This is to certify that the thesis entitled GALACTOSE OXIDATION IN RAT LIVER MICROSOMES presented by Nancy Ann Josef has been accepted towards fulfillment of the requirements for M . S . degree in Biochemistry 1/. / [4/ Q/JJA Major professor Date ,//{d/, . 07/3/71? 0-7639 GALACTOSE OXIDATION IN RAT LIVER MICROSOMES By Nancy Ann Josef A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biochemistry 1978 ABSTRACT GALACTOSE OXIDATION IN RAT LIVER MICROSOMES By Nancy Ann Josef Galactose oxidation was investigated in rat liver models to account for the production of galactonic acid in patients exhibiting classical galactosemia and in patients receiving galactose tolerance tests. A gas chromatographic assay was devised for the identi- fication and quantitation of galactonic acid. Rats main— tained on a 40% galactose diet excreted more than 30 mg of galactonate per day in the urine and accumulated galactonic acid in several tissues. Liver microsomes incubated with 30 mM galactose produced galactonic acid. This oxidizing activity was also seen in the mitochondrial fraction; the soluble fraction had no activity. Optimal activity occurred at pH 8.0, and inhibition was caused by heavy metals and sulfhydryl reagents. The apparent Km for galactose was 32.9 mM and Vmax was 160 nmoles of galactonic acid/4 hr/mg protein. The activity was specific for galactose although unidentified oxidation products of altrose, talose, and 2—deoxy—galactose were detected. Oxygen appeared to be the sole hydrogen acceptor; galactose dependent formation of hydrogen peroxide was demonstrated during the incubation period. These data suggest that an enzyme with galactose oxidase activity is present in rat liver microsomes. ACKNOWLEDGEMENTS I wish to express sincere appreciation to Dr. William W. Wells for his enthusiasm, patience, guidance, and finan- cial support throughout the course of these experiments. I thank my committee members Dr. Loran L. Bieber and Dr. Richard L. Anderson for taking the time to help in this endeavor. I wish to thank Thomas for his love and encouragement, and for having more confidence in me than I have in myself. ii TABLE OF CONTENTS List of Tables. List of Figures . List of Abbreviations . . . . . . . INTRODUCTION Rationale and Objectives Literature Survey. MATERIALS AND METHODS Materials . . . . . . . . . Hemoglobin Incubations Cell Incubations . . . . . . Animal Maintenance Tissue Collection. . . . Microsome Preparation Conditions of Incubations . Difference Spectra Galactonate Isolation . . . Galactonate Determinations. Hydrogen Peroxide Assay. Protein Separation RESULTS Hemoglobin and Blood Studies Ig_1;zg Studies Microsome Activities. Cellular Distribution . . . . . . . Galactonate Production with Time, Protein and Substrate . . . . . . . . . Effect of pH . . . . . . . . . . iii Page vi vii \OCDCI)\1 11 11 12 12 12 13 14 14 16 16 18 18 20 20 Activators and Inhibitors Substrate Specificity. Difference Spectra. Reaction with Hydrogen Acceptors . Formation of Hydrogen Peroxide. Effect of Age Distribution in Species Protein Separation. DISCUSSION. REFERENCES. iv 20 26 26 26 28 29 29 29 33 41 Table II. III. IV. VI. VII. VIII. LIST OF TABLES Diet Composition. Concentrations of Galactonate Found in Tissues of Rats Fed High Galactose Diets and Phenobarbital Water . . Galactose Oxidation Activity in Rat Liver Microsomes. . . . . . . Effects of Various Activators and Inhi- bitors on the Galactose Oxidizing Activ— ity of Rat Liver Microsomes . . . . Effects of Hydrogen Acceptors on the Galactose Oxidizing System in Rat Liver Microsomes. . . . . . . . . Effects of Age on Galactose Oxidation in Rat Liver Microsomes . . . . Galactose Oxidation in Various Species Purification of the Hydrogen Peroxide Producing Activity from Rat Liver Microsomes. . . . . . Page 10 17 19 24 27 3o 31 32 Figure LIST OF FIGURES Page Time Course of Microsomal Galactonate Pr0dUCtiono I O O I O O O O I 21 Lineweaver-Burk Analysis of Kinetic Data 22 Effect of pH on Microsomal Galactose Oxidation . . . . . . . . . . 23 vi BHT FAD FMN NAD NADH NADP+ NADPH SDS TMS Tris max LIST OF ABBREVIATIONS butylated hydroxytoluene flavin adenine dinucleotide flavin mononucleotide Michaelis constant nicotinamide adenine reduced nicotinamide nicotinamide adenine phosphate reduced nicotinamide cleotide phosphate dinucleotide adenine dinucleotide dinucleotide adenine dinu- sodium dodecyl sulfate trimethylsilyl tris (hydroxymethyl) maximal velocity vii aminomethane INTRODUCTION Rationale and Objectives The excretion of galactonic acid by patients with class- ical galactosemia and by subjects given galactose tolerance tests has been reported (1). The mechanism for this oxida- tion and the conversion of galactose to carbon dioxide in galactosemics has been widely disputed (2-7). Work in this laboratory has shown that high levels of galactose lead to impaired phagocytosis and oxygen-dependent killing of bacteria in polymorphonuclear leukocytes (8,9). In the course of those studies it was found that an in_yitrg Fenton reaction combining hydrogen peroxide and ferrous iron produced hydroxyl radicals that could convert galactose to galactonic acid (10). Since it is known that liver micro— somes are capable of producing hydrogen peroxide and hydrox- yl radicals (11), the liver microsome system was studied to determine if a similar hydroxyl radical mechanism could produce galactonic acid. To test this hypothesis, experiments were conducted using microsomes isolated from rat liver. Levels of galactose employed in these studies were comparable to those encountered in galactosemics (12,13). When it was shown that an ig_yitrg microsomal system could indeed produce galactonic acid, 2 various activators and inhibitors were used in attempts to elucidate the mechanism. The in_yitrg results of the micro— somal incubations were extrapolated to the in_yiyg system by dietary studies. The observed hydrogen peroxide production along with galactonic acid production led to the proposal that a protein with galactose oxidase activity was responsi- ble for the microsomal oxidation of galactose. Literature Survey Galactose is normally metabolized in most biological systems by conversion to glucose through the Leloir path- way (14—17): Galactose + ATP ——9 Galactose—l-Phosphate + ADP Galactose-l-Phosphate + UDP-Glucose--+ UDP-Galactose + Glucose—l-Phosphate UDP-Galactose'—-H*UDP-Glucose UDP—Glucose + PPi -6'UTP + Glucose-l-Phosphate Galactose reacts with ATP to form galactose—l—phosphate and ADP. The equilibrium is far in the direction of sugar phos- phorylation but the reaction is reversible (18). In the second step of the galactose-glucose interconversion, the enzyme galactose-l-phosphate uridylyltransferase reacts with uridine diphosphate glucose (UDP-glucose) and galactose-1- phosphate to give UDP—galactose and glucose—l—phosphate. The conversion of galactose to glucose by inversion of the hydrox— yl group at the fourth carbon of the hexose chain is catalyzed by uridine diphosphate galactose-4-epimerase. The last step 3 in the pathway is catalyzed by uridine diphosphate glucose pyrophosphorylase; UDP-glucose and pyrophosphate react to form glucose-l-phosphate and UTP. This reaction is reversible, allowing not only the original carbon chain of galactose to enter the pathway of glucose metabolism, but also enables UDP-glucose to be formed from glucose and UTP (19). Classical galactosemia is an inherited disease resulting from a deficiency in galactose-l-phosphate uridylyltransferase. The first detailed description of the syndrome by Mason and Turner in 1935 (20) was followed by numerous reports clearly establishing the clinical entity (21-25). The most common initial clinical symptom is failure to thrive, and occurs in almost all cases. Vomiting or diar— rhea usually starts within a few days of milk ingestion (23). Jaundice and hepatomegaly present after the first week of life and are accentuated by the severe hemolysis seen in some patients. Cataracts are observed a few days after birth, and retarded mental development is apparent in those first observed after the first several months of life (19). Chem- ical findings include elevated blood galactose, galactosuria, hyperchloremic acidosis, and amino aciduria (26,27). The acidosis may be secondary to the gastrointestinal disturbances, but may be a result of renal tubular dysfunction and a defect in urine acidification mechanisms (26). Albuminuria (27) and amino aciduria (28,29) are the result of a renal toxicity syndrome. Classical galactosemics who are never exposed to galactose should exhibit no abnormalities. The clinical symptoms are most likely the result of the accumulation of galactose-l-phosphate and the products of alternate path- ways of galactose metabolism (19,30), but in many instances the reason for the toxicity of an organ, especially the liver and brain, remain obscure (19). One alternate route of galactose metabolism involves the reduction of galactose to galactitol. This reaction requires NADPH and is catalyzed by aldose reductase. The Km value for galactose is between 12 and 20 mM (31,32), a level that is surpassed in galactosemia (12,13). Galactitol has been isolated from the tissues and urine of galactosemics (33—35) and administration of radioactive galactitol to normal patients has shown that the compound is not further metabolized or converted to carbon dioxide, but only excreted in the urine (36). The formation of galactitol in the lens has been directly linked to cataract formation (37,38). Since galactitol diffuses very slowly from the lens, it is osmotically active and there is an obligatory movement of water into the lens fiber, contributing to the cataract formation (39). Some patients who appear to have a total absence of galactose-l-phosphate uridylyltransferase are still able to oxidize small amounts or labelled galactose to carbon dioxide (40,41). It has been shown that galactosemic fibro- blasts produce 14 002 from galactose-l-luC (42), and radio- autography of cultured human galactosemic and normal cells also suggests the possibility of an alternate pathway for galactose metabolism (43). An auxiliary pathway for galactose metabolism was pro- posed by Isselbacher (44,45) where galactose-l—phosphate and uridine triphosphate react to form UDP- galactose and pyrophosphate. This reaction is catalyzed by uridine di- phosphate galactose pyrophosphorylase. The action of this enzyme could circumvent the block in transferase deficiency, but it was later shown that the activity of this enzyme in normal human liver is low (46), and the activity in liver biopsy specimens from galactosemics is insignificant (47). Inouye and coworkers reported the presence of galactose- 6-phosphate in galactosemic erythrocytes (48), but this has been disputed (49). Galactose-6-phosphate has been shown to be formed from galactose-l-phosphate by phosphogluco- mutase (50,51), or by hexokinase (52) in in yitgq systems, but the rate of formation is very small. The enzyme hexose- 6-phosphate dehydrogenase oxidizes galactose-6-phosphate to 6—phosphogalactonic acid (53); the latter was shown to be a dead end product of metabolism (5). Another liver enzyme, galactose-6-phosphate dehydrogenase oxidizes galactose-6- phosphate to a ketoaldose product (54). There is no direct evidence to suggest that this enzyme is involved in any 1n vivo pathway of galactose metabolism. Another pathway explaining galactose metabolism in galactosemics was proposed by Cuatrecasas and Segal (2): Galactose ——% Galactonolactone Galactonolactone --9Galactonic Acid Galactonic Acid -€*fl-Keto-Galactonic Acid fl—Keto—Galactonic Acid —-> Xylulose + 002 Xylulose + ATP ——> Xylulose-5-Phosphate Galactose is converted to galactonolactone by an NAD+ re- quiring enzyme, galactose dehydrogenase. A lactonase acts upon the lactone to produce galactonic acid, which proceeds through an unstable intermediate, A-keto-galactonic acid to xylulose and carbon dioxide. The enzymela-Lyhydroxy acid dehydrogenase was proposed to catalyze this step (2). Xylu- lose can be readily phosphorylated by liver xylulokinase to xylulose-5-phosphate. Galactose dehydrogenase was isolated and partially purified from rat liver cytosolic fraction (3) and character- ized as an enzyme inhibited by divalent metal ions, sulhydryl reagents, and exhibited a broad substrate specificity (4). However, other researchers have shown that the galactose dehydrogenase was actually alcohol dehydrogenase, and that the increase in absorbance at 340 nm.used to assay the activity was a result of alcohol contamination of sugar substrates (5-7). When alcohol was removed from the sugars, no activity was observed. In addition the isolation of galactonic acid from an elaborate reaction incubation in- volving hydrogen peroxide, peroxidase, and resorcinol, along with galactose, purified enzyme, and NAD+, could not be reproduced if only galactose, enzyme, and NAD+ were employed (5). Because of such conflicting data as outlined above, the mechanism of the oxidation of galactose to galactonic acid and carbon dioxide in totally transferase deficient galactosemics remains to be elucidated. MATERIALS AND METHODS Materials. All reagents used were analytical grade. Horse- radish peroxidase (EC 1.11.1.7) and catalase (EC 1.11.1.6) were from Worthington Biochemical Co. Catalase was purified before use by passing through Sephadex G-10. Superoxide dismutase (EC 1.15.1.1) was obtained from Sigma Chemical Co. Bovine hemoglobin was isolated earlier in this laboratory and stored at -20 degrees. DrGalactose, Demannose, Drglucose, Dexylose, Dearabinose, Defructose, nicotinamide adenine di- nucleotide (NAD+), nicotinamide adenine dinucleotide phosphate (NADP+), reduced nicotinamide adenine dinucleotide phosphate (NADPH), flavin mononucleotide (FMN), flavin adenine di- nucleotide (FAD), Dgribonic acid-E—lactone, 2,6—dichloro— phenolindolphenol, grdianisidine dihydrochloride, pyrazole, and butylated hydroxytoluene (BHT) were from Sigma. Galactose was found to be free of galactonic acid contamination by gas chromatography. Galactono-F-lactone was from General Biochemicals, hydrogen peroxide from Mallinckrodt, diphenyl- furan from Eastman, and diphenylisobenzofuran from Aldrich. DgTalose, Deallose, Dealtrose, and Lrgalactose were gifts from Dr. W. A. Wood. All rats were obtained from Holtzman (Madison, WI), and adult male guinea pigs were from Elm Hill Laboratories (Cambridge, MA). One week old chicks were were the gift of Dr. D. Polin. Mice of the C57BL/6J-+/+ strain were purchased from Jackson Laboratories (Bar Harbor, ME). Beef liver was obtained at a local slaughter house. Hemoglobin Incubations. Bovine hemoglobin was dissolved in 250 mM potassium phosphate, pH 7.4, to a concentration of 50 mg/ml and centrifuged to remove insoluble impurities. Sodium dithionite was added to reduce iron, and the solu- tion was passed over Sephadex G-25. Hydrogen peroxide solutions were made fresh and the concentration was deter- l). mined spectrophotometrically (6 =81 M” Reduced hemo- 230 nm globin (1.2 mg/ml), 2.0 mM hydrogen peroxide, and 30 mM galactose or glucose were incubated in 250 mM potassium phosphate, pH 7.4, in a total volume of 1.0 ml, for 2 hr at 37 degrees. Protein was precipitated with 200 ul of 30% trichloroacetic acid (TCA). Samples were centrifuged and the supernate was extracted three times with three volumes of diethyl ether. A few drops of 0.1 M NaOH were added to convert the products to the sugar anionic forms. The samples were dried under nitrogen and trimethylsilyl (TMS) derivatives were prepared as previously described (55). Gas chromatography was performed on the aldonate derivatives at 180 degrees using a Hewlett Packard model 5830 A gas chromatograph equipped with a 1.8 m x 2 mm glass column packed with 3% OV-l on Gas Chrom Q, 80-100 mesh (Applied Science Laboratories, Inc., State College, PA). Cell Incubations. Rats and guinea pigs were ether anesthe- tized and blood collected by heart puncture was placed in heparinized tubes. Blood was centrifuged at 5000 x g, plasma was removed, and red cells were resuspended to the original blood volume in Krebs-Ringer phosphate solution (without calcium or magnesium to prevent clumping), pH 7.4 (56). Red cells (3.0 x 109) or plasma (0.5 ml) were incubated with 1.0 mM hydrogen peroxide and 30 mM galactose in a total volume of 1.0 ml for 4 hr at 37 degrees. Samples were protein precipitated with 200 ul 30% TCA, and samples were treated as outlined above. Animal Maintenance. The effects of phenobarbital on micro- somal galactose oxidizing activity and the tissue distri- bution of galactonate were observed through diet studies. Male rats weighing 250-300 g were randomly divided into two groups, fed a commercial chow diet (Allied Mills, Inc., Chicago, IL), and were allowed to drink distilled water or water containing 0.1% phenobarbital, pH 7.0, ad libitum. After 7 days these groups were randomly subdivided, and the rats were placed in individual stainless steel metabolism cages. One half of each group was maintained on a control or 40% galactose containing diet (Table I) for an additional 72 hours, and the water regimen was continued. The diets were fed ad libitum. A supplemental salt mixture was added to bring the essential elements up to accepted levels. Urine was collected under toluene for the last 24 hours before sacrifice. TABLE I Diet Compositiona Component Concznfigation Galactose or Sucrose 400 Vitamin-Free Casein 250 Sucrose 226 Soybean Oil 50.0 Phillips-Hart Salt Mixtureb 50.0 Alpha Cellulose 12.5 Vitamin Mix 10.0 Choline Chloride 1.0 Supplemental SaltC 0.5 aNo galactonate contamination was found in the diets. bThe Phillips-Hart salt mixture (Teklad Test Diets) consisted of (% by weight): 30.00% CaCOB, 7.50% CaHP04°2H20, 0.005% CoClZ-6H20, 0.003% CuSOu-5H20, 32.2% KZHP04, 2.75% ferric citrate, 10.2% MgSOu-7H20, 0.08% KI, 16.7% NaCl, and 0.025% ZnClz. CThe supplemental salt mixture consisted of (% by weight): 6.123% CuSOu-5H20, 0.500% CoClZ-6H20, 14.004% ZnO, 38.348% MnSOu'HZO, and 41.035% alpha cellulose. 10 11 Tissue Collection. Animals were anesthetized with diethyl ether and blood taken by heart puncture was placed in hepa- rinized tubes. Livers were removed and samples (2 g) were minced in 10-15 ml ice cold 0.15 M potassium phosphate buffer pH 7.4, for microsomal preparation. Intestines were removed, rinsed with distilled water, and together with kidney, heart, brain, and spleen, were stored at -20 degrees until analysis. The metabolism cages were rinsed with 10 ml distilled water, rinses were combined with the urine, and urine was stored under toluene at 4 degrees. Microsome Preparation. Microsomes were prepared at 4 degrees by a modification of the method of May and McCay (57). Livers were removed from starved rats (those on special diets were not starved), minced in 25 ml of 0.15 M potassium phos- phate, pH 7.4, and homogenized with a Potter—Elvehjem homogenizer. Nuclei and cellular debris were removed by centrifuging at 300 x g for 12 min. The supernate was centrifuged again,20,000 x g for 12 min. The resulting supernate was thenwvow HonPcOo was mo anooaom mm commonmxo .monEMm oopnv op ozp mo apfl>apow owmno>m cap mpsomonmon osam> somm .cOflPoom mvonpos was 966:5 ponaaomov mm poamwmm was apa>flpom coapmcwxo omopomamwm o.ooa as sm.o assa so asssz c.6m as sm.o cscpccs s.~ms a on campssmas ceascscssm m.cs as o.H ceass ssascm H.NNH s cm sesameso n.6HH as ~.o aoa o.ooH as m 9am o.ooa as o.m seam o.ooa as m escasesesau o.cos as o.m maowa m.mm as ea caosssaa o.ss as m.o Names m.ms samssz .+Ncs-mm< s.:m as m.o «floss m.HNH eaassa .+mcs-mm< o as m.o semen o.ooH as o.H sassecsscecmaaascseam m.6a as m.o Hoescsseoasess o.oos as o.a sasssaascsesm m.mm as o.H msacpmao o.sa as ooH sessoasa o.mm as o.s csaemao m.sm as ea cescsscm w.om as H.o ceasacassaaspm-a o.ooH as ooa assasssa m.ms as o.a ceasseccsoscH m.soa as om Hosasem N.HH as m.o cascssceaesoscscscasoum gamma mass gawk mass mmmsomosoflz Hm>aq Pam mo hpa>apo< msflsflwflxo omopomaww map Co msoPHpflncH cam whopm>apo¢ msoaaw> mo mpoomwm >H mqm¢e 25 activity, suggesting that metal cofactors are not involved in the activity. Organic solvents had a variable effect; 0.27 mM acetone decreased the activity by 50% while up to 30 mM ethanol had no effect on the activity. The addition of NADPH and NADH had no effect on the oxidation of galactose. An isocitrate/isocitrate dehydrogenase system was used to regenerate NADPH, and NADH was added after every half hour of the incubation. The effects of several other inhibitors on the micro- somal galactose oxidizing system were studied in order to determine if the observed activity was the result of various known oxidative mechanisms. Singlet oxygen trappers, diphenyl- furan and diphenylisobenzofuran, had no effect on the galac- tose oxidizing system. Hydroxyl radical scavengers, ben- zoate (10 mM) and thiourea (100 mM), showed a 45% and 88% decrease in activity, but mannitol (100 mM) and ethanol (30 mM) had no effect. Thus, these data are not consistent with the involvement of a hydroxyl radical mechanism. Pyrazole (10 mM) was used to rule out alcohol dehydrogenase acti— vity. Addition of lipid peroxidation enhancers, ADP-Fe3+ and NADPH, showed a 20% increase in activity, but cyanide did not decrease the oxidative activity, arguing against a lipid peroxidation mechanism. Glutathione and BHT had no effect on the activity, ruling out a spontaneous oxidation. Addition of 90 units of catalase or 70 units of superoxide dismutase caused a 20% increase in activity. Sodium azide (1 mM) decreased activity 25%. 26 Substrate Specificity. Several sugars were tested as sub- strates in the microsomal galactose oxidizing system. No oxidation products were found in the gas chromatographic assay with Deglucose, D—mannose, D—allose, Q—fructose, D-xylose, Dearabinose, or 2—deoxy-D—glucose as substrates. Unidentified oxidation products were found with Q-altrose (15.6% of control), Detalose (17.1% of control) and 2—deoxy- Q-galactose (63.8% of control). Incubations with L-galactose showed 13.4% of control activity. The specific rotation of the Legalactose used here (measured with a Zeiss polarimeter after overnight equilibration) was -72.3 degrees, compared with a specific rotation of +83.0 determined with authentic Degalactose. It is possible that much of the activity seen with the Legalactose may be attributable to a D—galactose contaminant. Difference Spectga. Difference spectra obtained through- out the microsomal incubation period were of little value. The incubation mixture was turbid and the detergents Triton X-100 (0.5%) and sodium deoxycholate (0.5%) were added to solubilize the protein. Samples so treated did not produce useful spectra, and still showed turbidity due to glycogen. Reaction with Hydrogen Acceptgrs. Little change in galactonate production was observed with the addition of NAD+ or NADP+ to the microsomal assay system (Table V). FAD and FMN (0.37 mM) acted as inhibitors, reducing the galactose ox- idizing activity to approximately 50% of the control value. TABLE V Effects of Hydrogen Acceptors on the Galactose Oxidizing System in Rat Liver Microsomesa Substance Concentration Percent of Control Oxygen Ambient 100.0 NAD+ 0.37 mM 90.2 NADP+ 0.37 mM 92.8 FAD 0.37 mM 50.6 FMN 0.37 mM 48.0 2,6-Dichlorophenolindolphenol 0.20 mM 0 aGalactose oxidation activity was assayed as described under the Methods section. Each value represents the average activity of two to four samples, expressed as percent of the control activity. 27 g 28 2,6-Dichlorophenolindolphenol (0.20 mM) completely destroyed all galactose oxidation. Taken together, these data strongly suggest that oxygen is the sole hydrogen acceptor of the microsomal galactose oxidizing system. Formation gf_Hydrogen Peroxide. Microsome samples were incubated in sodium phosphate buffer, pH 7.4, containing horseradish peroxidase, grdianisidine hydrochloride, and galactose. In the presence of hydrogen peroxide and perox- idase, pedianisidine is oxidized by the peroxidase to an orange colored product which can be measured spectropho- tometrically. Microsome incubations without galactose were used as blanks and peroxide concentration was determined from a standard curve. Color formation in standard perox- ide samples was not affected by high galactose concentra- tions. Half of the microsome incubation samples were treated with Dowex resin and analyzed for galactonate by gas chromatography; the other half were analyzed for hydro- gen peroxide. The galactose oxidizing activity of the micro- somes was inhibited by the Chromagen, a result also seen in the assay of fungal galactose oxidases (65). With the microsome preparations it was possible to demonstrate color formation that was dependent on the presence of galactose, but the galactonate to peroxide ratio varied from five to one to one to one. Bean and coworkers (66) have reported that competition between en- dogenous substrates and pedianisidine for the horseradish perosidase make quantitation by this method unfeasible. 29 Effect 9: Age. Neonatal rats (20 g) of mixed sex showed the greatest microsomal galactose oxidizing activity (Table VI). Liver microsomes from young adults (130 g), adults (260 g), and very old males (540 g) all showed approximately 62% of the activity observed with neonatal animals. Distribution ip_Species. The oxidation of galactose was also observed in mice and beef liver microsomes (Table VII). No activity was observed in guinea pig or chicken liver microsomes. Protein Separation. Attempts were made to further separate this galactose oxidizing activity from rat liver microsomes. The hydrogen peroxide assay was used to monitor the peroxide forming activity. A summary of the preliminary purification scheme is shown in Table VIII. There was no observable activity in the whole homogenate from rat liver. Ammonium sulfate fractionation of the solubilized microsomal frac- tion resulted in an increase in the total activity, and a nine fold purification. However, dialysis overnight at 4 degrees resulted in the loss of 66% of the observed activity. No protein could be eluted from the Sephadex G—150 column after washing with three column volumes of buffer. TABLE VI Effects of Age on Galactose Oxidation in Rat Liver Microsomesa Animal Weight Galactose Oxidation (g) (nmoles/4 hr/mg protein) Neonateb 20 120 _+_ 12.2 Young Adultc 130 76.0 i 5.6 Adultd 260 65.8 1 1.5 018° 540 82.1 1:. 15-3 aMicrosomes prepared and galactonate analyzed as outlined under the Methods section. n=2 bMixed sex CMales dFemales 30 TABLE VII Galactose Oxidation in Various Speciesa Species Activity (% of Control)f Rat 100 b Mouse 85.9 CowC 45.3 Guinea Pigd 0 Chickene 0 aMicrosomes prepared and galactonate analyzed as outlined under Methods section. n=3, except cow (n=1). guinea pigs (n=4). bMice were 3 months old, C57BL/6J-+/+ strain. CBeef liver obtained at slaughter house. dGuinea pigs weighed 400-600 g, 1%-6 months old. eChickens were 1 week old. fActivity is expressed as percent of activity compared to rat control. 31 TABLE VIII Purification of the Hydrogen Peroxide Producing Activity from Rat Liver Microsomesa Fraction Volume Protein Total Units Specific (ml) (mg/ml) (nmoles/min) Activity (units/mg) Microsomes 13.0 35.8 28.0 0.0614 Ammonium Sul- fate (0-40%) 3.3 21.3 38.6 0.549 Dialysis 2.1 21.3 12.8 0.286 3Hydrogen peroxide formation monitored by using dianisidine- peroxidase assay described under the Methods section. assay mix: Typical 1.0 ml Chromagen-peroxidase buffer, 75 ul of 420 mM galactose, 50-75 ul protein fraction, 100 ul of 5% Triton X-100. b Units defined as nmoles/min. 32 Blanks were run without galactose. DISCUSSION Previous studies of mammalian galactose oxidation have suggested that the activity may be attributed to galac- tose-6-phosphate dehydrogenase (54), hexose-6-phosphate dehydrogenase (5), or galactose dehydrogenase (2-4). Galactose-6-phosphate dehydrogenase exhibits high specificity for galactose-6-phosphate and requires NAD+ for activity. It is found exclusively in liver cytosolic fractions (54). Hexose-6-phosphate dehydrogenase (formerly called glucose dehydrogenase) is a microsomal enzyme and shows minimal activity with galactose. However, this enzyme requires NAD+ or NADP+ for activity (5,53). Galac— tose dehydrogenase required NAD+ for activity and was a cytoplasmic enzyme (2—4). Its activity was later shown to be due to alcohol dehydrogenase and alcohol contami- nation of galactose reagents. Cuatrecasas and Segal reported, however, that they were able to isolate radioactive galac— tonic acid from a system containing liver supernatant fraction, NAD+, resorcinol, horseradish peroxidase, galac- tose-l-lfic, and hydrogen peroxide after 12 hours of incu— bation (3). Litchfield has shown that galactonate can be easily produced by an iglyitgg Fenton reaction involving hydroxyl radicals (10): 33 34 2+ Fe + H202 —9 Fe3+ + OH. + 0H" OSc-H + OH' -—> 0333' + H20 Osc- + OH' -—3 O\‘c'J—OH Galactose incubated in the presence of hydrogen peroxide and ferrous iron was partially converted to galactonic acid. In the reaction system used by Cuatrecasas and Segal, hydro- gen peroxide and potential ferrous iron contamination in the reagents or liver supernatant fraction may have generated hydroxyl radicals sufficient to cause the observed oxida- tion of galactose lg vitro. The results of Litchfield were verified here using the ferrous sulfate incubation system, and it was found that reduced bovine hemoglobin could serve as a source of ferrous iron to generate hydroxyl radicals, partially ox- idizing both galactose and glucose to their respective aldonic acids. However, neither intact red blood cells nor plasma could replace the hemoglobin. Apparently the level of cata— lase found in the red cell is sufficient to destroy the hydrogen peroxide before it can react with hemoglobin. Galactonate has been identified in the urine of human galactosemics and in the urine of controls given a large galactose load (1). It has not previously been identified to our knowledge in tissues of mammals maintained on high galactose diets. The removal of excess galactose from the tissue extracts through ion exchange chromatography proved important for the successful trimethylsilyl deriva- 35 tization and detection of the compound. Subsequent gas chromatography increased the sensitivity of the assay over the use of paper chromatography (3,5). Galactonate was found in the urine and tissues of all rats maintained on high galactose diets. In those animals drinking phenobarbital in addition to the galac- tose diet, more than two times as much galactonate was found in the liver, blood, and urine. Phenobarbital is known to cause a proliferation of smooth endoplasmic reticulum in liver, followed by an increase in rough endo- plasmic reticulum (62,63). This increase in liver micro— somal protein may explain the increase in liver galactonate levels found in those animals fed galactose diets and drinking phenobarbital. The higher values seen in the blood and urine of these animals are probably a reflection of the liver values. The action of phenobarbital in other organs is more obscure, but one explanation for the lower galactonate levels observed in all other organs tested in the galactose fed, phenobarbital drinking animals as compared to the animals eating only galactose may be that less galactose is transported to the other organs because of the liver's increased ability in detoxification. This hypothesis could be tested by measuring the galactose levels found in the blood and tissues of the animals eating galactose and drinking 0.1% phenobarbital and those eating only galactose. When microsomes were isolated from the livers of 36 the animals fed special diets, it was found that feeding galactose for 72 hours did not induce the galactose oxidizing activity. The addition of phenobarbital to the animals' water resulted in microsomes with decreased specific activi- ties. When 0.1% phenobarbital was added to microsomes prepared from rats eating chow diets, there was no signi- ficant effect on galactose oxidizing activity. Phenobarbi- tal induces the synthesis of some microsomal proteins espe- cially related to the metabolism of the drug more than it increases the synthesis of this galactose oxidizing en- zyme, resulting in the overall decrease in specific activity. Galactose is oxidized to galactonolactone in a number of other systems. An NAD+ requiring galactose dehydro- genase found in Pseudomonas saccharophilia (67) and a lactose dehydrogenase from Pseudomonas graveolens (68) are known to show galactonate production. An aerodehydrogenase isolated from citrus fruit oxidizes galactose and several other sugars to the corresponding aldonic acid with simul- taneous formation of hydrogen peroxide (66). The mold Polyporus circinatus produces an oxidase that catalyzes the oxidation of galactose at the C-6 position with pro- duction of hydrogen peroxide (69,70): this enzyme is now widely used for measuring galactose levels in various tissues. The microsomal galactose oxidizing activity reported here is not readily attributable to any currently known mechanism. Since singlet oxygen trappers, hydroxyl 37 radical scavengers, and antioxidants had little effect on galactose oxidation, these diffusion mediated oxidations may be ruled out as sources of activity. Pyrazole, a strong alcohol dehydrogenase inhibitor (71) also had no effect in this system, and the presented data are not consistent with a lipid peroxidation mediated reaction. The high degree of substrate specificity, the use of oxygen as the sole hydrogen acceptor, and the formation of hydrogen per- oxide together indicate liver microsomes as a new source of galactose oxidizing activity. The data here suggest an enzyme that catalyzes a reaction similar to that seen with bacterial glucose oxidase (72-76): D—Galactose + O --# Q-Galactonate + H O 2 2 2 To determine conclusively whether this is the observed reaction, a better method of quantitation of hydrogen peroxide must be sought where endogenous substrates do not compete with the Chromagen. A successful separation of the protein from endogenous catalase and peroxidases would also aid in the quantitation of the hydrogen peroxide produced in this reaction. Simultaneous purification of both the galactonate and peroxide producing activities would provide a strong argument for such a reaction. It is not certain whether galactonate is the initial product of the oxidation, since galactonolactone would be converted to the observed aldonic acid in its anionic form. The increase in activity observed with catalase 38 and superoxide dismutase provides support for the suggested reaction. Catalase may increase the rate of the forward reaction: azide may inhibit endogenous catalase or the enzyme itself. The increase observed with superoxide dis- mutase suggests the possibility of a superoxide anion inter- mediate in the formation of hydrogen peroxide. Liver microsomes from neonatal rats were found to have the highest galactose oxidizing activities. Older animals showed that the activity decreased with age by nearly 40% and reached a relatively constant value. Since only the neonatal rat is normally exposed to high galac- tose concentrations in the form of maternal milk, this correlates well with their activity. It is interesting to note that guinea pig and chicken liver microsomes did not show the galactose oxi- dizing activity. Chickens are not exposed to dietary galactose, but the enzymes of the uridine nucleotide pathway have been shown to be present in chick liver (77), although the activities are less than those seen in rat liver (78). Guinea pigs would normally encounter galac- tose in the neonatal diet. There remains the possibility that guinea pig and chicken liver microsomes may exhibit a difference in stability and another method of isolation using sucrose or other membrane stabilizers may be needed. Further investigation of this galactose oxidizing activity may be directed at finding an assay that is less time consuming than the galactonate assay and more reliable 39 than the peroxide assay. With a valid peroxide assay, the protein could be further purified. The results reported here are based on microsomal preparations; with a purified protein the effects of various activators and inhibitors may be altered. Another aspect to study would be the correlation of this activity with that seen in a fresh sample of human liver. The presence of this galactose oxidizing activity in human liver would help to explain the excretion of galactonate seen in human galactosemics and in those given a galactose load. In the dietary studies most of the galactonate appeared to be excreted in the urine, but further metabo- lism of the compound cannot be ruled out since previous reports have shown the conversion of small amounts of 14 galactose—1- C to labelled CO2 (26.40.79.80). It is also reasonable to suspect that the accumulation of galactonate (ranging from 0.045 to 3.35 mM) may affect the activity of other enzymes in the tissues; Meisler has shown that 1 mM galactonolactone is sufficient to inhibit human liver fi—galactosidase by 50% (81) and galactonolactone also competitively inhibits human gang- lioside GMl fl-galactosidase (82). Evidence has been presented here for a micro- somal enzyme with galactose oxidizing activity. While this activity may be the result of a previously unknown oxidase, the high apparent Km for galactose suggests that the activity is the result of a microsomal enzyme 40 with a different primary function. Galactose oxidation may be a reaction called upon when galactose is present in high concentrations, such as in human galactosemia or in galactose tolerance studies. REFERENCES 10. 11. 12' 13. 14. 15. REFERENCES Bergren, W. R., Ng, W. G., Donnell, G. N. and Markey, S. P. (1972) Science 176, 683-684. Cuatrecasas, P. and Segal, S. (1966) Science 153, 549-551- Cuatrecasas, P. and Segal, S. (1966) J. Biol. Chem. a1. 5904—5909. Cuatrecasas, P. and Segal, S. (1966) J. Biol. Chem. 241) 5910—5918- Srivastava, S. K. and Beutler, E. (1969) J. Biol. Chem. 233. 6377-6382- Beutler, E. (1967) Science 156, 1516—1517. Shaw, C. R. and Koen, A. L. (1967) Science 156, 1517-1518. Litchfield, W. J. and Wells, W. W. (1976) Infect. Immun. 63, 728—734. Litchfield, W. J. and Wells, W. W. (1977) Infect. Immun. 16, 198—202. Litchfield, W. J. (1976) Ph.D. Thesis, Michigan State University, pp. 121— 122 Fong, K. L., McCay,P . B., Poyer, J. L. , Keele, R. B. and Misra, H. (1973) J. Biol. Chem. 248, 7792- 7797. Penington, J. S. and Prankerd, T. A. (1958) Clin. Sci. 11, 385-391. Medline, A. and Medline, N. M. (1972) Can. Med. Assoc. J. 102, 877-878. Leloir, L. F. (1951) Arch. Biochem. Biophys. 33, 186-190. Leloir, L. F. (1951) in Phosphorus Metabolism (McElroy, W. D. and Glass, B., eds.), Vol. 1, pp. 67— 93, The Johns Hopkins Press, Baltimore. 41 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 42 Cardini, L. E. and Leloir, L. F. (1953) Arch. Biochem. Biophys. 25. 55-64. Kalckar, H. M., Braganca, B. and Munch-Petersen, A. (1953) Nature 112, 1038. Atkinson, M., Burton, R. and Morton, R. (1961) BiOChemo Jo ZQJ 813-820. Segal, s. (1972) in The Metabolic Basis 61 Inheriteg Disease (Stanbury, J. B., Wyngaarden, J. B. and Fredrick- son, D. S., eds.), 3rd ed., pp. 174-195, McGraw-Hill Book Company, New York. Mason, H. H. and Turner, M. E. (1935) Amer. J. Dis. Child. 56, 359-374. Hsia, D. Y. Y. and Walker, F. A. (1961) J. Pediat. 59, 872-883. Donnell, G. N., Collado, M. and Koch, R. (1961) J. Pediat. 53. 836-844. Nadler, H. L., Inouye, T. and Hsia, D. Y. Y. (1969) in Galactosemia (Hsia, D. Y. Y., ed.), pp.127-139, Charles C. Thomas, Springfield, IL. Komrower, G. M. and Lee, D. H. (1970) Arch. Dis. Child. ii. 367-373- Donnell, G. N., Koch, R. and Bergren, W. R. (1969) in Galactosemia (Hsia, D. Y. Y., ed.), pp. 247-268, Charles C. Thomas, Springfield, IL. Komrower, G. M., Schwarz, V., Holzel, A. and Golberg, L. (1956) Arch. Dis. Child. ll: 254-264. Holzel, A., Komrower, G. M. and Schwarz, V. (1957) Am. J0 MBd. gg, 703-7110 Holzel, A., Komrower, G. M. and Wilson, V. K. (1952) Brit. MGdO Jo l) 194-1950 Cusworth, D. 0., Dent, C. E. and Flynn, F. V. (1955) Arch. Dis Child. 16, 150-154. Sidbury, J. B. (1960) in Molecular Genetics and Human Disease (Gardner, L. E., ed.), p. 61, Charles C. Thomas, Springfield, IL. Hayman, S. and Kinoshita, J. H. (1965) J. Biol. Chem. 240, 877-882. Moonsammy, G. and Stewart, M. (1967) J. Neurochem. 14, 1187-1193. 33- 34. 35- 36. 37- 38. 39- 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 43 Wells, W. W., Pitman, T. A., Wells, H. J. and Egan, T. J. (1965) J. Biol. Chem. 240, 1002-1004. Quan-Ma, R., Wells, H. J., Wells, W. W., Sherman, F. E. and Egan, T. J. (1966) Am. J. Dis. Child. 112, 477-478. Wells, W. W., Pitman, T. A. and Egan, T. J. (1964) J. Biol. Chem. 232, 3192-3195. Weinstein, A. N. and Segal, S. (1968) Biochim. Biophys. Acta 156, 9-16. Kinoshita, J. H. and Merola, L. O. (1964) Invest. Ophthal. 3, 577-584. Sippel, T. 0. (1966) Invest. Ophthal. 1. 568-575. Sidbury, J. B. (1969) in Galactosemia (Hsia, D. Y. Y., ed.), pp. 13-29, Charles C. Thomas, Springfield, IL. Segal, S., Blair, A. and Roth, H. (1965) Am. J. Med. 8, 62-70- Se al, S., Blair, A. and Topper, Y. J. (1962) Science 12 , 150—151. Petricciani, J. C., Binder, M. K., Merril, C. R. and Greier, M. R. (1972) Science 125, 1368-1370. Hill, H. Z. and Puck, T. T. (1973) Science, 129, 1136-1140. Isselbacher, K. J. (1957) Science 126, 652-654. Isselbacher, K. J. (1958) J. Biol. Chem. 232, 429-444. Abraham, H. and Howell, R. (1969) J. Biol. Chem. 239.! 545'550 ' ‘ Segal, S., Rogers, S. and Holzapple, P. G. (1971) J. Clin. Invest. 56, 500-506. Inouye, T., Tannenbaum, M. and Hsia, D. Y. Y. (1962) Nature 122, 67-680 Ng, W. G. (1964) Ph.D. Thesis, University of Southern California. Posternak, T. and Resselet, J. P. (1954) Helv. Chim. Acta 37, 246-250. Lowry, O. H. and Passonneau, J. V. (1969) J. Biol. Chem- 244,H910-9160 52. 53- 54. 55- 56. 57- 58. 59- 60. 61. 62. 63. 64. 65. 66. 67. 68. 44 $015, A. and Crane, R. K. (1954) J. Biol. Chem. 210, 581-595- Beutler, E. and Morrison, M. (1967) J. Biol. Chem. 212. 5289—5293. Ray, M. and Bhaduri, A. (1975) J. Biol. Chem. 256, 3595-3601- Sweeley, C. C., Bentley, R., Makita, M. and Wells, W. W. (1963) J. Am. Chem. Soc. 85. 2497-2507. Ross, B. D. (1972) Perfusion Techniques 16 Biochemistry, p. 26, Clarendon Press, Oxford. May H. E. and McCay, P. B. (1968) J. Biol. Chem. 4}: 2288— 2295 Ernster, L. and Nordenbrand, K. (1967) Meth. Enzymol. fl: 574—580 ' Lowry, O. H., Rosebrough, N. J., Farr, A. D. and Ran— dall, R. J. (1951) J. Biol. Chem. 193, 265-275. Bernt, E. and Bergman, H. U. (1974) in Methods of Enzymatic Analysis (Bergman, H. U., ed. , Vol. pp. 2246-2248, Academic Press, Inc., New York. Stetten, M. R. and Ghosh, S. B. (1971) Biochim. Biophys. Acta 2}}, 163—175. Orrenius, S. and Ericsson, J. L. E. (1966) J. Cell Biol. 31, 243—256. Orrenius, S., Ericsson . and Ernster, L. (1965) J. Cell Biol. _6, J627— 639. Lineweaver, H. and Burk, D. (1934) J. Am. Chem. Soc. 56, 658-666. Hamilton, G. A. , DeJersey, J. and Adolf, P. K. (1973) in Oxidases and Related Redox Systems (King, T. E. et al, eds. ), Vol. 1, pp. 103— 119, University Park Press, Baltimore. Bean, R. C. Porter, G. G. and Steinberg, B. M. (1961) J. Biol. Chem. ”2 1235—1240 DeLey, J. and Doudoroff, M. (1957) J. Biol. Chem. ggz, 745-757- Nishizuka, T. and Hayaishi, I. (1962) J. Biol. Chem. 232, 2721-2728. 69. 70. 71. 72. 73- 74. 75- 76. 77- 78. 79- 80. 81. 82. 45 Cooper, J. A. D., Smith, W., Bacila, M. and Medina, H. (1959) J. Biol. Chem. 234, 445-448. Avigad, G., Amaral, D., Asensio, C. and Horecker, B. L. (1962) J. Biol. Chem. 232, 2736-2743. Teschke, R., Hasumura, Y. and Lieber, C. S. (1976) Arch. Biochem. Biophys. 125, 635-643. Muller, D. (1928) Biochem. Z. 122, 136-170. Muller, D. (1936) Ergebn. Enzymforsch. 5, 259-272. Franke, W. and Lorenz, F. (1937) Liebigs Ann. 532, 1-28. Franke, W. and Deffner, M. (1939) Liebigs Ann. 54 , Keilin, D. and Hartree, E. F. (1948) Biochem J. 42, 221—229. Mayes, J. S., Miller, L. R. and Myers, F. K. (1970) Biochem. Biophys. Res. Comm. 32, 661—665. Wells, H. J., Gordon, M. and Segal, S. (1970) Biochim. Biophys. Acta 222, 327—332. Topper, Y. J., Laster, L. and Segal, S. (1962) Nature 126. 1006. Baker, L., Mellman, W. J., Tedesco, T. A. and Segal, S. (1966) J. Pediat. 68. 551-558. Meisler, M. (1972) Meth. Enzymol. 26, 820—824. Sloan, H. R. (1972) Meth. Enzymol. 26, 868-874. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII lllllllllllllfllflllllylllllmljllllfllllllllH