of trypsin-derived fragments of X0. Column A indicates results from four dimension chromatography of acid hydrolysate of the dansylated 85,000-dalton species. Chromatograms of dansylated 136,000-, 35,000-, 18,000- dalton species were similar to column A. Column 8 indicates chromatograms of a blank gel. Legend: G- glysine, NH -dansyl NH , OH-dansyl 0H, L-lysine. Salvent systgm was described in EXPERIMENTAL. 102 o I O O i? " o ”3 g 0'- 06 3 0 «<3 01. 3 OH O 0 141430 06 0L 0 m oL of E 6 CD , .2 0 OH o 3 O .L I O V. 0L '2 :4 O on ‘5 O . 2 o W Solvenil -—-> A Figure 13 OH 103 However, the lysine content of X0 is, in fact, relatively high compared to other residues, e.g., 172 lysine residues per mole X0 calculated from 6.8 mole percent (see Table 6). Trypsin cleaves the peptide bond between the COOH group of lysine or arginine and the amino group of the adjacent amino acid except when proline is the adjacent amino acid. It is suggested that the first N-terminal peptide bond of the fragments as well as the enzyme is a lys-pro bond. Only a 60% recovery was obtained from the electrophoretic elution of trypsin-derived fragnents based on a comparison of the amount of X0 subjected to SOS-PAGE and the total amount of polypeptides eluted from the agarose gel system. The amount of each eluted fragment was not sufficient to determine the content of sulfur-amino acids and tryptophan. Table 8 represents the amino acid composition without methionine, half-cystine and tryptophan of the major polypeptide fragments of trypsinized X0. All fragments have low histidine, tyrosine and arginine contents and high aspartic acid and glutamic acid. A comparison of the results of each polypeptide revealed that the 136,000 dalton species contained relative high lysine, glutamic acid and leucine, whereas high contents of glycine and arginine were found in the 85,000 polypetide. Relative high contents of serine, lysine and glycine were found in the 35,000 dalton fragment, whereas the 18,000 dalton polypeptide contained a relative high leucine content. Although performic acid oxidation of fragment species was not performed, the amino acid chromatogram of the HCl-hydrolysate of the 18,000 dalton sample showed a peak at the half-cystine position. The chromatograms of the other fragments showed only a small or no peak at 104 Table 8. Amino acid compositions of the major fragments of trypsinized X0 Amino Acid Content (mole%) 136,000 85,000 35,000 18,000 Residue dalton dalton dalton dalton species species species species Lysine 8.7 5.3 6.8 3.9 Histidine 1.6 2.3 1.9 2.2 Arginine 1.8 3.5 2.6 2.9 Aspartic acid 12.8 10.0 8.4 12.1 Threonine 5.8 5.6 6.4 6.5 Serine 4.6 6.2 11.5 7.7 Glutamic acid 16.7 10.7 12.2 12.2 Proline 5.9 5.9 6.0 7.5 Glycine 5.0 15.8 12.8 5.1 Alanine 6.9 8.0 7.7 8.5 Half cystine --- --- --- --- Valine 7.2 7.1 6.0 8.0 Methionine --- --- --- --- Isoleucine 5.6 5.1 3.8 5.9 Leucine 12.4 7.7 7.9 10.2 Tyrosine 1.8 2.6 1.9 2.4 Phenylalanine 3.3 4.3 4.2 5.0 Tryptophan --- --- --- --- Average hydrophobicity 1272 1116 1062 1272 105 this position, indicating that the 18,000 dalton polypeptide contained significantly more free SH groups than the other three polypeptides. Coughlan (1980) suggested that a peptide with molecular weight of 20,000 containing about ten free sulfhydryl groups which possibly function to stabilize the dehydrogenase conformation of X0. This peptide could be removed by proteolysis. Since X0 lost its dehydrogenase activity after proteolysis by six proteases and the 18,000 dalton band was observed in all SOS-gel of proteolyzed X0 (except pepsin treatment), it is suggested that the 18,000 dalton species is the peptide described by Coughlan (1980). Average hydrophobicities calculated according to Bigelow (1967) were 1,272, 1,116, 1,062 and 1,272 cal for 136,000-, 85,000-, 35,000- and 18,000-dalton fragments, respectively. Average hydrophobicities of the 136,000 and 18,000 polypeptides are higher than the 1,135 cal calculated for intact X0, whereas the 85,000 and 35,000 polypeptides revealed slightly lower average hydr0phobicities than that of X0. These results led to the conclusion that the fragments of trypsinized X0 possess hydrophobic characteristics which account for their association in dissociating environments. Photometric scanning of stained SDS-gels of trypsin-digested X0 revealed peak area percentages of 6.89, 57.54, 25.59 and 13.22 for bands corresponding to the 136,000, 85,000, 35,000 and 18,000 dalton species, respectively. Assuming equal dye binding capacities for each of the fragments, their relative peak area distributions represent the relative weight concentrations. Therefore, the sum of the amino acid contents for each fragment should represent the amino acid composition of X0. The results of these calculations are given in Table 9. Column 106 Table 9. Comparison of amino acid composition of the purified X0 and its amino acid composition calculated from its trypsin-derived fragments Amino Acid Cotent_(mole%) Residue A B C D Lysine 7.3 5.5 5.3 5.6 Histidine 2.5 1.9 2.0 2.0 Arginine 4.6 3.0 3.0 2.5 Aspartic acid 9.1 9.8 9.3 12.3 Threonine 7.2 5.7 5.6 6.3 Serine 6.1 7.3 7.3 6.6 Glutamic acid 10.2 11.3 10.6 13.7 Proline 6.0 5.9 5.8 6.9 Glycine 8.6 15.8 12.7 5.1 Alanine 7.9 7.7 7.5 7.9 Valine 7.5 6.8 6.5 7.7 Isoleucine 5.4 4.8 4.5 5.8 Leucine 9.8 8.2 7.6 10.9 Tyrosine 2.3 2.3 2.3 2.2 Phenylalanine 5.4 4.1 4.1 4.4 A; Amino acid composition of the purified X0 recalculated the data from Table 6 without half cystine, methionine and tryptophan. 8; Calculated from data in Table 8. Each amino acid content is the sum of 6.89%, 57.54%, 25.59% and 13.22% of its content of 136,000, 85,000, 35,000 and 18,000 dalton species, respectively. C; Same calculation as B with 59.75%, 26.52% and 13.73% for 85,000, 35,000 and 18,000 dalton species, respectively. 0; Same calculation as B with 34.26% and 65.74% for 136,000 and 18,000 dalton species, respectively. 107 8 represents an amino acid composition calculated from all four major fragments. Data in column C was obtained by assuming that X0 is constituted from 85,000, 35,000 and 18,000 dalton species in weight ratio of 59.75%, 26.52% and 13.73%, respectively, recalculated from the peak-area percentage of the scanned gel. Data in column 0 were based on weight ratios of 34.24% and 65.74% for the 136,000 and 18,000 species, respectively. Here, it was assumed that these two polypeptides constitute the XOInolecule. When compared to the data for the original X0 (Column A), all three calculated results show differences in distribution of amino acids. The discrepancy may be due to the ignored small peptides produced by trypsin digestion. In fact, trace bands, of 85,000 and 35,000 dalton, derived from the 95,000 and 42,000 dalton bands, respectively, were observed in SDS-PAGE gels. Similar species were reported by Nalger and Vartanyan (1976). Quaternary Structure of Milk X0 Based on the above results from analyses of trypsinized X0, the model of polyglobular structure of milk X0 proposed by Nalger and Vartanyan (1976) can be further enhanced in Figure 14. Each subunit (150,000) of X0 is composed of three globulars segments, A, B and C, with molecular weights of 92,000 (85,000), 42,000 (35,000) and 20,000 (18,000), respectively. The unparenthesized numbers were suggested by Nalger and Vartanyan (1976) while numbers in parentheses were obtained in this study. A peptide bond between the amino group of lysine and the COOH group of other amino acid, possibly lysine or arginine (due to trypsin substrate specificity), linked each globule. In addition to 108 l l LYS-(PRO) (Paol-Lvs (P800 (PROO -_:YS 3' Ya LY:- A A LYS LYS Figure 14. Schematic representation of a suggested modification of the polyglobular structure of native milk X0 as proposed by Nalger and Vartanyan (1976). Molecular weights of each globule are: A-92,000 (85,000), B-42,000 (35,000), C-20,000 (18,000). Numbers in parentheses indicate molecular weight values obtained in this study. Legend: PRO=proline, LYS=lysine, +H--hydr0phobic binding, and arrows indicating the peptide bonds cleaved by trypsin. 109 the peptide bond between each globule, hydrophobic binding is probably involved in stabilizing the three globules, preventing their dissociation in urea and/or ME environments after proteolytic cleavage of peptide bonds. Nalger and Vartanyan (1976) suggested that globule C was the first released followed by the release of globule 8 during proteolysis. Attempts to prove this hypothesis failed since lysine was found to be N-terminal amino acid for all three globules and the intact enzyme in this study. It is not clear whether the N-amino terminal exists in globule A or C because lysine is N-terminal in the intact X0. Further research is required to provide evidence in support of this model. Free Riboflavin, FMN and FAD Contents and X0 Activity in Cow's Milk Swope et al. (1965) found a fluorescent material in aqueous suspensions of fat globule membrane preparations which contained riboflavin and its natural derivatives such as FMN and FAD. About 92- 96% of the total riboflavin in the fluorescent substance is FAD. They suggested that FAD is contributed by dissociation from X0 in the preparation of MFGM. Milk contains a system of enzymes including alkaline phosphatase capable of converting FAD to FMN and riboflavin (Manson and Modi, 1957). The FAD moiety released from X0, if it occurrs, can be converted to FMN and riboflavin, resulting in an increase in free riboflavin in milk. Table 10 shows the contents of FAD and its components and X0 activity in milk samples stored at room temperature for 0, 6, 20 and 110 Table 10. Flavins content and X0 activity in milk stored at room temperature X0 Free Total Hour activity FAD FMN riboflavin riboflavin (IU/ml) (ug/ml) (ug/ml) (us/ml) (ug/ml) 0 0.037 0.178 0.252 1.018 1.448 6 0.046 0.161 0.211 0.932 1.304 20 0.080 0.244 0.141 0.920 1.305 48 0.125 0.260 0.336 0.879 1.475 Table 11. Flavins content and X0 activity in milk stored at 4 C X0 Free Total Hour activity FAD FMN riboflavin riboflavin (IU/ml) (Hg/ml) (U9/ml) (Hg/ml) (Hg/ml) 0 0.037 0.178 0.252 1.018 1.448 6 0.152 0.120 0.217 1.041 1.378 20 0.152 0.164 0.309 0.9 1.417 48 0.132 0.250 0.378 0.812 1.440 111 Table 12. Flavins content and X0 activity in milk stored at room temperature in the presence of aprotinin (protease inhibitor) X0 Free Total Hour activity FAD FMN riboflavin riboflavin (IU/ml) (U9/ml) (ug/ml) (ug/ml) (Hg/ml) 0 0.010 0.081 0.197 0.456 0.733 6 0.013 0.062 0.159 0.462 0.683 20 0.032 0.079 0.176 0.470 0.722 48 0.065 0.108 0.128 0.585 0.820 Table 13. Flavins content and X0 activity in milk stored at 4 C in the presence of aprotinin. X0 Free Total Hour activity FAD FMN riboflavin riboflavin (IU/ml) (Hg/ml) (us/ml) (us/ml) (U9/ml) 0 0.010 0.081 0.197 0.456 0.733 6 0.158 0.134 0.105 0.431 0.670 20 0.162 0.084 0.155 0.471 0.710 48 0.151 0.091 0.143 0.474 0.708 112 Table 14. Flavins content and X0 activity in milk dialyzed at 4 C for 24hr and followed by stored at 4 C X0 Free Total Treatment activity FAD FMN riboflavin riboflavin (IU/ml) (us/ml) (ug/ml) (U9/ml) (us/ml) After dialyzed 24 hr at 4 C 0.107 0.100 0.034 0.082 0.216 1 day storage 0.100 0.101 0.034 0.083 0.218 2 days storage 0.085 0.101 0.037 0.083 0.221 Table 15. Flavins content and X0 activity in milk stored at 70 C X0 Free Total Hour activity FAD FMN riboflavin riboflavin (IU/ml) (ug/ml) (ug/ml) (ug/ml) (uglml) 0 0.037 0.178 0.252 1.018 _ 1.448 6 0.070 0.110 0.173 1.141 1.424 20 0.022 0.150 0.338 0.896 1.384 48 0.002 0.141 0.401 0.869 1.411 113 NADH- K3Fe(CN)5 Activity, AA420 /min Figure 15. NADH-ferricyanide reductase activity of X0 in milk stored for 0, 6, 20 and 48 hr at different temperatures. 114 48 hr. X0 activity increased gradually and reached about 3-fold of its original activity after 48 hr. Amounts of FAD and FMN decreased at first (6 hr), then increased after 48 hr, whereas the amount of free riboflavin decreased during the storage period. Assuming that free riboflavin is not further degraded into non-fluorescent products by environmental effects, the results indicate that increased X0 activity is accompanied by an increase in FAD content and a decrease in free riboflavin. It is suggested that the conversion of FAD to riboflavin by the enzymes in milk acted in a reverse direction to form FAD from free riboflavin. The FAD product, could have been picked up by inactive X0, which lacks FAD, resulting in its activation. However, this hypothesis was not supported by the data obtained from milk stored at 4 C and by other evidence to be discussed. Table 11 presents the data characterizing the flavin contents and X0 activities in milk stored for 0, 6, 20 and 48 hr at 4 C. X0 activity increased rapidly and reached the highest level after 6 hr with a slight decrease after 48 hr. Changes in contents of FAD, FMN and free riboflavin were random during the storage of milk. After 48 hr the contents of FAD and FMN were the highest and free riboflavin was the lowest when compared to different storage times. Changes in the flavin content were similar to that observed in the milk sample stored at room temperature. It is suggested that the interconversion of flavins in milk also occurred at low temperature. However, the increase of X0 activity at 4 C did not correspond to the change in flavin content. Since protease can degrade the enzyme without affecting enzymic activity, the experiments described above were repeated in the 115 presence of protease inhibitor(s). Chloroquine and aprotinin have been used as protease inhibitors to eliminate the protease effect on X0 in MFGM preparations (Eigel, 1980). Only aprotinin was used here since chloroquine was found to be a mixed-type inhibitor of X0 (see previous discussion in section on Enzyme Nature). Table 12 and 13 report results obtained with milk stored at room temperature and 4 C in the presence of aprotinin. Increased X0 activity similar to that of milk in the absence of inhibiotr was observed, whereas the contents of free riboflavin and its natural derivatives showed a random change. FAD and FMN contents did not increase to the extent observed in milk without the aprotinin. The results suggest that the increase in X0 activity was not related to the changes in FAD and its components. Milk samples were dialyzed for 24 hr to diminsh free riboflavin in milk and subsequently stored for an additional 1 and 2 days. It was found that X0 activity decreased with a very small increase of FAD, FMN and free riboflavin (see Table 14). The small change noted in the flavin content was attributed to the experimental error in fluorescence measurements. Therefore, the changes of FAD, FMN and free riboflavin are independent of the increase in X0 activity. Table 15 represents results for milk samples stored at 70 C for 0, 6, 20 and 48 hr. Decreased FAD and FMN contents accompanied by an increase in free riboflavin at 6 hr were due to the enzymes system promoting the conversion of FAD to FMN and free riboflavin in heated milk. The enzyme system was destroyed in milk subjected to prolonged heating at 70 C. Therefore, it is postulated that the FAD molecule released from X0, if, indeed, it occurs, increases the FAD content only and does not contribute to the increase in FMN and free 116 riboflavin. Based on this theory, the increased FAD content and decreased X0 activity in milk stored for 20 hr suggests that FAD is released from X0. However, the theory does not explain why the FMN and free riboflavin contents decreased. Also, the large decrease in X0 activity and decrease of FAD content in milk after heating for 48 hr could not be explained. Thus, the release of FAD from X0 during storage of milk at 70 C is inconclusive. The XO activity obtained by measuring the xanthine-uric acid reaction may not reflect the presence or absence of its FAD moiety since the active enzyme retains its reducibility by xanthine after removal of the flavin moiety. Therefore, to determine if FAD remains as a part of the active enzyme, its activity was measured, using NADH as a reducing substrate and potassium ferricyanide as an electron acceptor (Komai et al., 1969). An increase in the oxidation of ferricyanide was observed and the increasing rate of NADH-ferricyanide reductase activity of X0 was similar to that calculated from xanthine- uric acid measurements (see Figure 15). Thus, it appears that the prosthetic FAD group was not dissociated from the enzyme when employing xanthine-uric acid activity measurements. From the above results it was concluded that X0 is incapable of releasing its FAD molecule as a source of free riboflavin in normal cow's milk. Although it is possible that FAD is released from the enzyme as a consequence of heating milk, evidence supporting this proposition is inconclusive. Ho et al. (1978) studied the availability of riboflavin from bovine milk X0. They found that the rate of weight gain for chicks fed with 19.4 ug of riboflavin/day in the form of X0 was equivalent to that of chicks consuming 5 ug riboflavin/day. Based 117 on this observation, they concluded that the availablity of riboflavin in X0 for growth of chicks was approximately 25% of the potential riboflavin and suggested that only 4% of milk riboflavin is X0- associated available riboflavin. They concluded that X0 is not a good dietary source of riboflavin. Stabilization of Fat Globules by X0 Figure 16 illustrates the relationship between surface tension and concentration of X0, casein and whey proteins. Increasing protein concentrations up to 0.02 mg/ml resulted in a rapid decrease in surface tension of the solution whereas further increase in concentration produced only slight change in surface tension. These results indicate that the correlation between surface tension and protein concentration is typical of surface-active systems. The data reported herein show that X0 is more surface active than casein and whey proteins. All three protein samples demonstrated interfacial activity at a butter oil-simulated milk ultrafiltrate interface (see Figure 16). Rapid reduction of interfacial tension at low protein concentrations occurred, whereas the interfacial tension did not change at high concentration of proteins. The change of interfacial tension by casein was similar to results reported by Jackson and Pallansch (1961). The reduction of interfacial tension by whey proteins was less than that by casein. Major whey protein components may be less effective at reducing the free energy at the interface than casein. This explanation is supported by the work of Jackson and Pallansch (1961) 118 Figure 16. Surface tension of whey proteins, casein and purified X0 at room temperature. Figure 17. Interfacial tension of whey proteins, casein and purified X0 at butter oil/simulated milk ultrafiltrate interface at 45 C. 119 ./' // ‘ i // a K} Surface Tension, dyne/cm ~ .g T T '\ 'Whev - fi:\ .’ 50 \ oégein We 0 .04 .08 .12 .10 .32 Protein Concentration, mg/ml Figure 16 0 .04 .00 .12 .10 .32 Protein Concentration, mg/ml Interfacial Tension, dynelcm N Figure 17 120 Figure 18. Measurement of the size distribution stability of fat droplets in protein-butter oil emulsions dispersed in simulated milk ultrafiltrate at 45 C. Proteins employed were: casein, whey proteins, purified X0 and a mixture of each in a ratio of 80:19.9:0.1, repectively. SMU indicates the emulsion formed with butter oil and simulated milk ultrafiltrate only. . A420 Turbidity , 121 - casein mixture 122 who found that bovine serum albumin, alpha-lactalbumin, and beta- lactoglobulin showed poor interfacial activities in a butter oil- protein-free plasma system. The experimental data indicate that X0 is the most interfacially active of the three proteins studied and that the amount of X0 adsorbed on the butter oil Surface exceeded that of casein and whey proteins. The surface active properties of X0 as well as its hydrophobic nature indicate that X0 is more lipophilic than most of the other proteins in milk. This property of X0 may explain why X0 is concentrated in cream phase. Also, it has been postulated that X0 may involve the stabilization of milk fat globules (Mangino and Brunner, 1977a). Halstra (1965) introduced a turbidimetric technique to determine the size distribution of fat globule in milk. He suggested that the change in turbidity could be used to monitor small changes in fat globule size and the stability of fat emulsion. Based on his theory, turbidimetry was adopted to examine XO-stabilized fat globules. Figure 18 represents the results of studies with emulsions formed by homogenizing X0, casein and whey proteins with butter oil at 45 C. A decrease in absorbance of less than 0.1 units at 420 nm was noted for the XO-stabilized emulsion after 11 hr of storage at 45 C. The casein-emulsion showed a slight decrease in A after the first hour following homogenization which was followed by a drastic decrease after further aging. The whey protein-emulsion showed a large decrease of absorbance after 11 hr. These results indicate that the size of fat globules in the XO-emulsion was more stable than that in casein- and whey protein-emulsion. Although only a slight decrease in the 123 absorbance of the casein emulsion occurred during first hour, subsequent decreases in absorbance reflect the coalescence of small fat globules to form large globules. In fact, an oil layer was observed on the surface of casein- and whey protein-emulsion after 6 hr, whereas no oil film was found on the X0 sample. Obviously, the stabilization of fat globules by X0 is more effective than that by casein and whey proteins. The above evidence not only supports the hypothesis of stabilzation of milk fat globules by X0 but also indicates that X0 could be one of the first proteins to associate with naked, intracellular fat globules. Thus, the following hypothesis is suggested. Nascent fat globules in secretory cells are coated with X0 (may be coated concomitantly with a phospholipid/cholesterol film, Hood and Patton, 1973) since X0 is distributed throughout the cytoplasm of mammary epithelial cells which line the alveolar lumina (Jarasch et al., 1981). This coating prevents the coalesence of fat globules as they migrate from the basal to the apical region of secretory cells. Also, this XO-coat may prevent lipid oxidation in fat globules. After the XO-coated fat globules reach the inner surface of the apical cell membrane, butyrophilin and a layer of plasma membrane are added to the globules during exocytosis. Therefore, milk fat globules are surrounded by an outer layer of MFGM and an inner coat of X0. Evidence of two layers, an outside true membrane layer and a firmly bound inner layer enriched in X0, surrounding milk fat globules has been reported by several researchers (Freudenstein et al., 1979; Franke et al., 1981; Buchheim, 1982). Recently, Phipps and Temple (1982) reported that the inner coat possesses overall high interfacial 124 activity in milk system. Also, Keenan et al. (1982) found that fatty acids in milk lipid, such as palmitic, stearic and oleic acids, could bind X0 in alkali-labile ester linkages. They did not indicated whether active X0 was similarly involved in lipid binding since the enzyme was isolated from SDS-gels of MFGM and was enzymically inactive. However, their results suggested that x0 possessed a strong affinity to lipid which supports the above hypothesis of X0 layered on the naked fat globules. Their results and observations from this study suggest that the X0 coating on nascent fat globules involves hydrophobic binding and possible alkali-labile linkages. Since X0 functions as a redox reaction, the prevention of lipid oxidation in XO-coated fat globules is disputable. However, Bruder et al. (1982) found that no significant increase in lipid peroxidation was detected by adding hypoxanthine to MFGM preparations, supernatant fractions from butter milk and tissues fractions containing native active X0 including a specimen from the mammary gland. Malondialdehyde production was not measurable with NADH or NADPH as substrate unless EDTA, ADP and ferric ions were added to the lipid peroxidation assay. In the latter case promotion of lipid peroxidation was inhibited by cyanide or peroxide dismutase. They concluded that lipid peroxidation was not a biological function of X0 which is in contrast to the hypoxanthine-X0 system in_vlt£g, The cell must provide an additional mechanism for preventing the tendency of X0 to promote lipid oxidation to ensure the secretion of undamaged lipid components. X0 may stabilize fat globules in the milk system since milk lipolysis concides with a release of membrane-bound X0 into skim milk (Bhavadasan et al., 1982). 125 The release of X0 from the fat globule complex during homogenization of milk and the stability of homogenized fat globules in homogenized milk seem to conflict with the proposed working hypothesis. It is known that the layers surrounding fat globules are disrupted and replaced partially with casein and whey proteins during homogenization. Hhey proteins and casein were found to be less effective in stabilizing butter oil emulsions than was X0, see Figure 18. In fact, the stability of the fat globule size distribution in an emulsion formed by homogenizing milk fat with a protein mixture of casein, whey proteins and X0 (with ratio of 80:19.9:0.1) was superior to that for either casein- or whey protein-fat emulsion(see Figure 18, curve of mixture). Decrease in the absorbance reading for the mixture-emulsion sample was only 0.12 after standing for 11 hr at 45 C which was close to that observed for the XO-emulsion. Thus, although homogenization result in a change of components surrounding milk fat- globules, the newly adsorbed casein and whey proteins and the remained coat material including X0 serve to stabilize the fat dispersion. Recently, Keenan et al. (1983) reported that most of the original MFGM materials, especially the inner coat which contains abundant X0 and butyrophilin, were retained on the fat globule surfaces after homogenization. This evidence not only indicates the strong affinity between inner coat and fat globules but supports the hypothesis that X0 plays a prominent role in the stabilization of the milk fat emulsion in normal milk and, possibly, homogenized milk. 1. 2. 3. 5. 6. 10. 11. 12. 13. CONCLUSIONS The spectral results, specific activity and electrophoretic properties of the final enzyme preparation indicated that the purified X0 was close to 100% pure. A reversible oxidase form, possessing the detectable dehydrogenase activity of X0, was isolated without exposure to proteolytic enzymes and reducing agents. The dehydrogenase activity in the enzyme was enhanced by 1% ME or 10 mM DTT but was completely lost after treatment with 6M urea. The enzyme performed as an active monomer in 6M urea condition. Chloroquine is a mixed-type inhibitor of X0. The enzyme preparation contained 14.8% protein nitrogen, no lipid and 82 sulfhydryl groups/mole with 44 of these constituting disulfide bonds. Proteases did not change the oxidase activity of X0. Lysine was determined as the N-terminal amino acid of the intact X0 and its trypsin-derived fragments. Amino acid compositions of X0 and its trypsin-derived fragments indicated their hydrophobic character. A previous hypothesis that the X0 monomer consists of three globular subunits held tightly by hydrophobic binding in addition to peptide bonds was supported. X0 is not a source of milk riboflavin. X0 is a surface and butter-oil interfacially active protein. Stabilization of fat globules by X0 is superior to that by 126 127 casein and whey protein. 14. A theory for the stabilization of fat globules by X0 during secretion and in milk was advanced. RECOMMENDATIONS Questions raised by this study requiring further investigation are: 1. What is the original form of milk X0? Is the enzyme synthesized in mammary gland? If so, is it synthesized as dehydrogenase similar to liver xanthine dehydrogenase? What is the mechanism involving the conversion of this enzyme to the reversible type 0 form of X0 found in this study? 2. What is butyrophilin? Is it possible that butyrophilin is a product of degraded X0/dehydrogenase? 3. What is the physiological evidence to support the theory of stabilization of fat globule by X0 in secretory cells? 4. What is the actual binding order of the three globular subunits of X0? Which globular subunit possesses the active site(s)- Mo, FAD, or both? 5. Does the 20,000 (18,000) dalton subunit contain more free sulfhydryl groups than the other two subunits? 00 these sulfhydryl groups contribute to the stabilization of dehydrogenase activity.of the enzyme as suggested by Coughlan (1980)? 6. If milk X0 originates from a dehydrogenase form, is the tertiary structure of this form the same as that of the liver enzyme described by Coughlan (1980), or a three globular type described in this study, or do both forms exist due to the interconvertability of the enzyme? 7. Why does X0 activity increase after milk is subjected to environmental stress? Is it possible that XOInolecules surrounding fat 128 129 globules orient as a polymer which results in a restriction of the enzyme activity? 8. Is it possible that X0 reaches the intestinal tract as individual subunits rather than as an intact molecule? If so, is the subunit containing the Mo site still active before and after intestinal absorption? Is it possible that absorbed subunits resemble somewhere in the body and regain complete enzymic characteristics? 9. Is reconstituted X0 and/or the active subunit containing the M0 site involved in atherogensis? 10. Is the reaction of XO-antiserum to its globular subunits the same? Answer may provide a tool for investigating question 8 and 9. APPENDIX 130 Table A1. Chemicals used in this study and their sources Chemical Company Ammonium persulfate Potassium ferricyanide Selenium dioxide Sodium phosphate, tribasic Sodium pyrophosphate Sodium deoxycholate Citrate-H01 buffer, pH 2.2, pH 3.28 Acrylamide Ammonium persulfate Bisacrylamide Glycine Hydroxylapatite Sodium dodecyl sulfate Urea Ethyl acetate Photo-flo 200 Riboflavin N,N,N -N -Tetramethyl- ethylenediamine Benzyl alcohol Boric acid Bromophenol blue Phosphoric acid Potassium citrate monohydrous Sodium citrate-5H20 Sodium phosphate, dibasic Sodium phosphate, monobasic Agarose Folin-Ciocalteu phenol reagent Chymotrypsin Papain Pepsin Tnypsin Xanthine Ammonium sulfate Bartital Chloroform Copper sulfate Formic acid Glacial acetic acid Hydrochloric acid Hydrogen bromic acid Hydrogen peroxide 2-Mercaptoethanol Methanol J. T. Baker Chemical Co. Baltimore Biological Laboratory, Inc. Beckman Instruments Bio-Rad Laboratories Burdick & Jackson Laboratories, Inc. Eastman Kodak Co. Fisher Scientific Co. FMC Corp. Harleco ICN Nutritional Biochemicals Mallinckrodt 131 Table A1. (continued) Chemical Company chloride phosphate, dibasic Potassium phosphate, monobasic Potassium sulfate Sodium bicarbonate Sodium chloride Sodium hydroxide Sodium nitrite Sodium phosphate, dibasic Sodium phosphate, monobasic Sucrose Sulfuric acid Trichloroacetic acid Ammonium sulfamate Bromocresol green Calcium carbonate Magnesium carbonate Methyl green Sodium salicylate Blue dextran Low MW protein calibrate kit High MW protein calibrate kit Sephacryl S-200 Ninhydrin Dansyl amino acids Dansyl chloride Aprotinin Bovine serum albumin Chloroquine Coomassie blue R-250 Dithionitrobenzoic acid Ethylenediamine Tetraacetic acid Fibrinolysin (Plasmin) N-Leucine N-1-(naphthyl)ethylenediamine dihydrochloride Neotetrazolium chloride Nicotinamide adenine dinucleotide Pancreatin Thyptophan Tris (hydroxymethyl)aminomethan ( Potassium Potassium Matheson, Coleman & Bell (MCB) Merck & Co. Inc. Pharmacia Fine Chemicals Pierce Chemical Co. Sigma Chemical Co. Sigma 7-9) 132 Table A2. Equipment routinely used in this study Equipment Company Analytical balance, type 2463 Satorius Balance Top-loading balance, type K7T Mattler Instrument Corp. Carmera, MP-3 Land Camera Polaroid Corp. Electrophoresis set ‘ Bio-Rad Laboratories Preparative refrigerated centrifuge, Model RC2-B, type SS-34 and GSA rotors Sorvall Instruments Preparative refrigerated ultracentrifuge, Model L-2-65, type 21, 30 and 65 rotors Beckman Instruments Dialyzing tubing Union Carbide Corp. Disk milk separator, type LWA 205 Westfalia Separator Hand-operated homogenizer C. W. Logeman Co. Lyophilizer Laboratory-constructed, Dr. J. R. Brunner Power supple M158 MRA Corp. Research pH meter, Model 12 Coring Scientific Instruments Rotary evaporator Buchler Instruments Thermo-lift water bath Buchler Instruments 133 Table A3. T0tal riboflavin content and X0 activity in fresh raw milk from a specific cow at different milking dates during 1981 Date Total riboflavin (ug/ml) X0 activity (IU/ml) 8 - 24 1.713 0.020 9 - 2 0.976 0.022 9 - 14 1.011 0.015 10 - 5 1.448 0.037 11 - 3 0.878 0.010 11 -. 9 1.411 0.016 11 - 17 1.127 0.011 134 b'iFe/S)” ~(Fe/S),, “‘(Fe/S)°, “'(Fe/Sio, r-FAD ~FAO H- r-Mom r-Mom 'H—0 ’H—0 \ N;/\N-H ~1’—,—/\N-H _ ,,/’ —ses——4LJQ{ "S-S N= f N—:. \N O “OH \N O ' A A DXIDIZED XANTHINE INTERMEDIATE COMPLEX ENZYME "r (Fe/S10, l L-(Fe/Sl,,... 02 -FAD ”1- ”Mo:II i URIC ACID L—S-S- REDUCED ENZYME Figure A1. Proposed mechanism for the catalytic role of the persulfide in X0 (Edmondson et al., 1972). 135 OKIOI‘ SE (Irreversible) I 1 ,,z . l /’ I I I ’ I / I I ,l” Proteolym —. : I, I a " ’ 1" t , , ’ ____________________ , i ‘. DEHYDROGENASE v (INTERMEDIATE!) ‘ ’ \ \ -—-—--—-—-o—-——n—O. t / 'NACTIVE ._""_‘— """ "‘ '—“"”“' °" " ”--“ ”" —- ""- OXIDASE (reversiglc) / / 7’ C") ° EDTA Figure A2. Interconversion of X0 among its various possible forms (Della Corte and Stirpe,1972). H H . \d’ c/ /N 1 N ”N/ I ;>"".IC>5“’| 535's“ 1);}:26- es- s— +QT/:/ :>»0n il\ / N/ ——-—o ~/J H N/ 0 "_— b’} 9) —_T h 2 1 mm 3 H n _-Mo—N ___.J -—,“°\—~ ——’M'0-—~——-—J _-Mo_‘/~\ / \ I I \ I \ K kin , k x + 50) ° 5x0) 2 5xn+21 3 60+21+0 Figure A3. Chemical interpretation of the intermediates of the reaction of X0 (Olson et al., 1974). 136 ' 0 H ' HzN/C i \ +. 4\Enz (3113 gaziwzv Figure A4. Mode of productive binding to X0 od I-methylnicotinamide cation. 0 represents the postulated enzymic hydrogen-bond donor (Bunting et al., 1980). D 0 us“ 0‘“ 0 H x0. H a HN | N) ———"' HN o N N N 0)": O I H H D D 4 I,”o| 0‘ HAN .1 H H 0 Figure A5. Productive bindin of xanthine a and h oxanthine b to X0 (Bunting and ngaskara, 1982).) yp ( ) BIBLIOGRAPHY BIBLIOGRAPHY Alfa-Laval, A. B. 1977. Method of improving the keeping quality of milk and other lipids. British Patent 1 468 405. Allen, J. C. and Wrieden, W. L. 1982. Influence of milk proteins on lipid oxidation in aqueous emulsion II. Lactoperoxidase, lactoferrin, superoxide dismutase and xanthine oxidase. J. Dairy Res. 49:249. Al-Khalidi, U. A. S. and Chaglassian, T. H. 1965. The species distribution of xanthine oxidase. Biochem. J. 97:318. Anderson, M. and Cawston, T. E. 1975. Reviews of the progress of dairy science. The milk-fat globule membrane. J. Dairy Res. 42:459. Andrews, P., Bray, R. C., Edwards, P., and Shooter, K. V. 1964. The chemistry of xanthine oxidase. II. Ultracentrifuge and gel- filtration studies on the milk enzyme. Biochem. J. 93:627. Aurand, L. W. and Woods, A. E. 1959. Role of xanthine oxidase in the development of spontaneously oxidized flavor in milk. J. Dairy Sci. 42:1111. Aurand, L. W., Boone, N. H., and Giddings, G. G. 1977. Superoxide and singlet oxygen in milk lipid peroxidation. J. Dairy Sci. 60:363. Aurand, L. W., Chu, T. M., Singleton, J. A., and Shen, R. 1967. Xanthine oxidase activity and development of spontaneously oxidized flavor in milk. J. Dairy Sci. 50:465. Avis, P. G., Bergel, F., and Bray, R. C. 1955. Cellular constituents. The chemistry of xanthine oxidase. Part I. The preparation of a crystalline xanthine oxidase from cow's milk. J. Chem. Soc. 1100. Avis, P. G., Bergel, F., Bray, R. C., James, D. W. F., and Shooter, K. V. 1956a. Cellular constituents. The chemistry of xanthine oxidase. Part II. The homogeneity of crystalline metalloflavoprotein fractions. J. Chem. Soc. 1212. Avis, P. G., Bergel, F., and Bray, R. C. 1956b. Cellular constituents. The chemistry of xanthine oxidase. Part III. Estimations of the co-factors and the catalytic activities of enzyme fraction from cow's milk. J. Chem. Soc. 1219. 137 138 Baker, 8. R. and Sienbeneick, H-U. 1971. Irreversible enzyme inhibitors. 184. An affinity column for purification of rat liver guanine deaminase and xanthine oxidase. J. Med. Chem. 14:799. Baker, H. and Frank, 0. 1975. Analysis of riboflavin and its derivatives in biological fluids and tissue. In "Riboflavin", Ed. Rivlin, R. S. p.54-56. Plenum Press, New York and London. Ball, E. G. 1939. Xanthine oxidase: Purification and properties. J. Biol. Chem. 128:51. Barber, M. J. and Siegel, L. M. 1982a. Oxidation-reduction potentials of molybdenum, flavin, and iron-sulfur center in milk xanthine oxidase: Variation with pH. Biochemistry 21:1638. Barber, M. J., Salerno, J. C., and Siegel, L. M. 1982b. Magnetic interactions in milk xanthine oxidase. Biochemistry 21:1648. Basha, S. M. M. and Roberts, R. M. 1977. A simple colorimetric method for the determination of tryptophan. Anal. Biochem. 77:378. Battelli, M. G., Lorenzoni, E., and Stirpe, F. 1973. Milk xanthine oxidase type D (dehydrogenase) and type 0 (oxidase). Biochem. J. 131:191. Battelli, M. G. 1980. Enzymic conversion of rat liver xanthine oxidase from dehydrogenase (0 form) to oxidase (0 form). FEBS Letters 113:47. Bergel, F. and Bray, R. C. 1959. The chemistry of xanthine oxidase. 4. The problems of enzyme inactivation and stabilization. Biochem. J. 73:182. Bhavadasan, M. K. and Ganguli, N. C. 1980. Free and membrane-bound xanthine oxidase in bovine milk during cooling and heating. J. Dairy Sci. 63:362. Bhavadasan, M. K., Abreham, M. J., and Ganguli, N. C. 1982. Influence of agitation on milk lipolysis and release of membrane-bound xanthine oxidase. J. Dairy Sci. 65:1692. Biasotto, N. 0. and Zikakis, J. P. 1975. Isolation of low molecular weight active xanthine oxidase from bovine whole milk. J. Dairy Sci. 58:1238. Bigelow, C. C. 1967. On the average hydrophobicity of proteins and the relation between it and protein structure. J. Theoret. Biol. 16:187. Bjorck, L. and Claesson, 0. 1979. Xanthine oxidase as a source of hydrogen peroxide for the lactoperoxidase system in milk. J Dairy Sci. 62:1211. Bray, R. C. and Malmstron, B. G. 1964. The chemistry of xanthine oxidase. 12. The amino acid composition. Biochem. J. 93:633. 139 Bray, R. C. 1975. Molybdenum iron-sulfur flavin hydroxylases and related enzymes. In "The Enzyme", Vol 12, 3rd ed., Ed. Boyer, P. D., p.299. Academic Press, New York. Briley, M. S. and Eisenthal, R. 1974. Association of xanthine oxidase with the bovine milk-fat-globule membrane. Catalytic properties of the free and membrane-bound enzyme. Biochem. J. 143:149. Briley, M. S. and Eisenthal, R. 1975. Association of xanthine oxidase with the bovine milk-fat-globule membrane. Nature of the enzyme- membrane association. Biochem. J. 147:417. Bruder, G., Heid, H., Jarasch, E. D., Keenan, T. W., and Mather, I. H. 1982. Characteristics of membrane-bound and soluble forms of xanthine oxidase from milk and endothelial cells of capillaries. Biochim. Biophys. Acta 701:357. Brunner, J. R. 1974. Physical equilibria in milk: The lipid phase. In "Fundementals of Dairy Chemistry" 2nd ed., Ed. Weble, B. H., Johnson, A. H., and Alford, J. A. p.474. The AVI Publishing Co. Inc. Westport, Conn. Buchheim, W. 1982. Paracrystalline arrays of milk fat globule membrane-associated protein as revealed by freeze-fracture. Naturwissenchaften 69(S):505 Bunting, J. W., Laderoute, K. R., and Norris, D. J. 1980a. Specificity of xanthine oxidase for nitrogen heteroaromatic cation substrates. Can. J. Biochem. 58:49. Bunting, J. W., Laderoute, K. R., and Norris, D. J. 1980b. The pH dependence of xanthine oxidase catalysis in basic solution. Can. J. Biochem. 58:394. Bunting, J. W. and Gunasekare, A. 1982. An important enzyme-substrate binding interaction for xanthine oxidase. Biochim. Biophys. Acta 704:444. Burt, J. R., Murray, J., and Stroud, G. D. 1968. An improved automated analysis of hypoxanthine. J. Food Tech. 3:165. Carey, F. G., Fridovich, I., and Handler, P. 1961. Preparation of several forms of xanthine oxidase by enzymic proteolysis. Biochim. Biophys. Acta 53:440. Cavallini, D., Graziani, M. T., and Dupre, S. 1966. Determination of disulphide groups in proteins. Nature 212:294. Chu, A. E. and Chaykin, S. 1974. Structural requirement of ligands for affinity chromatography absorbents: Purification of aldehyde and xanthine oxidase. In “Immobilized Biochemicals and Affinity Chromatography", Ed. Dunlap, R. B. p.162. Plenum Press, New Yor . 140 Clare, 0. A., Blakistone, B. A., Swaisgood, H. E., and Horton, H. R. 1981. Sulfhydryl oxidase-catalyzed conversion of xanthine dehydrogenase. Arch. Biochem. Bi0phys. 211:44. Clark, A. J. and Pratt, D. E. 1976. Xanthine oxidase activity in rat serum after administration of homogenized bovine cream preparation. Life Sci. 19:887. Coleman, R. 1973. Membrane-bound enzymes and membrane ultrastructure. Biochim. Biophys. Acta 300:1. Corran, H. S., Dewan, J. G., Gordon, A. H., and Green, D. E. 1939. CCXI. Xanthine oxidase and milk flavoprotein. Biochem. J. 33:1694. Coughlan, M. P. 1980. Aldehyde oxidase, xanthine oxidase and xanthine dehydrogenase: Hydroxylases containing molybdenum, iron-sulphur and flavin. In "Molybdenum and Molybdenum-Containing Enzymes", Ed. Coughlan, M. P., p.121. Pergamon Press. Oxford and New York. Coughlan, M. P. and Cleere, W. F. 1976. On the interaction between sulphide and turkey liver xanthine dehydrogenase. In “Flavins and Flavoproteins“, Ed. Singer, T. P., p.576. Elsevier Scientific ( Publishing Co. Amsterdam, The Netherlands. Deeth, H. C. 1981. Homogenized milk and coronary heart disease: A review. (in press). Della Corte, E. and Stirpe, F. 1972. The regulation of rat liver xanthine oxidase. Involvement of thio groups in the conversion of the enzyme activity from dehydrogenase (type 0) into oxidase (type 0) and purification of the enzyme. Biochem. J. 126:739. DeMott, B. J. and Praepanitchoi, 0. A. 1978. Influence of storage, heat and homogenization upon xanthine oxidase activity of milk. J. Dairy Sci. 61:164. Dixon, M. and Thurlow, S. 1964. Studies on xanthine oxidase. I. Preparation and properties of active material. Biochem. J. 18:971. Dougherty, T. M. 1976. A seqaitive assay for xanthine oxidase using commercial available [ C] xanthine. Anal. Biochem. 74:604. Dowben, R. M., Brunner, J. R., and Philpott, D. E. 1967. Studies on milk fat globule membranes. Biochim. Biophys. Acta 135:1. Dvorak, J., Macha, J., Masek, N., and Hanakova, L. 1980. Xanthine oxidase variation in cow's milk. Acta Univ. Agric. (Brno), Fac. Agro. XXVIII 183 (USSR). Edmonsdon, D., Massey, V., Palmer, G., Becham III, L. M., and Elion, G. B. 1972. The resolution of active and inactive xanthine oxidase by affinity chromatography. J. Biol. Chem. 247:1597. 141 Eigel, W. N. 1980. Characterization of bovine milk fat globule membrane prepared in the presence of chloroquine and Kallikrein inhibitor. J Dairy Sci. 63, Suppl. 1:158. Ellman, G. L. 1959. Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82:70. Franke, W. W., Heid, H. W., Grund, C., Winter, S., Freudenstein, C., Schmid, E., Jarasch, E-D., and Keenan, T. W. 1981. Antibodies to the major insoluble milk fat globule membrane-associated protein: Specific location in apical regions of lactating epithelial cells. J. Cell Biol. 89:485. Freudenstein, C., Keenan, T. W., Eigel, W. N., Sasaki, M., Stadler, J., and Franke, W. W. 1979. Preparation and characterization of the inner coat material associated with fat globule membrane from bovine and human milk. Exp. Cell Res. 118:277. Fried, R., Fried, L. W., and Babin, D. R. 1973. Biological role of xanthine oxidase and tetrazolium reductase inhibitor. Eur. J. Biochem. 33:439. Gandhi, M. P. S. and Ahuja, S. P. 1979. Absorption of xanthine oxidase from the intestines of rats and rabbits and its role in initiation of atherosclerosis. Zbl. Vet. Med. A, 26:635. Gilbert, 0. A. and Bergel, F. 1964. The chemistry of xanrhine oxidase. 9. An improved method of preparing the bovine milk enzyme. Biochem. J. 90:350. Greenbank, G. R. and Pallansch, M. J. 1962. Inactivation and reactivation of xanthine oxidase in dairy products. J. Dairy Sci. 45:958. Greenlee, L. and Handler, P. 1964a. Xanthine oxidase. VI. Influence of pH on substrate specificity. J. Biol. Chem. 239:1090. Greenlee, L. and Handler, P. 1964b. Xanthine oxidase. VII. Inhibition by amino group reagents. J. Biol. Chem. 239:1096. Groman, K. and Groman, A. 1974. Methods of checking correctness of stirrer choice for dairy storage and fermentation tanks. Roczmiki Instytutu Przymstu Mlecyarrkiego. 16:45 (In Dairy Sci. Abst. 1975, 37:5441). Gudnason, G. V. and Shipe, W. F. 1962. Factors affecting the apparent activity and heat sensitivity of xanthine oxidase in milk. J. Dairy Sci. 45:1440. Habeeb, A. F. S. A. 1972. Reaction of protein sulfhydryl groups with Ellman's reagent. Methods in Enzymol. 25:457. Hart, L. T., McGartoll, M. A., Chapman, H. R., and Bray, R. C. 1970. The composition of milk xanthine oxidase. Biochem. J. 116:851. 142 Herald, C. T. and Brunner, J. R. 1957. The fat globule membrane of normal cow's milk. 1. The isolation and characterization of 2 membrane protein fractions. J. Dairy Sci. 40:948. Hicks, C. L. 1980. Occurrence and consequence of superoxide dismutase in milk products: A review. J. Dairy Sci. 63:1199. Hille, R., Fee, J. A., and Massey, V. 1981a. Equilibrium properties of xanthine oxidase containing FAD analogs of varying oxidation- reduction potential. J. Biol. Chem. 256:8933. Hille, R. and Massey, V. 1981b. Studies on the oxidative half-reaction of xanthine oxidase. J. Biol. Chem. 256:9090. Ho, C. Y. and Clifford, A. J. 1976. Digestion and absorption of bovine milk xanthine oxidase and its role as an aldehyde oxidase. J. Nutri. 106:1600. Ho, C. Y. and Clifford, A. J. 1977. Bovine xanthine oxidase, blood lipids and coronary plaques in rabbits. J. Nutri. 107:758. Ho, C. Y., Crane, R. T., and Clifford, A. J. 1978. Studies on lymphatic absorption of and the available of riboflavin from bovine milk xanthine oxidase. J. Nutri. 108:55. Hood, L. F. and Patton, S. 1973. Isolation and characterization of intracellular lipid droplets from bovine mammary tissue. J. Dairy Sci. 56:858. Horeker, B. C. and Heppel, L. A. 1949. The reduction of cytochrome c by xanthine oxidase. J. Biol. Chem. 178:683. Hwang, Q. S., Ramachandran, K. S., and Whitney, R. McL. 1967. Presence of inhibitors and activators of xanthine oxidase in milk. J. Dairy Sci. 50:1723. Jackson, R. H. and Pallansch, M. J. 1961. Influence of milk protein on interfacial tension between butter oil and various aqueous phases. J. Agri. Food Chem. 9:424. Jahn, F. 0., Howe, J. L., Codurie, R. J., and Rand, A. G. 1976. A rapid visual enzyme test to assess fish freshness. Food Technol. 30(7):27. Janolino, V. G. and Swaisgood, H. E. 1982. Partial characterization of skim milk membrane isolated by permeation chromatography on derivatized controlled-pore glass. J. Dairy Sci. 65, Suppl. 1:61. Jarasch, E-D., Grund, C., Bruder, G., Heid, H. W., Keenan, T. W., and Franke, W. W. 1981. Localization of xanthine oxidase in mammary- gland epithelium and capillary endothelium. Cell 25:67. 143 Jenness, R. and Koops, P. 1962. Preparation and properties of a salt solution which simulates milk ultrafiltrate. Neth. Milk Dairy J. 16:153. Johnson, J. L., Rajagopalan, K. V., and Cohen, H. J. 1974. Molecular basis of the biological function of molybdenum: Effect of tungsten on xanthine oxidase and sulphite oxidase in the rat. J. Biol. Chem. 249:859. Kaclker, H. M. 1947. Differential spectrophotometry of purine compounds by means of specific enzymes. J. Biol. Chem. 167:429. Keenan, T. W. and Huang, C. M. 1972. Membrane of mammary gland. VI. Lipid and protein composition of Golgi apparatus and rough endoplasmic reticulum from bovine mamnary gland. J. Dairy Sci. 55:1586. Keenan, T. W., Heid, H. W., Stadler, J., Jarasch, E-D., and Franke, W. W. 1982. Tight attachment of fatty acids to proteins associated with milk lipid globule membrane. Eur. J. Cell Biol. 26:270. Keenan, T. W., Moon, T-W., and Dylewski, D. P. 1983. Lipid globules retain globule membrane material after homogenization. J. Dairy Sci. 66:196. Kellogg III, E. D. and Fridovich, I. 1975. Superoxide, hydrogen peroxide, and singlet oxygen in lipid peroxidation by a xanthine oxidase system. J. Biol. Chem. 250:8812. King, R. L. and Dunkley, W. L. 1959. Relation of natural copper in milk to incidence of spontaneous oxidized flavor. J. Dairy Sci. 42:420. Kitchen, B. J., Taylor, G. C., and White, I. C. 1970. Milk enzyme- their distribution and activity. J Dairy Res. 37:279. Kitchen, 8. J. 1974. A comparison of the pr0perties of membranes isolated from bovine skim milk and cream. Biochim. Biophys. Acta 356:257. Komai, H., Massey, V., and Palmer, 0. 1969. The preparation and properties of deflavo xanthine oxidase. J. Biol. Chem. 244:1692. Krenitsky, T. A., Neil, S. M., Elion, G. B., and Hitchings, G. H. 1972. A comparison of the specificities of xanthine oxidase and aldehyde oxidase. Arch. Biochem. Biophys. 150:585. Kuramoto, S., Jenness, R., and Coulter, S. T. 1957. The oxidation of vanillin by unheated milk. J. Dairy Sci. 40:187. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriOphage T4. Nature 227:680. 144 Lewis, 0. A. M. 1966. Short ion-exchange method for the estimation of cysteine and methionine. Nature 209:1239. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265. Mangino, M. E. 1976. Characterization of xanthine oxidase in the fat globule membrane. Ph.D. thesis, Michigan State University, East Lansing, MI. Mangino, M. E. and Brunner, J. R. 1976. Homogenized milk: Is it really the culprit in dietary-induced atherosclerosis. J. Dairy Sci. 59:1511. Mangino, M. E. and Brunner, J. R. 1977a. Isolation and partial characterization of xanthine oxidase associated with the milk fat globule membrane of cow's milk. J. Dairy Sci. 60:841. Mangino, M. E. and Brunner, J. R. 1977b. Compositional homology of membrane-protein system and membrane-associated proteins: Comparison with milk fat globule membrane and "membrane"-derived xanthine oxidase. J. Dairy Sci. 60:1208. Manson, W. and Modi, V. V. 1957. Flavin adenine dinucleotide (FAD) metabolism and lactation. Biochim. Biophys. Acta 24:423. Marchalonis, J. J. and Weltman, J. K. 1971. Relatedness among proteins: A new method of estimation and its application to immunoglobulins. Comp. Biochem. Physiol. 38(B):609. Massey, V., Brumby, P. E., and Komai, H. 1969. Studies on milk xanthine oxidase. Some spectral and kinetic properties. J. Biol. Chem. 244:1682. Massey, V. and Edmondson, D. 1970. On the mechanism of inactivation of xanthine oxidase by cyanide. J. Biol. Chem. 245:6595. Mather, I. H., Nace, C. S., Johnson, V. G., and Goldsby, R. A. 1980. Preparation of monoclonal antibodies to xanthine oxidase and other proteins of bovine milk-fat-globule membrane. Biochem. J. 188:925. McCarthy, R. D. and Long, C. A. 1976. Bovine milk intake and xanthine oxidase activity in blood serum. J. Dairy Sci. 59:1059. McGartoll, M. A., Pick, F. M., Swann, J. C., and Bray, R. C. 1970. Properties of xanthine oxidase preparations dependent on the proportions of active and inactive enzyme. Biochim. Biophys. Acta 212:523. Melachouris, N. 1969. Discontinuous gel electrophoresis of whey protein, casein and clotting enzyme. J. Dairy Sci. 52:456. 145 Modi, V. V. E., Owen, E. C., and Proudfoot, R. 1959. Species differences in the occurence of xanthine oxidase in milk. Proc. Nutri. Soc. 18:i (Abstract). Moore, S., Spackman, D. H., and Stein, W. H. 1958. Chromatography of amino acids on sulfonated polystyrene resins. An improved system. Anal. Chem. 30:1185. Moore, 0. J., Franke, W. W., Deumling, B., Nyquist, S. E., and Ovtract, L. 1971. Golgi apparatus function in membrane flow and differentiation: Origin of plasma membrane from endoplasmic reticulum. In "Biomembrane", Ed. Manson, L. Plenum Press, New York. Morell, D. B. 1952. The nature and catalytic activities of milk xanthine oxidase. Biochem. J. 51:657. Morton, R. K. 1953. Microsomal particles of normal cow's milk. Nature 171:734. Morton, R. K. 1954. The lipoprotein particles in cow's milk. Biochem. J. 57:231. Nalger, L. G. and Vartanyan, L. S. 1973. The quaternary structure of milk xanthine oxidase. Dissociation in sodium dedocylsulfate. Biokhimiya 38:561. (Original article is in Russian and is translated by Consultants Bureau, Plenum Publishing Co. in 1975). Nalger, L. G. and Vartanyan, L. S. 1976. Subunit structure of bovine milk xanthine oxidase. Effect of limited cleavage by proteolytic enzymes on activity and structure. Biochim. Biophys. Acta 427:78. Nashino, T., Nashino, T., and Tsushima, K. 1981. Purification of highly active milk xanthine oxidase by affinity chromatography on Sepharose 4B/folate gel. FEBS Letters 131:369. Nathans, G. R. and Hade, E. P. K. 1975. Proteolytic activity in bovine milk xanthine oxidase preparations. Biochem. Biophys. Res. Comm. 66:108. Nathans, G. R. and Hade, E. P. K. 1978. Bovine milk xanthine oxidase. Purification by ultrafiltration and conventional methods which omit addition of proteases. Some criteria for homogeneity of native xanthine oxidase. Biochim. Biophys. Acta 526:328. Nelson, C. A. and Handler, P. 1968. Preparation of bovine xanthine oxidase and the subunit structure of some iron flavoprotein. J. Biol. Chem. 243:5368. Nishikimi, M. 1975. Oxidation of ascorbic acid with superoxide anion generated by the xanthine-xanthine oxidase system. Biochem. Biophys. Res. Comm. 63:463. 146 Olson, J. S., Ballou, D. P., Palmer, G., and Massey, V. 1974a. The reaction of xanthine oxidase with molecular oxygen. J. Biol. Chem. 249:4350. Olson, J. S., Ballou, D. P., Palmer, G., and Massey, V. 1974b. The mechanism of action of xanthine oxidase. J. Biol. Chem. 249:4363. Oster, K. A. 1971. Plasmalogen diseases: A new concept of the etiology of the atherosclerotic process. Am. J. Clin. Res. 2:30. Oster, K. A. 1976. The absorption and inhibition of xanthine oxidase. Am. Lab. 8(10):47. Patton, S. and Fowkes, F. M. 1967. The role of the plasma membrane in the secretion of milk fat. J. Theoret. Biol. 15:274. Patton, S. and Keenan, T. W. 1975. The milk fat globule membrane. Biochim. Biophys. Acta 45:273. Pederson, T. C. and Aust, S. D. 1973. The role of superoxide and singlet oxygen in lipid peroxidation promoted by xanthine oxidase. Biochem. Biophys. Res. Comm. 52:1071. Phipps, L. W. and Temple, 0. M. 1982. Surface properties of milk fat globule: Interfacial tension studies. J. Dairy Res. 49:61. Plantz, P. E. and Patton, S. 1973a. Plasma membrane fragments in bovine and caprine skim milk. Biochim. Biophys. Acta 291:51. Plantz, P. E., Patton, S., and Keenan, T. W. 1973b. Further evidence of plasma membrane material in skim milk. J. Dairy Sci. 56:978. Porras, A. G., Olson, J. S., and Palmer, G. 1981. The reaction of reduced xanthine oxidase with oxygen: Kinetics of peroxide and superoxide formation. J. Biol. Chem. 256:9096. Rajan, T. S., Richardson, G. A., and Stein, R. W. 1962. Xanthine oxidase activity of milks in relation to stage of lactation, feed, and incidence of spontaneous oxidation. J. Dairy Sci. 45:933. Reiter, B. 1978. Antimicrobial systems in milk. J. Dairy Res. 45:131. Ross, 0. J., Sharnick, S. V., and Oster, K. A. 1980. Liposomes as a proposed vehicle for the persorption of bovine xanthine oxidase. Proc. Soc. Exp. Biol. Med. 163:141. Segel, I. H. 1975. Rapid equilibrium partial and mixed-type inhibition. In "Enzyme Kinetics", Ed. Segel, I. H., Chapter 4. John Wiley and Son, New York. 147 Shipe, W. F. 1977. Enzymatic modification of milk flavor. In "Enzymes in Food and Beverage Processing", Ed. Ory, R. L. and St.Angelo, A. J., p.57. Am. Chem. Soc., Washington, D. C. Silver, M. R. and Zikakis, J. P. 1979. Comparative kinetic characterization of proteolytic and non-proteolytic prepared milk xanthine oxidase. J. Dairy Sci. 62, Suppl. 1:245. Smith, G. J. and Dunkley, W. L. 1960. Xanthine oxidase and incidence of spontaneous oxidized flavor in milk. J. Dairy Sci. 43:278. Smith, G. J. and Dunkley, W. L. 1962. Ascorbic acid oxidation and lipid peroxidation in milk. J. Food Sci. 27:127. Spence, J. T., Barber, M. J., and Siegel, L. M. 1982. Determination od the stoichiometry of electron uptake and the midpoint reduction potentials of milk xanthine oxidase at 25 C by microcoulometry. Biochemistry 21:1656. Stannard, D. J. 1975. The use of marker enzymes to assay the churning of milk. J. Dairy Res. 42:241. Stewart, P. S., Puppione, D. L., and Patton, S. 1972. The presnece of microvilli and other membrane fragments in the non-fat phase of bovine milk. Z. Zellforsch 123:161. Stirpe, F. and Della Corte, E. 1969. The regulation of rat liver xanthine oxidase. Conversion in vitro of the enzyme activity from dehydrogenase (type D) to oxidase (type 0). J. Biol. Chem. 244:3855. Stirpe, F. and Della Corte, E. 1970. The regulation of rat liver xanthine oxidase: Conversion of type D (dehydrogenase) into type 0 (oxidase) by a thermolabile factor, and reversibility by dithioerythritol. Biochim. Biophys. Acta 212:195. Swope, F. C., Brunner, J. R., and Vadehra, D. V. 1965. Riboflavin and its natural derivatives in the fat-globule membrane. J. Dairy Sci. 48:1707. Swope, F. C. and Brunner, J. R. 1968. The fat globule membrane of cow's milk: A reassessment of isolation procedures and mineral composition. Milchwissenschaft 23:470. Ultmann, J. E., Feigelson, P., and Harris, S. 1962. The effect of specific antibodies on xanthine oxidase from various sources. J. Immunol. 88:113. Volp, R. F. and Lage, G. L. 1977. Studies on the intestinal absorption of bovine xanthine oxidase. Proc. Soc. Exp. Biol. Med. 154:488. Walstra, P. 1965. Light sacttering by milk fat globules. Neth. Milk Dairy J. 19:93. 148 1 Waud, W. R., Brady, F. 0., Wiley, R. 0., and Rajagopalan, K. V. 1975. A new purification procedure for bovine milk xanthine oxidase: Effect of proteolysis on the subunit structure. Arch. Biochem. Biophys. 169:695. Waud, W. R. and+Rajagopalan, K. V. 1976a. Purification and properties of the NAD -dependent (type D) and 0 -dependent (type 0) forms of rat liver xanthine dehydrogenase. Argh. Biochem. Biophys. 172:354. Waud, W. R. and Rajagopalan, K. V. 1976b. The mechanism of conversion of rat liver xanthine dehydrogenase from an NAD -dependent form (type D) to an 02-dependent form (tyoe 0). Arch. Biochem. Biophys. 172:365. Waud, W. R. and Rajagopalan, K. V. 1976c. Rat liver xanthine dehydrogenase: Studies on dehydrogenase to oxidase conversion. In “Flavins and Flavoproteins“, Ed. Singer, T. P., p.566. Elsevier Scientific Publishing Co. Amsterdam, The Netherlands. Weber, K. and Osborn, M. 1969. The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J. Biol. Chem. 244:4406. Weiner, A. M., Platt, T., and Weber, K. 1972. Amino-terminal sequence analysis of proteins purified on a nan0mole scale by gel electr0phoresis. J. Biol. Chem. 247:3242. Weltman, J. K. and Dowben, R. M. 1973. Relatedness among contratile and membrane proteins: Evidence for evolution from common ancestral gene. Proc. Nat't. Acad. Sci. 70:3230. Williams, J. W. and Bray, R. C. 1981. Kinetic and e.p.r. studies on the inhibition of xanthine oxidase by alloxanthine (1 H- pyrazolo[3,4,d]pyrimidine-4,6-diol). Biochem. J. 195:753. Woods, K. R. and Wang, K. T. 1967. Separation of dansyl-amino acids by polyamide layer chromatography. Biochim. Biophys. Acta 133:369. Worden, A. N. 1943. Xanthine oxidase activity of cow's milk. Nature 152:505. Zikakis, J. P. and Treece, J. M. 1971. Xanthine oxidase polymorphism in bovine milk. J. Dairy Sci. 54:648. Zikakis, J. P., Dougherty, T. M., and Biasotto, N. 0. 1976a. The presence and some properties of xanthine oxidase in human milk and colostrum. J. Food Sci. 41:1408. Zikakis, J. P. and Rzucidlo, S. J. 1976b. Antigenicity of bovine milk. xanthine oxidase in guinea pig. J. Dairy Sci. 59:1051. 149 Zikakis, J. P., Rzucidlo, S. J., and Biasotto, N. O. 1977. Presistence of bovine milk xanthine oxidase activity after gastric digestion in vivo and in vitro. J. Dairy Sci. 60:533. Zikakis, J. P. 1979a. Preparation of high purity xanthine oxidase from bovine milk. US Patent 4 172 763. Zikakis, J. P. 1979b. Method for converting polarographic units of enzyme activity to international units. Am. Lab. 11(2):57. Zikakis, J. P. and Wooters, S. G. 1980. Activity of xanthine oxidase in dairy products. J. Dairy Sci. 63:893. Zittle, C. A., Dellamonica, E. S., Custer, J. H., and Rudd, R. K. 1956. The fat-globule membrane of milk: Alkaline ph05phatase and xanthine oxidase in skimmilk and cream. J. Dairy Sci. 39:528. Zittle, C. A., Dellamonica, E. S., Custer, J. H., and Rudd, R. K. 1959. Determination of xanthine oxidase in milk with triphenyl tetrazolium chloride. J. Dairy Sci. 39:522. ’4‘